Pitfalls and Options For Business-Desktop Linux

swhiser writes "Tom Adelstein dispassionately surveys the remaining fixes that will put desktop Linux through in the enterprise. Peer-to-peer networking, functional printing, laptop support, single sign-on to Active Directory and a better Device Manager (with a driver-get mechanism) are among the things companies are asking for. He says, 'The Linux desktop could fail if companies continue to pilot programs and conclude that it's less trouble to buy Microsoft. Everyone loses in that scenario.'" Pre-loaded systems are no longer a pipe dream or an obscurity, though; read on for one reader's mini-survey of Linux systems from large computer vendors.

Acidus writes "I called around today to the big OEMs (Gateway, Dell, HP, IBM) seeing who offered systems with Linux pre-installed, and the results were good. 3 of the 4 offered Linux on workstations. While no one offered Linux preloaded on laptops, Dell has some references nn how to install Linux on their laptops, while IBM has a scattering of docs on their website about installing Linux on systems. The reps at Dell, even though they have a series of Linux workstations, had to ask me what Linux was, and how to spell it. "Is that L-Y-N-I-C-S?""

Comments

As long as tech-knownothing PHBs keep making

[Megaweapon]

decisions in IT departments, Linux won't make much inroads on the desktop. They generally make decisions based on paid consultants and glossy magazine ads. Now, if the word spreads that companies can negotiate with MS based on threats of migration that'll keep some IT costs (somewhat) lower. Of course this can only work in bigger shops. Smaller companies can't do this.

Re:As long as tech-knownothing PHBs keep making

[_the_bascule]

From yesterdays slashdot story it is coincidental that this article addresses the same issues that a lot of the problems the IBM scholarship award posed.

link to PDF of problems ,provided by MBCook yesterday.

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

Tech know nothing PHBs know something you don't: if it's going to cost 2000 man hours of work at a $30 an hour average to redesign internal systems, templates, and procedures to work on a non-Microsoft system, that more than wipes out the cost of licensing the desktop systems. That doesn't include the cost of the lead up in which you have to test, deploy, and integrate all of your servers and desktops, plus the lost productivity from people needing to be retrained or retraining themselves on the shortcuts to use Linux.

People here act like a platform migration of this scale is simple as flipping a switch, and I think that really highlights how little experience in practical technology the Slashdot collective really has. You can reformat your system at home and install Linux in an hour depending on options and system speed. It's not that simple when you're talking about a business with 20 locations and 5000 workstations to migrate. It's not that simple when you have internal customer service apps to migrate. When you have internal template and procedures to rewrite. When you have to audit your hardware to ensure compatibility and then repurchase anything that might be too much hassle to fiddle with.

Migration to Linux isn't speading like wildfire for the same reason Windows shops don't jump ship to run to the superior UNIX systems even when that's cheaper: it's not as simple as you people think. It's not free. It's not even necessarily cheap. If it's going to cost you $250,000 to migrate and you're only going to be saving an average of $25,000 in license fees and support each year, it will take you ten years just to break even. Linux is not a magic bullet. You people whine and whine like little babies, but I'm willing to put my money where my mouth is that 99% of you only whine because you don't have the slightest clue what you're talking about. And the more you whine about your complete and total lack of knowledge, the more steam you give to other companies to muscle in on place where Linux could be making inroads.

What you need to do, if you really want Linux to succeed that badly, is address its single biggest shortcoming: the difficulty in migrating systems from Windows to Linux. No, it's not your fault that it's so hard. Microsoft intentionally makes it difficult to leave the nest. However, since you keep bitching about it, it IS your problem.

Quit being a whiny little bitch and contribute some code, documentation, consultation, or just shut the hell up. Your whining isn't going to change the fact that Linux just plain isn't a good solution for a lot of shops, but if you'd actually do some freaking work you chould change that.

Re:As long as tech-knownothing PHBs keep making

[megarich]

Your right but your example doesn't also apply to linux. Every new Windows upgrade and each new product MS feels the need to come out with, theycompanies need to test that too. And all those nifty little patches, big companies test them out because they need to. I'm sure we all read and saw results of what happens if you don't.

So remember, you have to take account BOTH sides of the coin and go with the best scenario. It'll be different for every company since each has their own needs...

Re:As long as tech-knownothing PHBs keep making

[killjoe]

If you want to migrate away from windows you need to start divorcing MS. Take a look at how Novell is doing their internal migration for example.

1) Do away with office. Replace office with openoffice the desktops (still windows).
2) Do away with outlook/exchange. Lucky for novell they have groupwise.
3) Set up a CMS system (novell used thei ifolder product) which keeps track of documents the employees create. This trains the employees to go to an abstract location for all their documents rather then "my documents".
4) Set up a desktop distro with open office, groupwise, ifolder and you are done.

It could be done with small gradual steps. Novell has done it, IBM is doing it and neither one of them is a small company.

Re:As long as tech-knownothing PHBs keep making

[MartinB]

Ahhh, consultants. Obviously they don't get Linux. And would never, ever recommend Linux on the Desktop.

Disclaimer: is my employer, but this is personal opinion.

Re:As long as tech-knownothing PHBs keep making

[einhverfr]

decisions in IT departments, Linux won't make much inroads on the desktop. They generally make decisions based on paid consultants and glossy magazine ads. Now, if the word spreads that companies can negotiate with MS based on threats of migration that'll keep some IT costs (somewhat) lower. Of course this can only work in bigger shops. Smaller companies can't do this.

The decisions have to be made on the basis of the business, not the basis of the technology itself. So I don't blame them for hiring experts (consultants) to do this sort of work.

THe fact is, there are Linux consultants too (I am one of them) and we can create decent marketing material. But the best marketing is done face to face where you can actually communicate with your customer.

Most of the problems talked about here are due to trying to fit a square peg (Linux) in a round hole (Windows workstation concept). Yes, it can be done, but it is not really optimal. Yes, we could use some more tools for allowing easy sharing of printers, for example.

Linux workstations are viable today, and have been viable for the past five years. THey offer unparalleled ease of use (given a little bit of effort to ensure that the desktop has the best icons, etc) and ease of administration, especially when given to someone who knows absolutely nothing about computer (Linux or Windows). For people who have become comfortable with Windows, this is another matter-- the computers are different enough that some frustration can be expected.

For small to midsize businesses who don't have an existing IT infrastructure, I tend to recommend Linux. For those who are not dependent on commercial vertically targetted applications, I recommend Linux. There are some gaps (payroll software being one) but these are relatively easy to mitigate.

Linux is not at the moment the desktop for everybody. Many businesses have way too much invested in Microsoft to make the move easily. And this is OK. It is up to consultants to counsel customers when their preferred solution is *not* likely to succeed too (far few take this road though).

The window of opportunity for desktop Linux is just beginning to open. Don't expect it to close too soon.

Re:As long as tech-knownothing PHBs keep making

[mpe]

People here act like a platform migration of this scale is simple as flipping a switch, and I think that really highlights how little experience in practical technology the Slashdot collective really has. You can reformat your system at home and install Linux in an hour depending on options and system speed. It's not that simple when you're talking about a business with 20 locations and 5000 workstations to migrate. It's not that simple when you have internal customer service apps to migrate. When you have internal template and procedures to rewrite. When you have to audit your hardware to ensure compatibility and then repurchase anything that might be too much hassle to fiddle with.

The thing is that it dosn't really matter what the migration is. Many of these things related to there being a migration. Yet there are plenty of people, including those in the PHB catagory, who take the view that so long as both the before and after involve Microsoft then it should be trivial.

Migration to Linux isn't speading like wildfire for the same reason Windows shops don't jump ship to run to the superior UNIX systems even when that's cheaper: it's not as simple as you people think. It's not free.

This is also the reason you find plenty of commercial systems running on ancient versions of MS Windows, even MSDOS...

It's not even necessarily cheap. If it's going to cost you $250,000 to migrate and you're only going to be saving an average of $25,000 in license fees and support each year, it will take you ten years just to break even.

As opposed to following the Microsoft prefered route of spending that $250,000 plus the licencing fees every 2-3 years.

Quit being a whiny little bitch and contribute some code, documentation, consultation, or just shut the hell up. Your whining isn't going to change the fact that Linux just plain isn't a good solution for a lot of shops,

The same applies to MS Windows. The difference is that Microsoft can pay lots of people to whine on their behalf...

Re:As long as tech-knownothing PHBs keep making

[DF5JT]

I have always wondered about that.

Why approach the PHB? Why not approach the guy familiar with the numbers? Go to the CFO and ask him how much money he spills down the drain for software licenses each and every year, inquire about software upgrades and the costs for that, loss of productivity through viruses, trojans and server downtimes. Talk money.

Take along your notebook, show him an OO-presentation and ask him about the necessity to throw away tons of money for a basic office setup that usually consists of no more than a webbrowser, a spreadsheet application and a word processor. Make him add up the numbers and show him that an enterprise is throwing away tons of money for stuff they can have for free at the same or better level of quality.

A truly dedicated CFO will have an open ear and mind for anything that will save money in the long run. Show him how to save money, tons of money and he will be the one to support you when PHB interfers with his glossy brochures.

Linux IS ready for the corporate desktop when one considers that maintaining a Windows based environment costs money, too. Migrating is the hard and painful step, but maintenance is easier and the cost savings are immense.

No reason to be defensive about Linux: It's a cost saver and that's something no CFO will ignore. CFOs are not the ones who need a glossy desktop, the latest IT-gadgets or shiny brochures. They are the sober type with an understanding for numbers.

Re:As long as tech-knownothing PHBs keep making

[passthecrackpipe]

real sad thing is - most of those issues have been sorted some time ago - in the shortlist of TFA, all except 8 are happening without problems today, and number 8 is happening but needs engineering. Rather clueless article, if you ask me.....

Re:As long as tech-knownothing PHBs keep making

[evdhn]

Migration to linux may not be easy or even desirable for a given comany, but while you stress what it may cost to migrate from windows to linux, you do not seem to take into account what migrating from one version of windows to another costs.

I vividly remember the long preparation and research time when a company I worked at moved from Windows '98 to Windows NT. Had to "to audit your hardware to ensure compatibility and then repurchase anything that might be too much hassle to fiddle with" then too.
I vividly remember the nightmares some of my colleagues went through when migration from NT to 2000 (I was on working on a UNIX project then). And do you remember the mix of proprietary workstations many companies used before Windows? They managed to migrate away from those too.

Just want to point out that running Windows also requires you to migrate and plan ahead and invest.

Emma

Re:As long as tech-knownothing PHBs keep making

[mgkimsal2]

It could be done with small gradual steps. Novell has done it, IBM is doing it and neither one of them is a small company.

They may be investing more heavily than other companies would for 'dogfooding' reason ('eat our own dog food', etc) They have services and products to sell and need to back up the claims of the products by proving it can be done, first and foremost. How they present accounting matters (if they do) is another issue altogether.

For most companies it's probably still cheaper just to buy MS than to move to Linux, due to all the migration issues. 'Cheaper' as measured in quarters, not years, but that's all most people can justify.

Re:As long as tech-knownothing PHBs keep making

[ckaminski]

Thank you for such an informative, and ultimately true, post. I've seen this working in large multinationals.

Remember, All. Microsoft won, not because they're technology was the best, or because of tying contracts. They won because of polish, and commercial developer support.

And Linux has trouble with both of those at the moment (although the polish is getting better).

Re:As long as tech-knownothing PHBs keep making

[99BottlesOfBeerInMyF]

You are claiming that ignorance of management is not a major reason for Linux uptake? You are dead wrong. Managers and executives are not, for the most part, tech saavy or well informed.

I have worked at two different companies where managers from on high decreed that Windows was going to be the ubiquitous platform for the whole company. In one case they wrote rules banning all mail clients other that Outlook and banning the use of any free software. This was for a company that wrote software that only ran on UNIX/Linux/BSD machines and was, of course, completely developed with free software. The rules they decreed would have completely stopped all work had the engineering middle management not lied through their teeth.

I have no faith that any given manager will have a clue, or even bother to inform themselves or talk to the people using the systems before making technical decisions. Undoubtedly some of those will be to Linux where it is inappropriate. For the most part, however, managers will just use Windows because that is what they know, and they don't want to look stupid.

The company I work for now, is a UNIX development shop, but even here you notice that the farther up the chain of command you get, the more windows machines you see. Luckily, we have some very clueful middle managers who have a hands-off style. As a result, there are maybe 3 windows machines among actual coders, and as many apples as X86 systems.

In summary, if you think the technical ignorance of managers is a myth, and that they have more of a clue than the the average slashdotter, I'm sure a thousand horror stories will be presented to prove you wrong.

Re:As long as tech-knownothing PHBs keep making

[simon_hibbs2]

My favourite bit is this: "The Linux desktop could fail if companies continue to pilot programs and conclude that it's less trouble to buy Microsoft. Everyone loses in that scenario.'" Does Microsoft lose? If the customer decides that the MS solution is genuinely better value for them (and in laptops especialy they may have a point), are they losing? Simon Hibbs

Re:As long as tech-knownothing PHBs keep making

[mpe]

Most of the problems talked about here are due to trying to fit a square peg (Linux) in a round hole (Windows workstation concept).

Which would be really ironic in cases where Windows machines are themselves a "square peg in a round hole". From the POV of the task which needs to be done. (Even more so if a unix style workstation was a better tool for the job in the first place.)

Yes, it can be done, but it is not really optimal.

Linux isn't Windows. But nor is Windows Linux, Unix, a mainframe, a glass teletype, an X-Term, etc.

Yes, we could use some more tools for allowing easy sharing of printers, for example.

There are also situations where you might want to make sharing of printers hard. e.g. a printer which prints on official stationary and automatically puts the result into an addressed envelope...

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

That's a highly irrelevant point you made. Migrating from 9x to NT was a complete platform migration. NT functioned significantly differently than 9x did, it just retained the surface elements that made it look familiar. It was well known that much of what ran on 9x would not run on NT or XP because they were, in fact, different platforms. The change is not as drastic as moving from Windows to a UNIX like system, but it's pretty bloody close. Upgrading from NT to XP, by comparison, results in far fewer problems because they share a common base.

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

...$250,000 plus the licencing fees every 2-3 years.

Uh. No. The point stands as I made it: if you're paying $25,000 in licensing fees and support every year to maintain your windows platform and it's going to cost you $250,000 to migrate to a UNIX type system, it's going to take you 10 years just to break even on the migration vs. maintaining a Windows base and you're going to carry a deficit for nine of those years.

Re:As long as tech-knownothing PHBs keep making

[killjoe]

Most enterprise customers use accounting programs that run on oracle. Novell isn't running their books on quickbooks.

"For most companies it's probably still cheaper just to buy MS than to move to Linux, due to all the migration issues."

Since most companies are mom and pop that's not really true. Enterprise customers do tend to have high migration costs but they have shown willingness in the past to spend a lot of money to gain long term advantage. Do a google for "SAP migration" and see what I am talking about.

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

You are dead wrong. Managers and executives are not, for the most part, tech saavy or well informed.

Nor should they be as that's YOUR job, not theirs. I'd also like to note that the rest of your post shows the astounding arrogance displayed daily by the Slashdot community. Who cares what management uses? Maybe they prefer to be able to type up their documents and drag them onto an email to create an attachment rather than launch off a two line, archaic sendmail command with pipes and redirects sticking out all over the place. Did it ever occur to you that maybe the reason non-technical people use Windows is that it's for non-technical people, and technical people use UNIX like sytsems because they're for technical people?

And, management is interested in the bottom line. As a technical person presenting a migration proposal, it's your job to clearly explain the process, it's benefits, and it's risks. As such, failure to justify the process is entirely the fault of the presenter. If you have a management team that's honestly not interested in the bottom line, however, you have much bigger problems than your software platform.

Re:As long as tech-knownothing PHBs keep making

[mpe]

The point stands as I made it: if you're paying $25,000 in licensing fees and support every year to maintain your windows platform and it's going to cost you $250,000 to migrate to a UNIX type system, it's going to take you 10 years just to break even on the migration vs. maintaining a Windows base and you're going to carry a deficit for nine of those years.

Where do you get support for a 1994 vintage MS Windows from? You appear to be missing the point that migrations are part and parcel of "sticking" with MS Windows.

Re:As long as tech-knownothing PHBs keep making

[mpe]

Upgrading from NT to XP, by comparison, results in far fewer problems because they share a common base.

In theory. Thing is that the theory dosn't matter much if the result is that some piece of "business critical" software won't work afterwards. The technical argument of "well NT and XP are similar" really dosn't matter.
There are even businesses which have to test every single patch which comes out of Microsoft. As well as circumstances in which Linux can run Windows programs better than Windows.

Re:As long as tech-knownothing PHBs keep making

[Anonymous Coward]

The guy is right, all is broken and printing under Linux is a nightmare (ever tried to print a presentation using paper from a different tray from another tray for the first page ?).
All this things don't work right out of the box, nobody wants lost nights reading cryptic comments in kernel sources and n compiles to get it working.
People like you who refuse to see the problem are our real problem.
Ah, and the GUI tools are a sad joke, the usual problems break the abstraction layer and stand out right trough them.

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

Yes, and all of these things count against Microsoft when you do your cost analysis. I fail to see what the point is. If you consider all of these factors and it's still cheaper to maintain Windows systems, what does it matter?

Re:As long as tech-knownothing PHBs keep making

[99BottlesOfBeerInMyF]

You claimed in your original post that managers were not switching to Linux, because they were more informed about the cost, than the typical slashdotter. Now you state that they should not have to know about technology, that is my job. Don't you think that is a bit of a contradiction? You further claim that management is interested in the bottom line, and you're right. Unfortunately they are rarely informed enough to make correct decisions regarding the bottom line, and the bottom line they really care about is their own. The CEO of a company I left had an employment history that should have made him a red flag for every prospective employer. Three companies had tanked under his command, and two suffered losses, then fired him. For some reason, however, that hard work made him a millionaire. That is because he was looking after his own interests, not those of the company. This, in my experience, is par for the course. Managers are interested in short term gains, giving themselves bonuses and raises, with all available cash, and bailing once they have ruined a good and profitable company.

As far as it being my job to convince management otherwise, no it is not. My task is to accomplish specific goals, and improve our product. Managements job is to plan strategy, and provide reliable tools cheaply. In corporate America, however, their goal tends to be to make money for themselves at all costs. Few managers care about making a quality product, or a long lasting and stable company. The reason Linux will eventually win, is because companies that don't use it will be undercut by those who do. The process will take a while, and may be more socialist than Americans are used to. Look to China, Europe, and South America. These are places where innovation can still happen.

Re:As long as tech-knownothing PHBs keep making

[CumInHerTaco]

Microsoft may own the market and are the Mega Monopoly. But that will eventually change.

FOSS will eventually win.

It just takes longer, FOSS is the inexorable machine. It has no required capital, no market expectations. It will keep going as long as people are different. MS has to worry about market share and profits, FOSS does not. FOSS has the upper hand, you know this because no matter how much they try, MS can't make FOSS go away.

Re:As long as tech-knownothing PHBs keep making

[the_mad_poster]

You clearly don't understand your position in the company. Your job is to know technology, not theirs. You wouldn't blame yourself for being stupid and ignorant if you didn't understand the complexities of the company's accounting procedures, now would you? Of course not, it's not your job. However, if there were reason for you to know it, you would expect that the accountants would explain things in clear, concise terminology and present it in such a way that you can understand the concepts, if not the niggling details.

Likewise, it is not their job to understand the difference between NTFS and ReiserFS. If there are benefits and negatives, they must be phrased in relation to the companies operation. You don't talk about how great it is to have a jounraling file system that organizes on the disk in such a way that fragmentation is not an issue. That's stupid. You talk about how this filesystem will allow the company to cut losses related to lost data which will save $x per year by saving y hours of work on backup tape recovery.

Secondly, it is not management's job to provide tools, it's management's job to ensure that the tools each department says it needs are really what it needs. You can't sit back and be completely passive about your needs and expect management to come drop everything you want in your lap. You need to make proposals and justify them or you won't get ANYTHING. It's management's job to approve your decisions in the context of the entire business, it's not their job to make them for your in the first place.

Regarding your bitch fest about your incompetent management: big whoop. Weak companies run by weak people fail. What exactly was your point in complaining about the profiteers? If they're that incompetent then Linux is not an issue at all, the lack of direction is the problem, and no software platform was ever going to change that. You can't use overall weak management in a company as justification for your position that Linux only fails in businesses because of management. That doesn't even make any sense.

Re:As long as tech-knownothing PHBs keep making

[99BottlesOfBeerInMyF]

You clearly don't understand your position in the company.

You don't think this statement is a little presumptuous? You don't know who I am, what my company is, or what my position within it is. Unless, of course, you're one of them!

You can't use overall weak management in a company as justification for your position that Linux only fails in businesses because of management. That doesn't even make any sense.

I'm not providing reasons why Linux fails, I'm providing reasons why Linux is not deployed. It has nothing to do with technical merit, and everything to do with the people making decisions, their motivations, and their decision making process. It is a cultural issue, and one that will resolve itself, given time and some good old capitalist competition.

Re:As long as tech-knownothing PHBs keep making

[einhverfr]

There are also situations where you might want to make sharing of printers hard. e.g. a printer which prints on official stationary and automatically puts the result into an addressed envelope...

Sure, but for those cases, those tools can be uninstalled or locked down. And SE-Linux is a great tool is it not?

Re:As long as tech-knownothing PHBs keep making

[Xerp]

OK, this is pretty true but at the same time as divorcing MS you need to keep one thing in mind, and one thing alone: Is it simple enough for a PHB to understand and use?

You know the sort of questions you'll get. Of course, here we are assuming you're not using Linux as a multi-user server system and are instead running it in as a single-user "desktop" machine system. The real issue is that Linux is ready, but everyone else isn't.

Ease of installation and availability of software is a big one.

Can the PHB "Hotsync" with their Handspring / Blackberry? Can they just download some software and install it themselves? How about their iPod? Their digital camera? Their scanner? The latest and greatest gadget that does who-knows-what?

I'm afraid that even downloading and installing Firefox on a Linux desktop would be too much for a PHB. Until is it pretty much exactly the same as Microsoft Windows you will have a hard time getting anywhere with a PHB.

Take another user group as an example: designers. Lots of designers will only use Macromedia software. Big problem.

In summary, Linux is already (and has been for a long time) ready as a desktop. It just needs the support of hardware manufacturers, software vendors, and hence their nifty little pre-packaged yummy installers / drivers.

Until hardware and software vendors put the same time and money into developing solutions for Linux as well as Microsoft Windows, it just isn't going to get the recognition it deserves. It could just take one big player to get the ball rolling though; who knows what the future will hold? I love Linux, and would never use Microsoft software at home (I've been through Windows from version 3 up to Xp so I have a experience of both systems) but until "the whole gang" gets "with it", then those who are "technically challenged" won't be able top be part of it.

Re:As long as tech-knownothing PHBs keep making

[killjoe]

"Is it simple enough for a PHB to understand and use?"

Yes. If a PHB can use office they can use open office. IF they know how to use outlook then can use groupwise.

"Of course, here we are assuming you're not using Linux as a multi-user server system and are instead running it in as a single-user "desktop" machine system."

No I am not presuming that. Novell has full linux desktops.

"Ease of installation and availability of software is a big one."

This is handled by the IT dept.

"Can the PHB "Hotsync" with their Handspring / Blackberry? "

Yes.

"Can they just download some software and install it themselves"

Yes but they probably should not be allowed to.

"How about their iPod? Their digital camera? Their scanner? The latest and greatest gadget that does who-knows-what?"

At work? Why?

"I'm afraid that even downloading and installing Firefox on a Linux desktop would be too much for a PHB"

Why would it be any harder then windows. Fire up YAST, click on firebird and it installs. That's a lot easier then downloading MS software, unzipping it and then installing it. With YAST you don't even need to hunt it down on some web site.

"Until is it pretty much exactly the same as Microsoft Windows you will have a hard time getting anywhere with a PHB."

Linux is growing at an explosive pace. It's definately not having problems getting anywhere. WHen the growth of linux adoption slows down even a little bit then I'll worry. Until then it's all sheets to the wind and full force ahead.

"Until hardware and software vendors put the same time and money into developing solutions for Linux as well as Microsoft Windows, it just isn't going to get the recognition it deserves"

It's already getting the recognition it deserves. It's growing wildly despite the best efforts of MS and SCO stop it. They can't even seem to slow down a little bit. Tens of millions spent on lawsuits and it did not even slow down the growth of linux by .01%.

Honestly what planet do you live on? Are you under some impression that the adoption of linux is slowing down or has stopped? If so where did you get that impression.

Re:As long as tech-knownothing PHBs keep making

[wcdw]

Actually, I've found one of the biggest problems to be that PHBs won't even _consider_ Linux on the desktop. I have even gone so far as to propose a voluntary trial effort (backed by several interested parties), and using old machines due to be recycled.

I thought that would be a good test of viability, with a safety net and very little downside. (How much IS a PC-for-recyle worth?)

And yet the response to this zero $$$ low-effort proof-of-concept - from the CIO, or my boss' boss - was effectively "shut up and sit down".

This WAS a large enough company to leverage threats about switching, which might have played a role.

So the fact that there are uneducated PHBs who do make decisions based on glossy ads and golf conversation is perfectly valid.

And personally, I think it's a mistake to focus on 'making it easy to migrate from Windows'. To the nth degree, this means recreating the Windows desktop. BLEAH. By all means, work on making an easy-to-use-as-possiible desktop, but please, let's try and move the world forward, not backward!

Re:As long as tech-knownothing PHBs keep making

[Xerp]

How about their iPod? Their digital camera? Their scanner? The latest and greatest gadget that does who-knows-what?"

At work? Why?

At work. Yes. Its a PHB!

Honestly what planet do you live on? Are you under some impression that the adoption of linux is slowing down or has stopped? If so where did you get that impression.

I live on planet Earth, thanks. I'm under the impression that the growth of Linux is ever increasing.

I do, however, have a great deal of exposure to extremely non-technically people. Please remember that I am simply expressing an opinion based on my own personal experiences.

Re:As long as tech-knownothing PHBs keep making

[Anonymous Coward]

Instead of migration, use Linux for greenfield applications. Linux is genuinely cheaper than Microsoft for most computing, but wildly cheaper for new applications. Also, instead of dropping Microsoft like a hot potato, you can drop them slowly, although most Microsoft sales reps work hard to prevent this, and many present customers with the ultimatum "either you use ours exclusively, or you use none of ours at all". Many companies don't want to go "none at all" and so adopt excllusively. Other companies, pissed off with Microsoft shakedowns, go for the "none at all". As more adopt "none at all", the OSS pie grows larger, Microsofts grows smaller. At some point, Microsofts "standards" which aren't standards at all, become the minority. Then the whineing and complaining begins.

Re:As long as tech-knownothing PHBs keep making

[killjoe]

"I live on planet Earth, thanks. I'm under the impression that the growth of Linux is ever increasing."

Good. We both agree. Linux is growing and will continue to grow.

"I do, however, have a great deal of exposure to extremely non-technically people. Please remember that I am simply expressing an opinion based on my own personal experiences."

So do I. Trust me those people can't use windows either. We have a team of IT people that do nothing but answer helpdesk questions from clueless users who can't seem to be even able to check their email regularly because outlook is so confusing to them. These people would welcome something like evolution of thunderbird because it's so much simpler to use.

Re:As long as tech-knownothing PHBs keep making

[homer_ca]

"Can the PHB "Hotsync" with their Handspring / Blackberry? Can they just download some software and install it themselves? How about their iPod? Their digital camera? Their scanner? The latest and greatest gadget that does who-knows-what?"

Well, is it a toy that's supposed to accomodate a PHB's every wish for a gadget and game or is it a tool to get your work done? You could reasonably consider a Blackberry or PDA to be used for legitimate business purposes. If that's the case, it should be a company-issued device that's officially supported by the IT dept. Don't even think about going down the road of supporting personally owned PDAs or the more frivolous gadgets. You have a point though. PDA and Blackberry support is severely lacking on Linux right now.

Re:As long as tech-knownothing PHBs keep making

[Anonymous Coward]

You people whine and whine like little babies, but I'm willing to put my money where my mouth is that 99% of you only whine because you don't have the slightest clue what you're talking about

By my calculations the true # is 98.463% so give your money... your mouth is where? Uh, nevermind keep it.

Re:As long as tech-knownothing PHBs keep making

[Specter]

It's not just about the business apps; they're only the customer facing aspect of supporting an internal PC setup. You also need centrally managed:

1) Asset control
2) Software package and patch deployment
3) Anti-virus/spyware
4) Firewall
5) Backup and restore

There's a lot of infrastructure setup cost right there whether or not enterprise ready solutions exist in the Linux space for those 5 things.

Baking a cake for 4 people is one thing, making it for 15,000 is quite another.

Jared

Re:As long as tech-knownothing PHBs keep making

[mpe]

Sure, but for those cases, those tools can be uninstalled or locked down. And SE-Linux is a great tool is it not?

But can they be easily uninstalled or locked down with Windows?

Re:As long as tech-knownothing PHBs keep making

[mpe]

Yes, and all of these things count against Microsoft when you do your cost analysis.

Actually it counts against any migration. There is a basic engineering rule of thumb that you change something which is working.

I fail to see what the point is.

The point is that there is no single entity called "Windows". Even applying a Microsoft patch, let alone a "service pack", can effectivly be a "migration".
Windows is especially problematic to patch, probably more so than any other OS. Because Microsoft have deliberatly avoided producing a well structures system.

If you consider all of these factors and it's still cheaper to maintain Windows systems, what does it matter?

The point is, as with so called "TCO studies", many of the relevent factors tend to be ignored.
Remember you were the poster who effectivly claimed that Windows would always be a cheaper option.

WiFi support

[ChrisMDP]

From the article:

"Broader WiFi card support needs to be introduced to Linux. WiFi card support for the large and important group of laptop users hardly exists. The expedient solution here would to use something like Linuxant's DriverLoader which has the elegance of being a single point solution that's applicable to the great majority of user/device scenarios."

This is the single reason that stopped my from installing Linux on my laptop. Until I discovered ndiswrapper, that is, which wraps windows wireless drivers...

Now if ndiswrapper worked out of the box, that *would* be a step forward.

Re:WiFi support

[Anonymous Coward]

Try FreeBSD 5. NDISulator included, right out of the box. I use it on my laptop with a Linksys (Broadcom) NIC and it works like a dream.

Re:WiFi support

[MobyDisk]

Instead of writing wrappers so that drivers from other OSs work on Linux, we should turn those wrappers into development kits and make them available to manufacturers. By "make them available" I don't mean place them on Sourceforge and announce them on Freshmean. I mean they should be marketed to hardware manufacturers so that they know that they exist, and that they can gain market share by using this simple tool to make a Linux driver.

(But then we get into the whole binary driver thing and it all goes to hell. Maybe that has to be resolved first.)

We also need to partner with manufacturers, like Microsoft does, so that kernel developers know when new hardware and drivers are available. This will fix the catch-up game where Linux drivers aren't deployed until the hardware is already on the shelf. We can also ease driver development using the feedback from manufacturers.

Re:WiFi support

[EnronHaliburton2004]

Try FreeBSD 5.

Don't like your bathroom? Purchase a new house!

Stable driver API

[TheToon]

A stable driver API is one of the things that is much needed. This is even a problem for server environments. In a perfect world, all drivers would be open source and easy to include, but that is just a pipe dream at the moment. There is a need for binary only drivers for several reasons, where a) support and b) it includes patented/licensed code are two of the biggest.

As it is now, Linux on the Desktop is only feasible for very specific desktop environments. And on laptops? Power management and wireless networking are not automatic, and with several different hardware versions and with users that roam the world... it's a pain.

Linux is getting there though, but slowly. The support cost for linux on desktops and laptops in corporations today would be too high I fear.

Re:Stable driver API

[rcw-work]

In a perfect world, all drivers would be open source and easy to include, but that is just a pipe dream at the moment.

You're 100% there on probably 90% of the hardware being sold, and you call it a pipe dream?!

Re:Stable driver API

[TheToon]

Yes, because those last 10% is what gives you problems. If you just go to your local electronic store and buy a Wifi PC Card (both for the Radius servers at work and with WPA for the users home nets, and open or WEP or WPA encrypted customer/coffee shot nets), you buy a MP3 player where you want do up/download music and use it as a portable storage device, you buy a label printer and a scanner for desktop use. Will it all "just work"? Nope.

Sure, you can find stuff that will work in Linux, but some requires 3rd party drivers (madwifi? how can you support that in a corporation) or binary only drivers (video cards, custom high end storage devices) or you have to use "vi" to configure it.

It has to be easily installed even by Joe Sixpak, else your support costs will skyrocket. IMO, this is the largest stumbling block for Linux Desktops.

Re:Stable driver API

[MichaelIhde]

Why not create a kernel module that wraps the current API into some stable API? Instead of each vendor trying to do this (NVIDIA, etc.) you could create one standard that could be shared by all. Certianly you might have trouble getting it into the mainstream kernel, as Linus will oppose it.

Re:Stable driver API

[Otter]

Yes, because those last 10% is what gives you problems.

For consumer use, absolutely. For business desktop use, you just don't buy that last 10%. And if you need it, or if your investment in it is too large to write off -- well, that's a good reason to stick with Windows.

Re:Stable driver API

[cr0sh]

I don't think I would want binary-only drivers on my server, if I was an admin. How do I know those drivers are secure? What if it isn't, and the code is bypassed or otherwised compromised to allow an intruder into my machine? What if I need to fix it because of this? How do I explain to my boss that "yeah, it is all open source, and we can fix anything - well, except this binary driver over here that was bypassed by an intruder who stole all of our IP"?

Yes, the same issues apply to Windows, on an even greater scale - and look how well security works (or doesn't) in that environment...

On home machines and possibly desktops, I could accept it (though ideally this wouldn't be the case) - but on servers...? I don't think I could honestly live with that...

Re:Stable driver API

[Libor Vanek]

Linux is so far stable and quick also thanks to the "no binary driver API". There is NO (read "NO!!!") reason for your driver to be closed source (maybe except ATI/nVIDIA which are doing some crazy optimalizations - but hell, why don't they devide they driver into open source (2D + basic 3D) part and binary (gaming 3D)?)

Re:Stable driver API

[Anonymous Coward]

And if you need it, or if your investment in it is too large to write off -- well, that's a good reason to stick with Windows.

You mean, like if your competitors are already getting the productivity improvements of that 10%?

News flash: business is competitive. You don't just throw away 10% of your business requirements to get a penguin t-shirt.

Re:Stable driver API

[Lumpy]

It has to be easily installed even by Joe Sixpak, else your support costs will skyrocket. IMO, this is the largest stumbling block for Linux Desktops.

HUH?? for a company with 60,000 different brands and configurations of machines I would agree with you. We are talkinga bout enterprise and corperate.

I just ordered 100 lattitude D800 laptops. every one of them will be 100% identical so as soon as I configure one to the base line, all other 99 are configured, all I need to do is image from the first. so if I need to change the WiFi settings for a machine I can easily replicate those changes to the other 100 machines.

this takes 1 tech sitting at his desk in a cube 1 hour. he does not even need to touch the laptops.

In windows this will take that 1 tech an entire day or more travelling to each laptop. 3 techs if you want to get it done without overtime.

so how does linux increase costs again?

Re:Stable driver API

[TheToon]

Yes, and an abstraction level like this is what is in use in many cases. You get a binary only driver and a small software stub you can compile against your driver to integrate it.

If the kernel provided such a layer by default, that would surely solve the problem -- because then you would have a standard API layer you make your binary only driver for. There is work going on for this with big-name IT companies behind it, so let's cross our fingers :)

Re:Stable driver API

[TheToon]

But that's hard. Cell phones, PDAs, home printers and equipment already installed, hotel equipment like broadband and printer services... it's easy to define a standard desktop and laptop, make a preloaded image that works and install that. But esp. for laptop users, when they are out there and have to interface with the real world, then you get problems.

Like on a hotel room I was last week with a special combo USB and Ethernet connector... only worked in Windows. How do you explain that to your CFO that will have a presentation tomorrow for a large financing group? Ask the receptionist? Dial up using his GSM cell phone and download this driver and compile it?

It got to be plug-and-play -- ideally just play -- the "plug" part can be hard too sometimes ;)

Re:Stable driver API

[Anonymous Coward]

Pretty amazing, deploying Linux to laptops that are still in cardboard boxes. Remember -- your Linux tech "does not even need to touch the laptops".

FUD

Re:Stable driver API

[fwitness]

Why is Joe Sixpack responsible for installing his own hardware? I can't think of an IT department that would incourage this.

On that note, why does he need an MP3 player? Get back to work Joe, or I'm sending out your pink slip.

Slacker.

Re:Stable driver API

[TheToon]

It's relatively easy to configure linux on a specific laptop with a specific wifi card to work on a specific wifi network. Having it work seamlessly with *all* wifis out there is much much harder.

We're talking about standard drivers and something like Windows XP SP2's "Choose a Wifi" application. Search, point, enter WPA passphrase and off you go. Corporate enterprise installations are not entirely static and predictable.

Corporate users expects the WiFi to just work, whatever the hotspot the user is in.

Re:Stable driver API

[_Sprocket_]

Yes, because those last 10% is what gives you problems. If you just go to your local electronic store and buy a Wifi PC Card (both for the Radius servers at work and with WPA for the users home nets, and open or WEP or WPA encrypted customer/coffee shot nets), you buy a MP3 player where you want do up/download music and use it as a portable storage device, you buy a label printer and a scanner for desktop use. Will it all "just work"? Nope.

When you bought your 10% piece of hardware, did you even bother to check what it will work with? Does it require some wierd plug you've never heard of? Does it have drivers for your OS? If that OS is Linux, do they note Linux compatability? And if not, why are you buying it?

I feel the pain. My household systems run either Linux or Win2K. In both cases, I've had hardware that didn't support my desired OS' (or took a bit of work to get more-or-less working).

Having said all that - even when hardware claims support for your OS, things aren't always set. Read hardware reviews. You'll find plenty of cases where a product's drivers have noted issues. And while that's not QUITE the issue here, it does further highlight the fact that a lot of this compatability issue belongs to the hardware manufactorer / vendor.

Buy what supports your system, and supports it well.

It has to be easily installed even by Joe Sixpak, else your support costs will skyrocket. IMO, this is the largest stumbling block for Linux Desktops.

Your support costs are going to skyrocket the minute you start trying to support Joe Sixpak plugging in random hardware and associated software.

As an aside, if this was really an issue, MacOS would have dominant marketshare.

Now... having made those statements, I agree with the general sentiment. I'd like Linux support to be better. But I just don't see it as insightful or productive to demand it of Linux developers.

Re:Stable driver API

[Peer]

...or you have to use "vi" to configure it.

It has to be easily installed even by Joe Sixpak, else your support costs will skyrocket.

Hmmm.. I like joe as an editor, but installing stuff with the sixpack option is not documented well enough.

Re:Stable driver API

[TheToon]

OK, real world example:

Imagine you're in a large corporation. You have windows, linux and unix boxes, with multi terrabyte storage systems. You use one of the few available backup/storage solutions that is delivered from one of the large players in enterprise level computing, IBM. With your Tivoli management solution you have Tivoli Storage Manager (TSM).

To those TSM servers in your storage hierarchy that are attache to tape libraries and tape drives you need special drivers that are supplied with TSM. These drivers are kernel modules and they are compiled against specific kernel levels of Red Hat Enterprise Linux and SUSE Enterprise Linx. The drivers are fully supported by IBM, but they require you to run them on the exact kernel version supplied by Red Hat or SUSE. And they are binary only.

So what do you do in your multi-million dollar SAN/storage solution? Port to amanda? Use cpio and library scripts in perl? Remember, you need full support for your backup solution, and it must work on Linux, Windows, AIX, HP-UX, Solaris, Mac and Netware servers.

Know if the drivers are secure? You don't care. They are supported, that's what count. (besides, your SAN/backup network should be closed and secured anyway :)

Re:Stable driver API

[Anonymous Coward]

You get the PFY to image the other 99, idiot.

Re:Stable driver API

[Anonymous Coward]

pretty amazing how fricking stupid you are. are you in marketing or upper management?

DUD

Re:Stable driver API

[TheToon]

Clearly the lawyers disagree with you (IP rights and patents) and lawyers are important in corporations.

Don't get me wrong, I would love all open source drivers. But it's just not feasible at the moment. We might get there some day.

Read the /. article here for reasons why nvidia has not open sourced their driver. And that is just one example, you would get the same mantra from most providers of binary only drivers

Re:Stable driver API

[Anonymous Coward]

and when every computer in your offices use that "specific" wifi card...

your point again is?

Lumpy is right, Corperations do not order computers with every one different. they order a base approved computer so that they are all the same.

so he is right, in a corperate setting there is no disadvantages.

the WiFi hotspots that the corperate users are ALLOWED to connect to are all named the same and properly designed so they hand off properly (we use real stuff not the crap you can buy at Compusa) so I can take ONE configuration and it will work everywhere in every office. no it will nto work at sally's donut joint. it is not supposed to.

Re:Stable driver API

[Anonymous Coward]

If you think things just work with Windows XP, you've been hiding under a rock. In the last couple of weeks, I've had all of the following fail to just work under XP:

HP Multifunction Printer
Sharp Zaurus Sync
2 different USB mice

I would have had no problems getting any of them to work under Linux, but my wife wanted them connected to her Windows box.

In the end, the HP printer would never work with out new drivers (HP had incorrectly coded the USB ID in the driver), since HP tech support is a toll call, we just returned it and bought a lexmark. The zaurus worked after I manually found and specified the drivers and the USB mice just started working after a few hours for no reason.

I wasted many hours trying to get the hardware wokring under Windows. Under Linux, it would have been trivial. And the Zaurus sync, does not allow the zaurus to share the windows internet connection because windows has ip addresses hardcoded for sharing and they are already in use.

Just works, in Gate's wet dream.

Re:Stable driver API

[Sylver Dragon]

In windows this will take that 1 tech an entire day or more travelling to each laptop. 3 techs if you want to get it done without overtime.

Um, bullshit.
Its basically the same process either way. Configure one laptop, sysprep, image, and deploy on the rest. Then, if you have any changes, you deploy them through Active Directory.
Yes, Linux does have advantages, but let's not going lying about how "hard" it is to administer Windows. Getting a bunch of laptops/desktops, with the same hardware/software up and running is easy with imaging, no matter which OS you choose. Where the difference is, is after they are running. With MS, you have to deal with patching regularly, which can be a nuciance, though SMS helps with this. Viruses are a problem, though again a good enterprise solution will have this sorted in no time. The main problem for keeping a system running is the interface between the chair and the keyboard; and this can be mitigated by liberal use of policies. Linux will help here, as the default would be to lock people out of installing software, but if you're in an environment where you want your users to be able to install software, they're gonna have the root password to the box, and they're gonna use it.
Yes, Linux is more stable and secure by default, this is why the FTP server I just built for my office is running Debian. But on the desktop, easy of use, which includes transitioning people from Windows; supporting nearly anything "out of the box"; and central password/authentication management are going to be killer apps for businesses. Yes, some companies could go back to a server and thin client setup, thus having central management; but this still leaves a lot of companies out in the cold. And <insert dieity here> forbid that whatever storage array you have all of this working off of goes tits-up (I've been dealing with this particular problem all week, its a sore spot at the moment).
Linux is making inroads on the desktop, and this is a very good thing. With any luck the developers who are working on it (thank you guys) pay attention to articles like this one. If the goal of Linux is to be a serious competitor to Windows, these types of things are going to have to happen.

Re:Stable driver API

[Anonymous Coward]

deploying through active directory NEVER works 100%. Nice try, but I use it here every day and there is ALWAYS some machines that the changes do NOT get to.

take off your rose colored glasses with the microsoft logon on them.

there are GOBS of problems with active dir. Most of us that have to deal with it daily know this.

and he is a BIT off, it takes 2 techs 2 days. they have to drive to seperate offices to fix the problems.

we switched AWAY from AD about a month ago to perl scripts (yes, PERL) running as a service on the local machines that self update and also self replicate out changes and then run the changes.

I can deploy OFFICE 2003 this way to all desktops. Active dir. CAN NOT do this sucessfully. (we tried for 2 weeks befoer rolling out the perl solution from a O-rieley book for windows NT.)

Re:Stable driver API

[thepoch]

It has to be easily installed even by Joe Sixpak, else your support costs will skyrocket. IMO, this is the largest stumbling block for Linux Desktops.

This is like saying that Joe Sixpak should be able install their own structured cabling in the office just to be able to surf the Net. It isn't Joe Sixpak's job to install or configure an OS. It's the OEM's and the SysAd's job to do that. In fact, the SysAd should lock down the OS. Joe Sixpak shouldn't even be allowed to install any additional apps. Only the SysAd should be allowed to install additional applications after approval from higher management, and after evaluation of said app. Besides, wasn't the article about Business-Desktop Linux and not Joe Sixpak-Desktop Linux?

Of note I use a Thinkpad T30 with the recently released Fedora Core 3. I'm a financial officer and systems administrator (yeah, strange combination). Upon install I can access USB Flash drives, I can use my crappy Mustek scanner, I can print on our companies networked printer, I can access my GF's MP3 player as a USB Flash drive as well. With a little configuration, I have a working WiFi connection.

Not only that, I've gained the advantage of knowing Linux and Windows. So if I ever switch jobs, they can rely on me for both technologies. I won't be locking a company into one-sided technologies.

Re:Stable driver API

[tchuladdiass]

Isn't that what ProjectUDI was for? Of coures, the various open source leaders rejected the idea because they were afraid, among other things, that it would allow people would use superior, free linux drivers to replace buggy or non-existant windows drivers, and that might upset the driver developers. Other complaints is that it would encourage binary-only drivers by removing the public demand for open source drivers.

Re:Stable driver API

[mikefe]

The WiFi area is being tackled by OpenBSD:

Thanks to their efforts both all all OSS systems (including Linux) will soon be able to have wifi on install without hassles.

OpenBSD: Trying To Contact Texas Instruments

Feature: OpenBSD Works To Open Wireless Chipsets

OpenBSD: Intel Refuses To Open Wireless Chipsets

Re:Stable driver API

[TheToon]

DOn't be silly. Of course it's supposed to work in Sally's Dounut Shop, and airports and hotels and visitor centers at other companies. What's the point of a mobile worker if not being mobile?

Re:Stable driver API

[rcw-home]

lawyers are important in corporations.

But ultimately not more important than customers.

Re:Stable driver API

[chthon]

IMHO the OEM's are the hardest part to crack.

Most of them like too much to make propaganda for Microsoft.

Yes, that is what I call it, propaganda. Because probably they are in a situation where they either are in blissful ignorance, or are downright paid by MS.

That is the feeling I get from several computershops here around (you want names ? You need to be from Brugge, Belgium to know them, Computerwinkeltje, Silicon Center).

Single sign-on to what ?

[Walrusss]

Why not single-sign-on to OpenLdap ?

Feel ready to own one or many Tux Stickers ?

Re:Single sign-on to what ?

[blowdart]

Why not single-sign-on to OpenLdap ?

Because companies want to use their existing infrastructure? Are you seriously suggesting you can sell linux on the workstation by telling companies to throw away their windows "investment" server side at the same time?

Re:Single sign-on to what ?

[Havokmon]

Because companies want to use their existing infrastructure? Are you seriously suggesting you can sell linux on the workstation by telling companies to throw away their windows "investment" server side at the same time?

AD supposedly supports LDAP. So you LDAP to AD.

I did it with NDS years ago. I assume AD would do it. Then again, MS just introduced salvage to Win2003.-shrug-

Re:Single sign-on to what ?

[ReelOddeeo]

Let me get this straight.

Gradually migrate desktops to Linux. Make them do sign on and authentication to a Windows server.

End result: Linux on the desktops, Windows as the server.

That way, each platform is being used for what it is best at.

Re:Single sign-on to what ?

[jmulvey]

Why not single-sign-on to OpenLdap ?

Is OpenLdap kerberized? (in other words, can you tie Kerberos security to permissions on the retrieval and setting of LDAP attributes?)

(hint: the answer is NO)

And because of this, OpenLdap authentications solutions are NOT secure, as they pass credentials in CLEARTEXT. Yes, you can use certificates but now you've introduced the thorny issues of key distribution.

Microsoft's Active Directory has smartly tied Kerberos and LDAP together, so LDAP queries can be encrypted with Kerberos... so no certificate distribution problems and secure from sniffing.

Sorry to rain on the MS-bashing parade. They did a good job here, too bad Slashdot isn't a merit-based rating.

Re:Single sign-on to what ?

[blowdart]

Nice sarcasm

My point was, I guess, that telling a company they must replace both parts of their infrastructure in order to use Linux is not the way to market it. Like it or not AD is being used in large companies, in concert with Exchange, ISA, Sharepoint, VPNs and whatever else. Lack of single sign on from workstations can well be a deal breaker, and saying "Oh just migrate your backend too" is not a helpful attritude.

Re:Single sign-on to what ?

[AstroDrabb]

MS's AD is as usual busted up LDAP. I have had no problems connecting a Linux box to every LDAP server I have tried, except for AD. I wrote a custom Java app that works with LDAP. Again, every LDAP server I pointed it to worked fine, until I pointed to AD. I had to make custom changes to work with AD. AD is typical MS "embrace, extend and break".

Re:Single sign-on to what ?

[demented]

If I'm not mistaken, according to this official OpenLDAP Admin Guide, OpenLDAP supports SASL framework, which, in turn, supports Kerberos V authentication via GSSAPI. It took me some time to put all the pieces in the right places on my Gentoo server installation, but after some Google searching, I finally had OpenLDAP over Kerberos authentication, and on top of all that, an AFS cell (using OpenAFS, of course).

Re:Single sign-on to what ?

[jrcamp]

Who moderated this lunitic +5? He is absolutely WRONG. My sibling poster already pointed this out but anybody who can google for "openldap kerberos" sees that people do use it.

A HOW-TO is even available here:
http://www.bayour.com/LDAPv3-HOWTO.html

Moderators need to only moderate on what they know about. These ones don't.

Re:Single sign-on to what ?

[PoochieReds]

Is OpenLdap kerberized? (in other words, can you tie Kerberos security to permissions on the retrieval and setting of LDAP attributes?) (hint: the answer is NO) Sorry, but this is just plain wrong...

OpenLDAP fully supports kerberos authentication (and many other types) via Cyrus (and maybe GNU) SASL libraries.

See the OpenLDAP SASL Instructions that document how to do it.

Re:Single sign-on to what ?

[jmulvey]

Try reading the document you linked to.

Let me say it again, this time slowly: You cannot encrypt LDAP communication using Kerberos tokens with OpenLdap. If you need secure LDAP communication with OpenLdap, you must distribute certificates to the client computers and use SSL/TLS. That's why your magic document has a section entitled, "Testing OpenLDAP, using your Kerberos ticket, with SSL/TLS" but it doesn't have one that says, "Testing OpenLDAP, using your Kerberos ticket, and encrypting with your Kerberos ticket".

I may be a "lunitic", but I know what I'm talking about.

Re:Single sign-on to what ?

[killjoe]

Both kerberos and LDAP are available. There is no reason why you can't do this on your own. If you don't feel like doing it just go get a Xserve and use the one that apple made for you. No big deal.

Re:Single sign-on to what ?

[fsmunoz]

Is OpenLdap kerberized? (in other words, can you tie Kerberos security to permissions on the retrieval and setting of LDAP attributes?) (hint: the answer is NO)

Er, the answer is YES. I have it working here. You can use the Kerberos tickets to authenticate to OpenLDAP and have ACL's in the LDAP server to define the permissions. It's done trough SASL and it works transparently.

And because of this, OpenLdap authentications solutions are NOT secure, as they pass credentials in CLEARTEXT. Yes, you can use certificates but now you've introduced the thorny issues of key distribution.

Not so. Understand that this is however seperate from the availability of Kerberos. Other methods can be used to pass the crendentials (Digest MD5, etc). Aditionally you can force the use of SSL, so even cleartext passwords are not problematic. You can actually define that the server won't accept cleartext from non-TLS connections.

I use OpenLDAP integrated with Kerberos and both integrated with the authentication and authorization of several different things (including machine logon). I also have a cross-realm trust relation between AD and the Unix LDAP which allows AD users to use their Windows tokens in the Unix environment (user "bar@WINDOWS.NET" assumes "bar@UNIX.NET" identity trough cross-realm). Aditionally, as a last resort for use in non-kerberized apps one can use the password '{KERBEROS}boo@UNIX.NET' or '{KERBEROS}boo@WINDOWS.NET' to make the LDAP server check the user supplied password in the Kerberos server.

Re:Single sign-on to what ?

[jmulvey]

Right. You can use Kerberos for authentication, but OpenLdap will not encrypt its data stream using the Kerberos token. So if I have a secure attribute, say "Salary" available in OpenLDAP and I use Kerberos/GSSAPI to query it, the value of the "Salary" field must pass over the network in cleartext. This is not the case with Active Directory.

Re:Single sign-on to what ?

[jmulvey]

Right. You can use Kerberos for authentication, but my point was that OpenLdap will not encrypt its data stream using the Kerberos token. So if I have a secure attribute, say "Salary" available in OpenLDAP and I use Kerberos/GSSAPI to query it, the value of the "Salary" field must pass over the network in cleartext. This is not the case with Active Directory.

Re:Single sign-on to what ?

[PoochieReds]

Wrong again! I'm not sure if you understand what kerberos does. Repeat after me:

kerberos is only for authentication

All it does is verify (securely) whether you are who you say you are. Everything else must be done with other mechanisms.

So you want to make sure that certain attributes are only visible over encrypted links? See the section on using the SSF parameter in your access controls in the slapd.conf documentation

OpenLDAP is a very capable LDAP server. I've found very little that it can't do (with some research).

Re:Single sign-on to what ?

[killjoe]

I thought the point of moderation was to weed out the nonsense. If somebody who out and lies or is severely misinformed gets modded up to +5 there is something severely wrong the moderation system.

The odd thing is that he is still modded at +5 despite being proven wrong. I guess all those MS shills have mod points today.

Re:Single sign-on to what ?

[PoochieReds]

Again, false...

You can make it so that certain attributes are only visible when the SSF (security strength factor) is above a certain value (indicating an encrypted channel of relative strength).

See the slapd.conf documentation at the OpenLDAP site (esp. the section on access controls).

Re:Single sign-on to what ?

[PhilipPeake]

You are not understanding the point he is making.
Yes, you can specify that SSL be used, but that isn't what he is saying or asking for.

What he wants is that the session key negociated with the Kerberos authentication be used for communication with LDAP, NOT a completely different key negociated via SSL. SSL requires certificates, using the kerberos session key removes any need for certificates and the associated pain of their management.

For examples of this look at the kerberized apps provided by the standard kerberos disribution - which include a telnet server/client which can be configured to encrypt communication unsing the kerberos session key.

Re:Single sign-on to what ?

[fsmunoz]

Right. You can use Kerberos for authentication, but my point was that OpenLdap will not encrypt its data stream using the Kerberos token. So if I have a secure attribute, say "Salary" available in OpenLDAP and I use Kerberos/GSSAPI to query it, the value of the "Salary" field must pass over the network in cleartext. This is not the case with Active Directory.

As PoochieReds said in the first reply this is also incorrect. The SSF he mentioned allows for that sort of data protection. An example from the docs he was refering to (and I'm only pasting because this could be useful for more people):

The server uses Security Strength Factors (SSF) to indicate the relative strength of protection. A SSF of zero (0) indicates no protections are in place. A SSF of one (1) indicates integrity protection are in place. A SSF greater than one (>1) roughly correlates to the effective encryption key length. For example, DES is 56, 3DES is 112, and AES 128, 192, or 256. A number of administrative controls rely on SSFs associated with TLS and SASL protection in place on an LDAP session. security controls disallow operations when appropriate protections are not in place. For example: security ssf=1 update_ssf=112 requires integrity protection for all operations and encryption protection, 3DES equivalent, for update operations (e.g. add, delete, modify, etc.). See slapd.conf(5) for details.

As you can see the mechanism is actually quite powerful and allows for very fine grained data protection (as do the ACL's).

All in all the most needed thing in OpenLDAP is a GUI tool to configure it :). Just about any LDAP browser is enough to make things work once setup, but maybe some graphical tool to do the setup would be useful. As far as capabilities go OpenLDAP is very good, one just has to actually read a lot of docs to deal with the more "advanced" ones.

Re:Single sign-on to what ?

[jmulvey]

You're right. Thanks, and I wasn't aware of that.

I found this to clarify better:
http://www.novell.com/documentation/suse91/suselin ux-adminguide/html/ch19s04.html

Search for "Security Strength Factor".

Re:Single sign-on to what ?

[Etyenne]

You, sir, are a total ignorant.

Kerberos is an authentication protocol. Kerberos ticket are not used to encrypt communication, they only convey authentication data. Encryption of the datastream is up to the application. SSL and STARTTLS are two popular way to encrypt various communication protocol. OpenLDAP happen to support both.

I may be a "lunitic", but I know what I'm talking about.

You wish you did. But you are entirely in your right to make a fool yourself publically, if you so wish.

We've been running Linux for quite a while now

[YodaToo]

We currently use a kickstart install of Fedora for our EE workstations. Customized it with everything we need including configs for the various workstation/networked printers.

We use NIS so that workstations are completely interchangable. Had an EE harddrive meltdown, grabbed a spare machine, ran the kickstart, and the user logged back in via NIS within 15 minutes with no data loss! Could have had him backup instantly if he wanted to go to a spare office.

I can't believe how much easier workstation admin is now that we use Linux.

Re:We've been running Linux for quite a while now

[media_Assassin]

How is this different from using Ghost (or free alternatives), authenticating to a Windows domain, with their data on a network share?

It would have been exactly the same outcome.

Re:We've been running Linux for quite a while now

[sporty]

This can also be done with roaming profiles on windows. Just a heads up. Windows loses on cost though..

Re:We've been running Linux for quite a while now

[YodaToo]

Its quite a bit cheaper and in my opinion a lot more secure!

Re:We've been running Linux for quite a while now

[media_Assassin]

But that's not the point - the conclusion point the parent made was "I can't believe how much easier workstation admin is now that we use Linux."

But the situation outlined would have been identically easy using the appropriate Windows tools.

Mind you, I'm replying from Firefox running on a Dell Laptop loaded with Fedora Core 2, so I'm not biased either way. I'm just saying, you can't claim it's easier to rebuild a Linux workstation vs. a Windows workstation if you use imaging or other similar technology.

Security, cost, etc are a whole different discussion.

Re:We've been running Linux for quite a while now

[Havokmon]

This can also be done with roaming profiles on windows. Just a heads up. Windows loses on cost though..

IMHO. Both are messy substitutes for Directory integration. Your applications should be installed on the server, and save data to the server. Any information they require that's user based needs to come from the directory schema.

Home directories mount from the file server.

No more workstation specific info, and no more passing 'crap' around to all the workstations.

See Netware and Pegasus Mail for an easy to follow implementation of this method.

Re:We've been running Linux for quite a while now

[sporty]

You are right, they are messy. Thus things like IMAP, webdav and the likes. But mind you, what you gain in portability, you lose in performance. Things like photoshop might not be happy when you deal with multiple multi meg files. Well.. not unahppy. .just slower.

Re:We've been running Linux for quite a while now

[YodaToo]

I didn't mean that workstation admin was easier for "just" the reasons I outlined, its my overall impression of Linux workstation vs Windows workstation admin. I'm sure almost anything can be automated in Windows as well, but I've personally found things easier to automate in Linux. Overall maintainence is less of a hassle and we have fewer user issues. Maybe we never had knowledge of or access to (financial or otherwise) the "right" Windows tools/software, but one way or another my overall impression of Linux is "better."

Re:We've been running Linux for quite a while now

[mpe]

This can also be done with roaming profiles on windows.

With Windows roaming profiles if your workstation goes bang then you might easily lose whatever you are working on. "Folder redirection" is probably better here. Though it appears to be far less well known...

Re:We've been running Linux for quite a while now

[mpe]

Home directories mount from the file server.

No more workstation specific info, and no more passing 'crap' around to all the workstations.

As well as minimal stuff actually on the workstations. Yet this appears to be contrary to the Microsoft way...

See Netware and Pegasus Mail for an easy to follow implementation of this method.

Dating from before Microsoft even got into client-server networking. I always found it ammusing that Microsoft couldn't even manage to copy Novell very well...

Re:We've been running Linux for quite a while now

[mpe]

Thus things like IMAP, webdav and the likes. But mind you, what you gain in portability, you lose in performance. Things like photoshop might not be happy when you deal with multiple multi meg files. Well.. not unahppy. .just slower.

IIRC network file systems started to outstrip local file systems with 100M switched ethernet. It's easier (and cheaper) to upgrade a few servers than to upgrade lots of workstations to have a disk system capable of a high I/O rate (or a lot of memory for cache).

Re:We've been running Linux for quite a while now

[Anonymous Coward]

I had roaming profile for some time. It works okay in win9x, but in XP, it takes a while to copy all the profile data back and forth, which means I have to wait about a minute when I login and also when I logout.

It was so much hassle that I only do network directory mapping and all profiles are now local. Win9x does okay with roaming profile only because the profile data is very small. But some win9x profiles don't work across heterogeneous machines.

With linux, since it was built with multi-user in mind, it feels so robust in heterogeneous environment with NIS, automount, etc. Nothing large needs to be copied back and forth either.

In the end, I was able to maintain several tens of linux machines part-time, while we had to hire 2 full-time and several part-time just to manage the Windows server and client PCs, and not even close to the advanced services linux provided.

Windows platform did create jobs, but jobs that took away people's skills to do something better when the technology moved. I wonder what would have happened if those people worked on other fields rather than being laid off from IT bust.

HP has a Linux laptop

[robyannetta]

http://h10010.www1.hp.com/wwpc/us/en/sm/WF06a/3219 57-64295-89315-321838-f33-395654.html

Re: HP has a Linux laptop

[Anonymous Coward]

Author: While no one offered Linux preloaded on laptops,

Reply: http://h10010.www1.hp.com/wwpc/us/en/sm/WF06a/3219 57-64295-89315-321838-f33-395654.html

Mod parent up! He just proved the author wrong!

Re:HP has a Linux laptop

[saur2004]

Giggle: Page not found

Re:HP has a Linux laptop

[robyannetta]

Try this: http://h10010.www1.hp.com/wwpc/us/en/sm/WF05a/3219 57-64295-89315-321838-f33-395654.html

Re:HP has a Linux laptop

[SassyDave]

Try this link instead. The laptop apparently comes with "SUSE® Linux HP Edition 9.1". Odd. Wonder what they've modified.

Re:Just keep using Windows

[Anonymous Coward]

Exactly! Why will I even bother to have Active Directory compatability if I can use kerberos and LDAP for this stuff?

Re:Just keep using Windows

[Bandman]

What's the point of using Linux, 'just because'?

Cost of licensing? Upward compatibility? Freedom of choice? Hardware requirements? Ability to customize workspace? Freedom from Microsoft inspections, like the ones MS has forced on city buereaucrats before? Better security?

Do I need to continue? I can...

Risk aversion

[Octagon Most]

Risk vs. reward for the decision-maker is going to be a key factor. If I am a CIO or CTO I am likely unwilling to bet my career on the risk of the unknown. There are possibly great cost advantages to deploying Linux on the desktop in the enterprise, but if that's not a primary focus area for the head of corporate technology then it is better to stay with what is know to work. Security factors are another big consideration, but in both of these cases it's a bit of a leap of faith. Windows is the known quantity and there is a massive budget in place around it. In other words, the main technology decision-maker is not likely to be rewarded as a hero for the advantages that Linux might bring, but would be sacrificed for any unforeseen downsides. One does not have to be too risk-averse to see why Microsoft remains entrenched.

Re:Risk aversion

[tbedolla]

I currently work closely with the CIO in a medium to large restaurant company where we are installing a POS system with Linux as the OS...Suse to be exact although we'll probably be using IRES from IBM which is just a imaging and deployment package around Suse. While we are working out the kinks of the POS, our online ordering system running IIS and Windows 2000 was comprimised and has left us waiting for a green light with our payment processor that has resulted in approximately 1 million in losses. I don't care how much of a budget Windows has if being a "known quantity" makes you a target.

Re:Risk aversion

[cliffyqs]

heck, we run win2k because we don't want to deal with the potential problems of going to xp. we have a couple win98 machines because the software they use does not run right under win2k. Linux is right out. this is funny because the hp mainframe will be migrating to a unix-shaped OS in a year or two. now I need to get an old castoff computer, install Linux, and learn. I have a Knoppix cd, but it doesn't like our integrated video much and will not talk to my modem (it's a winmodem, grr)

Re:Risk aversion

[Octagon Most]

"... our online ordering system running IIS and Windows 2000 was comprimised and has left us waiting for a green light with our payment processor that has resulted in approximately 1 million in losses. I don't care how much of a budget Windows has if being a "known quantity" makes you a target."

Good point. An alternative to a rewards for the positive benefits for deploying Linux on the desktop is punishment for sticking with something that is not working or causing problems. Security concerns with Windows are certainly an example of the latter. I still feel the path of least resistance is to throw more money at the entrenched product in an effort to fix it, but that's not a viable long-term strategy. Exposure in the corporate server arena and "pioneers" such as you will set the stage for Linux to gain wide-spread desktop deployment. But then again, you are talking about what sounds to me like a specific application-focused environment. The typical corporate cubical land is full of desk-bound workers who need a wide variety of standard office applications and some homegrown custom apps written in VB. The cost-savings vs. the desktop OS and office suite has to be compared with retraining costs and development of new customer apps. And all that has to be evaluated with productivity loss from, and overall cost to contain security breaches and attacks.

Re:Risk aversion

[Neil Watson]

That's what's wrong with people today. Everyone 'plays it safe'. No wonder some argue that innovation has slowed.

Re:Risk aversion

[Wudbaer]

There are risks that are worth to take. The question is if a Linux migration is one of those.

Especially if you are heading IT of some larger organization where any kind of migration becomes a multi-month or even multi-year event with respective costs (just consider the timeframe for the planned Munich Linux migration, IIRC correctly it's something going until 2007 or 2008).

In the end the old saying most of the time stays true: Never change a running system.

Re:Risk aversion

[tbedolla]

I certainly agree with your assessment, but with respect to the need of standard office applications and homegrown custom apps, I find that OpenOffice suits just about every need that MS Office will, and any useful app that gains momentum in the Windows arena is usually followed by a similar open source option not long there after (e.g., gimp, gaim, thunderbird, firefox etc). Let Windows do the trial and beta of new ideas, and then open source can take the best ideas and improve on them, while weeding out the unnecessary fluff (not that Linux is free of bulk either).

Re:Risk aversion

[killjoe]

There are whole books about why it's bad to innovate. Large businesses don't believe in innovation. THey leave that to the smaller companies and then buy them, steal their ideas or destroy them. History is filled with innovators that die or go nowhere.

Re:Risk aversion

[mpe]

heck, we run win2k because we don't want to deal with the potential problems of going to xp. we have a couple win98 machines because the software they use does not run right under win2k. Linux is right out. this is funny because the hp mainframe will be migrating to a unix-shaped OS in a year or two.

It's also "funny" because you probably could operate your 98 software using vmware (under Linux or WinXP) or win4lin (under Linux). It might even run better, because in such a case Windows is talking to a "Windows friendly" VM.

Domino

[cdimart]

The only thing that I am waiting for is a Linux Domino Client and Admin Client (not iNotes). One would think IBM could get this taken care of.

Re:Domino

[Jaegar]

You're not the only one. If IBM ever started/finished that project, I could finally replace Windows on my work machine. IBM seems to take their sweet time when releasing one of their products on Linux.

Actually, I couldn't remove Windows from my work machine. I still need it for Battlefield......I mean......intensive graphics benchmark program. Yes. That's it exactly.

Re:Domino

[mangu]

The only thing that I am waiting for is a Linux Domino Client and Admin Client (not iNotes)

Lotus Notes works fine for me under wine. Wouldn't other Domino programs work as well?

Re:Domino

[cdimart]

I would like to avoid using Wine and go native. Given IBM's Linux push, I would have thought this would have been done ages ago. Maybe I should go back to Warp 4. ;)

Re:Domino

[IamTheRealMike]

Nonetheless, Domino Administrator does work on Crossover at least. I worked on it a bit. The admin console crashes apparently but the rest works. Notes itself also works pretty well.

Re:Domino

[Anonymous Coward]

I wonder if IBM will have any interesting announcements at Lotusphere (wink,wink)

Cut Dell some slack!

[stratjakt]

The outsourced tech support probably couldn't spell "Windows" either. They don't even have the same letters on their keyboards as you do.

If someone called you up and asked you to spell some random word in Hindi I bet you'd mess up too.

As for the first topic, it should be no shock to any one that linux needs a whole shitload of stuff, Samba and others are great projects, and provide a lot of the desired functionality, but getting them installed and set up and "playing nice" with your Windows network can be a real bitch.

I mean, who here has jumped through the hoops of adding a linux server to an AD domain? Compare to adding a Windows server to an AD domain. Now imagine Betty McOfficeGirl trying to follow some written instructions to set up her fancy new linux desktop. Not all offices have a team of IT guys swarming around taking care of everything. Most people are on their own.

Linux needs to fight this battle in the small businesses of the world. They got a toe in the door as far as POS machines and kiosks, that type of thing. But linux needs to be running on the PC in the back office of every mom and pop grocery store or restaurant or doctors office, etc...

Everytime I criticize linux I get modded down and shouted at by morons for being a MS "fanboy" or "astroturfer". It's all obvious to anyone who cares to look, though.

Frankly, I don't think linux can do it (replace windows). I don't think linux will do it. I don't think we should be trying to shoehorn Windows compatibility into a Unix clone. Linux' strength comes from its Unix roots, and I think it should stay close to them, and stay focused on conquering the backend.

I see something like ReactOS developing into the horse to bet on.

To me, a Windows killer is something you install over some guys copy of Windows, and they never even notice that some of the icons are in different spots, or the Windows logo is replaced with something new. Everything works as it always did, albeit with all the transparency a GPL'ed project gives us.

Just my 0.02. I really don't think linux could ever replace Windows any more than a tractor trailer could replace a honda civic. All those regular non-mechanical folk don't want to drive a tractor trailer, and don't want to learn.

Re:Cut Dell some slack!

[stratjakt]

See, I called it.. Flamebait.. Waste some mod points on this message too.

Listen here, no flavor of unix will ever be a "windows killer". Just like no flavor of gameboy will never be a "PS2 killer". Why are you all so friggen blind?

Re:Cut Dell some slack!

[Kphrak]

Because the same kind of people as you laughed at the idea of a PC operating system ever growing into something that could compete with Solaris workstations. Or that a gameboy could ever be as powerful as a Super Nintendo. Incidentally, a couple of generations of gameboys from now, a gameboy might actually be comparable with a PS2.

The Linux desktop might not be ready yet, but that does not imply that it will never be a Windows killer.

Re:Cut Dell some slack!

[killjoe]

Windows does not play nice with other operating systems. If you have an operating system in your enterprise which refuses to place nice with other operating systems in your enterprise you should get rid of it.

Samba will never seamlessly integrate with AD. The minute it does MS will change AD. If you pursue that goal MS will keep shifting the ground underneath you and keep you off balance. We should not even try.

Get rid of AD, move to something else that works with ALL your operating systems, get rid of exchange and move to something that works with ALL your operating systems. Get rid of office and move to something that works on ALL your operating systems.

Windows is "just another operating system". If you keep on insisting that your software only works on one of your operating systems then you are doomed. Doomed to be vendor locked to a company which will jerk you around like a puppet.

Re:Cut Dell some slack!

[Tristan7]

Complain about outsourcing if you will, but when I send in a question to Dell tech support, they get back to me within 12 hours. Granted, it's in broken, formalized English, but it's quick, and it's helpful.

Re:Cut Dell some slack!

[tirnacopu]

who here has jumped through the hoops of adding a linux server to an AD domain? Compare to adding a Windows server to an AD domain.

I have. It's just as difficult. No, it's more difficult with a Windows machine versus a Linux/Samba one. Think of the following:
1. You need to change the IP address or DNS name of (one of) the domain controller(s). The first is VERY difficult, the second is simply impossible with Win 2000 (and VERY difficult with Win 2003).
2. Your hardware fails and you want the HDD plugged quickly in the first available PC-style machine, just to keep things going. Should the mainboard chipset differ a little (VIA instead of Intel or SiS), and you'll get a wonderful crash, not being able to at least boot it in safe mode.
Samba couldn't care less how and where you move it.
3. You need to add a nt4/2k/2k3 server to a 2k3/nt4/2k domain (just shuffle the configurations a little). LOADS of fun will follow.
Samba pays a lot of attention to legacy clients, but you can anyway upgrade each machine for a very good price ;)
4. You want to use several different OSs on clients/servers. All hell breaks loose when you get to the M$ ones.

Re:Cut Dell some slack!

[electroniceric]

Let me second this excellent point:

Linux needs to fight this battle in the small businesses of the world. They got a toe in the door as far as POS machines and kiosks, that type of thing. But linux needs to be running on the PC in the back office of every mom and pop grocery store or restaurant or doctors office, etc... ...
I don't think we should be trying to shoehorn Windows compatibility into a Unix clone. Linux' strength comes from its Unix roots, and I think it should stay close to them, and stay focused on conquering the backend.

There are a lot of areas will OSS is already strong and will get stronger, but an polished, integrated desktop is not one of them. OSS's pillars of strength are its openness, robustness, and the way that it commoditizes previously pricey functionality. We should build forward from there.

Savvy tech observers have pointed out that attacking Microsoft where it thinks you will attack is always a recipe for failure. They have spent millions upon millions of developer hours and billions of dollar producing a desktop that is tightly integrated. If anything, it is their key selling point. For disparate groups of OSS developers to try to accomplish anything like that is lunacy.

When you stick the UNIX method of laying one brick on another to build a wall, you get slow, robust development. The article is right that several individual bricks must be easier to lay mortar for - his points about version dependencies and driver installation are right on, and these basically about gluing one more thing on at a time. General support for wireless, for example, is now feasible, as the technology has started to stabilize, and represents one more piece to add on.

A great example of growth through commoditization is in the database market, from embedded systems to big iron installations. The vast majority of businesses need nothing more than a simple, functional database server. MySQL and Postgres on Linux are both close to stepping in nicely to fill this void. Postgres needs another brick layered on it - that being a simpler install process, but most other pieces are there.

As long as Linux sticks to its roots, it will grow slowly and steadily, and obligate other market players to react to its strengths.

Once again ... Why bother.

[JPyObjC Dude]

The truth is that Linux is no where near the quality and refinement of OSX. The only thing that is holding me back from recommending OSX as the switch to platform is the lack of native OOO and in particular a stable 2.0 trunk.

Once native OOO comes out next year, OSX will be the `switch` platform I am recommending to all my friends relatives colleagues...

Regarding Linux, OOO 2.0 is again a main switching point. OOO 1.1.n is still too limited to be useful for power users to switch.

Another HUGE blocking point for switchers to any platform other than win32 is the lack of a `all in one` netmeeting'ish application. Sure there is gnomemeeting but it still does not support an secure integrated vnc server/client p2p component. This is greatly limiting.

JsD
[dreaming of programming Java/obj-c/Python on an apple at apple while programming in Java/tcl/JS on a dell at another]

Re:Once again ... Why bother.

[slushbat]

Yes why bother ... posting a comment which nobody will understand? WTF is OOO?

Re:Once again ... Why bother.

[Anonymous Coward]

I assume he is meaning OOo -> OpenOffice.org.

Re:Once again ... Why bother.

[JPyObjC Dude]

As any Open source hacker should know...

OOO == OpenOffice.org

Re:Once again ... Why bother.

[Geccoman]

Open Office (.org) for those who might be wondering.

Re:Once again ... Why bother.

[TheRaven64]

I like OS X. My main computer is a PowerBook. I can honestly say, that I have found no other computing environment to be as easy to use.

This said, I would have trouble recommending Apple in a corporate environment due to the lack of a second source. If Apple licensed their OS and allowed third parties to make OS X compatible workstations (and could work out a way of doing this without going bust) then the situation would be slightly different.

Re:Once again ... Why bother.

[Anonymous Coward]

OpenOffice.org

Re:Once again ... Why bother.

[Anonymous Coward]

Hehe oops only 2 hours late. Refresh is your friend. >8p

VPN

[Anonymous Coward]

One major headache for us would be out of the box PPTP use. Most of the issues Tom mentions would go a long way to start phasing Linux into the environment I am responsible for.

Re:VPN

[stratjakt]

Linux needs out of the box L2TP, not PPTP. PPTP is insecure and shitty, MS abandoned it. L2TP isn't perfect, but it's better.

Of course, if the OpenVPN client for Windows worked better (no friggin WinPcrap dependencies), and the architecture on both sides better supported dynamic "road warrior" scenarios, it could render the whole issue moot.

Re:Just keep using Windows

[webzombie]

Sure... like just
- keep fighting terrorism
- losing indivdual freedoms
- stop thinking

For christ sakes... just because something isn't point click and done doesn't make it any less viable.

1. Windows Network Neighborhood visibility and UNIX/Linux visibility in the same panel.

XANDROS 2.0, Lindows, Lycoris, MEPIS

2. Active Directory password management which includes single sign-on and password expiration policies.

Novell Evolution embraces mail, calendar and address book standards to ease data sharing.

Supported mail protocols include IMAP, POP, SMTP and Authenticated SMTP, as well as Microsoft Exchange 2000 and 2003. Novell GroupWise support is currently in our development branch.

3. Interoperability with Exchange 5.5 and Exchange 2000.

See above

4. Font compatibility with Microsoft Office and Openoffice.org and/or StarOffice.

Crossover

5. Windows Terminal Server clients using RDP out of the box for home grown applications and special Windows applications.

Xandros, Lycoris, SUSE, RedHat... or just install VNC...

6. Ability to click on a file in a Windows or Samba share and initiate the associated application.

Fille association is not a roadblock. Simply a minor configuration issue.

7. Device management for hardware compatibility.

XANDROS 2.0, Lindows, Lycoris, MEPIS, RedHat, Suse

8. Compatible Windows Media player Codecs.

Crossover, MPlayer, XINE

Fundmental hardware problems

[samberdoo]

Until the common Linux flavors have better support for USB and Firewire devices, people will keep getting frustrated with these implementations. Trying to fit in to the wintel box way of doing things is a slow process. It's been evolving for over 20 years. Now if Intel started to actively support Linux.......

Re:Fundmental hardware problems

[Anonymous Coward]

Until the common Linux flavors have better support for USB and Firewire devices...

Wow, I didn't realize that this was still an issue with the major distros. I loaded Fedora onto my PC with a USB scanner, a USB laser printer, and a USB flash-memory reader. All three worked flawlessly, right off the bat.

What devices are causing problems, anyway?

Re:Fundmental hardware problems

[Anonymous Coward]

Wow, I didn't realize that this was still an issue with the major distros. I loaded Fedora onto my PC with a USB scanner, a USB laser printer, and a USB flash-memory reader. All three worked flawlessly, right off the bat.

I call Bullshit, unless you call having to edit /etc/fstab and manually mount your flash disk "working flawlessly"

Re:Fundmental hardware problems

[Anonymous Coward]

This worked fine for me. YMMV.

http://users.actrix.co.nz/michael/usbmount.html

Re:Fundmental hardware problems

[jadel]

I'm currently trialling Fedora Core 3 for a possible deployment, it incorporates HAL and Udev which allow automatic mounting of CD-ROMs and USB storage devices. I haven't tried any firewire drives yet though.

Single sigon

[Anonymous Coward]

I was charged with implementing single signon between our fedora box and ADS. Just use Samba, winbind, and pam. If someone wnats to configure these tools into a single offering, I'll use it, but it didn't take too long to implement.

Mattrock

Re:Single sigon

[buchanmilne]

Yeah, so I've been doing winbind (without Kerberos) on Mandrake since samba-2.2.2, and Mandrake 10.1 or 10.0 Corporate will setup auth to ADS (ie with Kerberos) during install or using drakauth, but that's still not single sign-on.

Single sign-on means you enter your password once per session, not that you enter the same password 15 times during the day.

In other words, it's the full Kerberos thing, including support for Kerberos auth in all the preferable mail clients, web browsers, network browsers (not just smbclient -k), FTP clients, IM clients etc etc.

We're not there yet ...

Re:Single sigon

[dbIII]

We're not there yet ...

ssh and X windows - we've been there for many years.

I have users that work on a dozen hosts daily, and they only have to enter their password once - keys handle the rest. "ssh hostname" and they are in as the same user they logged into on the local machine.

Also, if you take the basic step of screensaver screen locking, you DO want the user to enter their password several times daily - but only to do the initial login and to unlock the screen.

In other words, it's the full Kerberos thing

MS didn't invent Kerberos or LDAP (Active Directory is a subet of this), it was available on all breeds of *nix before they used it - just a pain to set up. If little knocked up in a weekend application B that only compiles on linux can't work with it you use something that can.

Re:Single sigon

[buchanmilne]

I have users that work on a dozen hosts daily, and they only have to enter their password once - keys handle the rest. "ssh hostname" and they are in as the same user they logged into on the local machine.


Oh, then I guess I should be using that. Oh wait, I am ... (and happen to maintain keychain - among other packages - in a popular linux distribution, which makes using passphrase-encrypted keys much more convenient). I hope you're not using passphrase-less keys ...

But, I hope you're not suggesting users should ssh in to the mail server to read their mail, ssh in to the proxy server to use the internet, or ssh to the authentication server as root to add/remove/modify user accounts.

MS didn't invent Kerberos or LDAP

That's right, but I didn't say that. MIT invented one, UoM the other. I am sure you knew that though.


(Active Directory is a subet of this)


Misconception. AD is a lot more than LDAP and Kerberos ... (and so, BTW, is NDS/eDirectory).


, it was available on all breeds of *nix before they used it - just a pain to set up.


Assuming you mean LDAP + Kerberos (and not AD), that was my point. Seems you missed it. We're not there yet, because it is a lot more pain to get *all* the same features as AD provides out-the-box working on *any* Unix (except maybe OS X ... but I am not sure).


If little knocked up in a weekend application B that only compiles on linux can't work with it you use something that can.


Well, then tell my why there is only one GUI Unix mail client that supports GSS-API authentication (I am not sure how many of the console-based ones support GSS-API, but I suspect it's also only 1), no browsers that support GSS-API authentication to proxy servers, no open-source proxy server that supports GSS-API authentication (but the most popular one supports NTLM) and why Samba only recently (3.0.6) got sufficient Kerberos support to even contemplate adding support for some of the basic features in Windows 2000.

Sorry, but, we're not there yet. Anyone who thinks we are either doesn't know where "there" is, or doesn't know where we are.

OpenLDAP + Heimdal + nss_ldap + pam_krb5 + NFSv4 gets us a lot closer than NIS+NFS (for a Unix-only solution with similar features), but still not close enough.

Don't get me wrong, I deploy Free Software solutions, but they're still not as good as they should be to compete. I'm doing my bit ... but there's a lot of work to do. For example, there is currently no equivalent to Group Policy Objects ...

Re:Single sigon

[dbIII]

keychain - among other packages

Yes, it works well.

ssh in to the mail server to read their mail

They are not a bunch of secretaries, they log into a variety of machines with shell access to run various programs that sometimes run in parallel on several machines - so ssh replaces telnet and ftp, not mail readers etc. Since they use awk and grep all the time, please do not suggest some setup with a few dozen icons to launch programs on different machines - the MS way is in a totally different niche and doesn't come close.

*all* the same features as AD provides

If you need the MS specific features then you run an MS box somewhere on the network - sort of like the difference between using email and MS Exchange (non standard email plus lots of other stuff and other third party software to get it to run well).. If you don't have anything that needs to interoperate with different MS ways of doing things then you don't need it. Name one really usefull thing that you can do with Active Directory that you can't do with a well set up LDAP + (several authentication methods) setup that is useful in a non-MS environment - it doesn't come "out of the box" due to the different ways of doing things, *nix programs aim to do one thing well, and get something else to handle other things. As a consequence you can set up LDAP to be authenticated in a wide variety of ways.

When you think of *nix, you also have to think of more than linux and more than GPL software.

Re:Single sigon

[buchanmilne]

They are not a bunch of secretaries

So, you're saying Unix doesn't cater for secretaries? Unfortunately, we have secretaries. We don't want them all to have to use Windows, but from what you're saying I guess we should use AD and Windows only?

Since they use awk and grep all the time, please do not suggest some setup with a few dozen icons to launch programs on different machines

You're the one proposing solutions. But, many organisations need to cater to users who would be totally lost running awk.

If you need the MS specific features

I don't need features that are only feasible on MS products. I want to have more automated software management, easier tools to set default settings for groups of users (and not just initially). Take the example of going from not using a proxy server to using one (and not doing a transparent proxy server because you want authentication too). You can deploy the settings with AD ... you can't deploy the settings easily on Unix, even if you deploy scripts that set http_proxy etc.

sort of like the difference between using email and MS Exchange

Well, we would like enterprise calendaring, but we don't want Exchange either.

Name one really usefull thing that you can do with Active Directory that you can't do with a well set up LDAP

-push software updates to all clients (ok, so you could script this ... but it works out-the-box on AD)
-deploy user settings (ie the proxy example above)
-manage the rights users have on their workstations (ok, since sudo 1.6.8 it's easier since we can now deploy our sudo rules in LDAP ... but it doesn't really work out-the-box yet although Mandrake 10.1 almost has it)
-management of SSL certificates
-disconnected authentication (so your laptop users can also use LDAP). This is almost working with pam_ccreds and nss_updatedb, but there are still 3 isues).

Show me the typical Unix admin who can set up a full LDAP and Kerberos implementation with sudo rules etc. and some means of configuration management and software maintenance in one day ... and for each one of them I will show you 100 MCSEs who can do that and more with Windows.

The fact remains, it is still a lot more effort to get even the basic single sign-on featuresof LDAPv3 running on Unix than it is to get all the AD features working sufficiently on Windows 2000.

Re:Single sigon

[dbIII]

So, you're saying Unix doesn't cater for secretaries?

No, I'm simply saying that your "ssh to the mail server, ssh to the web proxy" comment was simplistic, sarcastic and making unwarranted assumptions.

Show me the typical Unix admin who can set up ... I will show you 100 MCSEs

It looks like you agree with me, you can do it either way. The difference is when the unix admin has an error, their rewrite their config, while the same admin (assuming equal ability in MS and *nix) may have to log on, off and back on to fix a problem (new machine added to domain) and other unexpected behaviour.

What I am talking about is what the USER sees, not whether someone that has a peice of paper that says Microsoft will let them pretend to be an engineer after a multi-choice test can set it up.

It may be a lot of work to use LDAP, but AD is not some magic new thing that has never been done before.

"We are good, but everyone is against us"

[Trurl's Machine]

If you read a lot of Polish press - I do - you will often find this kind of reasoning, especially whenever Polish national soccer team coach explains his latest failure (and in Polish soccer, there's always a failure to explain). My favorite is "we actually won the first half, but...". There ARE some important issues with Linux in corporate environment - laptop support, printing and device managing among the most important ones. Don't comfort yourself with easy explanation that corporations reject Linux migration only because someone is "tech-knownothing". Maybe they know something - namely that the overall cost of the whole hit-and-miss game with installing Linux on laptops might cancel the benefits of such migration?

Re:"We are good, but everyone is against us"

[jdray]

Kudos for that. I'm a definite proponent of Linux implementation, particularly in enterprise systems, but it's not ready for the desktop yet. Say what you will about them, but Microsoft has a fairly polished desktop product, even if the polish hides significant flaws.

One thing that bothered me about the article is that many of the complaints the author had about Linux in general seem to be solved in SUSE 9.1 (KDE 3.2) that I'm using at home. I seriously doubt that the likes of Mandrake and RedHat have much that's different, as I haven't heard any news flashes that SUSE had some hot technology that no one else does. I don't know about other wireless cards, but the Atheros-based built-in card on my ThinkPad works like a charm, and was autodetected on OS install. Printing seems to work well enough through the KDE control panel applet, and the driver for my Canon S500 was in the very long list that came with the system. Autodetection works there, too.

Wandering between wireless networks is definitely a pain, though. I'm sure there are better options than what I have (the depths of YaST) for ad hoc network switching, but they didn't come with the system, and I have to say that the Wireless Networks applet on my wife's XP laptop would be excellent if only it were about 10% more configurable and actually remembered what you told it. In general, its detection of networks is reliable, if not its ability to get connected to them. That's a little spotty.

Regardless, the article's author is generally correct, inasmuch as Linux needs polish to be accepted in the enterprise as a replacement for Windows on the desktop. Furthermore, it (or a distro) needs to address a slow migration path instead of a wholesale replacement strategy. That means interoperability with almost everything Windows uses now, plus all the added benefits of using Linux today, such as attachment to NFS shares on Unix systems.

Re:"We are good, but everyone is against us"

[pnutjam]

SuSE 9.2 is supposed to handle wireless better. I haven't tested it. A nice feature of SuSE for roaming networks is hardware profiles. I have one for home and one for roaming, but you can set up as many as you need.

I tested Mandrake, Red Hat (Fedora), Debian, Lindows, and several other smaller distributions. None of them handled wireless like SuSE did at version 9.0. I couldn't even get my wireless to work on most of the distributions, but SuSE picked it up by default.

This is backasswards

[RealProgrammer]

The way Linux will make inroads on the corporate desktop is not by some big push to get it over the top, but by steady, incremental improvement. Not to mention any names (lest I be accused of flamebaiting) but targeted super-projects will not work.

Reacting to the perceived needs of corporate users is fine, but that's not a good fit for the Open Source way. You need someone who has enough pull with a developer to get a single feature or bug worked on. In the early stages of a project, that person is the developer or people he knows personally, with the circle expanding outward as the project grows.

Companies with perceived needs for a Linux desktop can sponsor development of those needs. Sure, the rest of us can try to guess what to create based on surveys and hearsay, but it's way better for the people close to the problem to come up with the solution.

The best way to promote Linux on the desktop is with apps. If a killer app appears, people will adopt Linux and be motivated to fix whatever perceived flaws they find.

Re:This is backasswards

[cliffyqs]

that's how they made progress so far, word of mouth. It's hard to be patient when a product you see as inferior and more expensive is more widely used than one that is good, shows great promise, and costs less. Slow and steady may not "win" anytime soon, but it will build support and credibility.

sing with me: we shall overcome...

Are you sure about that?

[lukewarmfusion]

...and conclude that it's less trouble to buy Microsoft. Everyone loses in that scenario.

And Microsoft loses...how?

Re:Just keep using Windows

[pqdave]

The eventual goal might be to switch to an all-Linux environment, but in the meantime Linux users need access to all the things they use now. Microsoft would love it if the only practical way to switch to a Linux desktop was to throw out all your existing backend software and start over.

Re:Just keep using Windows

[Mr. Slippery]

What's the point of using Linux, 'just because'?

For one, not wanting to have your business rely on a single supplier, especially a criminal monopolist. Also better security and lower TCO.

Re:Just keep using Windows

[marsu_k]

I don't have quite any experience with Windows networks, but let's take a few of these...

3. Interoperability with Exchange 5.5 and Exchange 2000.

Evolution Connector (formerly Ximian Connector) was recently GPL'd.

4. Font compatibility with Microsoft Office and Openoffice.org and/or StarOffice.

I have no problems using TTF fonts should I want to.

7. Device management for hardware compatibility.

I'm not quite sure what this is supposed to mean, but there are several distributions that offer very good hardware detection and a "control panel"-type of view/configuration of your hardware.

8. Compatible Windows Media player Codecs.

Mplayer. Although their legal status can rightfully be questioned.

I'm sure someone with more experience on Windows<->Linux networking can fill up the rest. So I recon the question should be why use Windows, 'just because'? IMHO Linux would make a lot of sense on a corporate desktop (less fiddling with malware and viruses, no more solitaire). Home users, no, not yet. Soon though (I've heard next year is going to be _the_ year of the Linux desktop ;-)

Re:Just keep using Windows

[BJH]

Most of these already exist in one form or another.

1) Use smb:// in Nautilus.
2) Dump Active Directory and use something that's a bit more cross-platform. There's plenty of LDAP-compatible stuff out there, and Novell will sell you a drop-in solution for single signon. If you do it right, you get single signon across Windows, Linux, Solaris and HPUX.
3) Evolution Connector.
4) Just set OOo to use the MS TT fonts.
5) Terminal Server Client or rdesktop (I'm guessing they mean a RDP client here).
6) Nautilus can handle file associations just fine.
7) Not sure what the hell they mean by this.
8) Mplayer using MS codecs ;)

Basically, it sounds like a list drawn up by someone who hasn't considered that introducing a new platform into a corporate environment means that they're supposed to exploit the advantages of that platform, rather than force it to conform to whatever existing platforms they have.

Some Insightful, Some Not So Insightful

[Anonymous Coward]

Article Myth: Linux doesn't do P2P networking.
Fact: Linux just doesn't have a Net Neighbourhood/Places GUI. There is nothing that requires Linux (or BSD) to have to have a domain controller. In the past week, I've provided support in online forums where the problem is stated that on Windows they can't see the other Windows box - because they are using Network Places, which relies on NetBIOS and can take up to 45 min for a computer to show up in. This is the reality of the userbase - GUI.

Myth: Printing sucks
Fact: No argument - it sucks. No central tie-in into the system so all programs use the same printing config. I shouldn't have to setup CUPS, and then setup each and every program I want to use to use CUPS.

Myth: Laptop support is non-existant
Fact: There's sites dedicated to it; as long as the hardware is available, for the most part there is no trouble booting linux on a laptop. Rather, the article says that there's just not enough wifi support in laptops...

Myth: No Terminal Services client
Fact: rdesktop worked fine for years now

There's other issues, but those are the most visible. Not to say the article isn't overall wrong in it's assertion - that in order for Linux to get to the point where drivers are listed with hardware along with Windows, the hobbyist programmer mantra of "it works for me, so fsck you" keep stagnating Linux where it is today - where it's been for the last couple of years ever since "this will be the year of the Linux desktop...No, THIS will be..."

It's not acceptable to have to install 3+ programs in sequence to get an app to work - bundle the bloody stuff already, quit being lazy. Funny from the crowd who chastizes closed source about how bad their software design is...

Re:Some Insightful, Some Not So Insightful

[stratjakt]

Article Myth: Linux doesn't do P2P networking.
Fact: Linux just doesn't have a Net Neighbourhood/Places GUI. There is nothing that requires Linux (or BSD) to have to have a domain controller

It's all about single sign-on and "zero configuration". Sure you can manually configure user lists in 900 linux machines, or you could set up a seperate LDAP for linux and have AD for windows, and manually sync them. But thats not what businesses want. That's twice as much time to add a user in their minds.

As for the Net Neighbourhood thing, it's not even just that. I rarely use it on windows, but I'm always typing stuff like \\SPIKE\Shared\Sourcecode\ into the address bar of explorer.

The closest thing to getting that behaviour to work in linux is that LiSa daemon which I positively can't stand. I shouldn't have to tell some daemon how big my subnet is to be able to browse local shares. There should be no configuration needed to do that, it should work "out of the box".

Similarly, Samba should install with a default config that works out of the box, rather than having HOWTOs tell you to edit the smb.conf file by hand because only morons and idiots use swat. WTF is that? This is how we're trying to entice switchers?

Integration is lacking here. I should be able to right click and select "share via SMB" or something in my file manager to create a new share, just like we do in Windows. Having to edit a text file, then restart the daemons is kind of ridiculous.

Myth: Laptop support is non-existant
Fact: There's sites dedicated to it; as long as the hardware is available, for the most part there is no trouble booting linux on a laptop. Rather, the article says that there's just not enough wifi support in laptops...

Wireless is the major selling point of a laptop these days. It's the whole ballgame. IMO, wireless is the only thing a laptop is useful for with their tiny little cramped keyboards and endlessly frustrating touchpads.

Don't you see all those businessmen at the starbucks in the airport all wirelessly doing their important business? They didnt get the company to agree to buy them so they could sit there disconnected and play Tux racer (even if that's what they really are doing).

Myth: No Terminal Services client
Fact: rdesktop worked fine for years now

TFA is talking about a client, not a server. We need to be able to start a windows terminal session from a linux desktop.

I can tell you that I couldn't use linux on my desktop box at work for this very reason, I regulary have to connect to clients machines via Terminal Services, or PCAnywhere.

I may have some techie cred in our office, but I have no say in what OS our clients want to run, and I can't tell them to install VNC or anything.

Re:Some Insightful, Some Not So Insightful

[doj8]

Myth: No Terminal Services client
Fact: rdesktop worked fine for years now

stratjakt commented:
TFA is talking about a client, not a server. We need to be able to start a windows terminal session from a linux desktop.

I can tell you that I couldn't use linux on my desktop box at work for this very reason, I regulary have to connect to clients machines via Terminal Services, or PCAnywhere.

I may have some techie cred in our office, but I have no say in what OS our clients want to run, and I can't tell them to install VNC or anything.

Uh. RDesktop is a Windows Terminal Server client. I've used it to connect to Windows Servers for several years.

Re:Some Insightful, Some Not So Insightful

[bluekanoodle]

I agree with you on most the points here, except for rdesktop. Rdesktop is a client to connect to a terminal services server. Compared to Windows RDP app, or Citrix ICA Agent, it's pretty bare bones, but it does work

Re:Some Insightful, Some Not So Insightful

[Anonymous Coward]

As for the Net Neighbourhood thing, it's not even just that. I rarely use it on windows, but I'm always typing stuff like \\SPIKE\Shared\Sourcecode\ into the address bar of explorer.

As do I.
But here's the clue - we're not the typical user.

I recently moved a client from a Win98 network to a Server2003 domain - and every single person in the office lived in Net. Neighbourhood. Same with the recent people I've helped online in numerous forums.

This myopic view of "works for me" - the elitism will lead to extinction, not proliferation.

Wireless is the major selling point of a laptop these days.

Wifi nics have the same issue, laptop or desktop.

But development for laptops is harder due to cost to get the hardware - which is why it's only just now that OpenBSD has got SMP support.

TFA is talking about a client, not a server. We need to be able to start a windows terminal session from a linux desktop.

If you don't know, google it before responding.
Do yourself a favour and check rdesktop.org - it's a client.

Re:Some Insightful, Some Not So Insightful

[Anonymous Coward]

"As for the Net Neighbourhood thing, it's not even just that. I rarely use it on windows, but I'm always typing stuff like \\SPIKE\Shared\Sourcecode\ into the address bar of explorer.

The closest thing to getting that behaviour to work in linux is that LiSa daemon which I positively can't stand."

What are you talking about? Have you ever tried Konqueror? If you have a share up, you connect the exact same way. Where's the problem? You don't need LISA. You can even do this in Mozilla. I think you're making it harder than it is.

Re:Some Insightful, Some Not So Insightful

[WhiteWolf666]

Use SuSE. All of these problems are corrected.

Net Neighborhood is present on the desktop. KDE desktop, anyways. Also, you can right click on a folder, and share it----it appears as a SMB share.

Priting has a central tie-in in Yast. Printers are automagically configured, if they are USB plug-n-play. Network printers are automagically discovered. Sharing networking printers is easy.

Laptop support.
Blar Blar. Yast installs ACX100, as well as several other drivers automatically. Ndiswrapper is still not automatica, but some manufacturers have licensed linuxant's driverloader, which works nicely.

ACPI is still a problem, though APM works on laptops that offer it.

Terminal service: Yup, its a myth.

Re:Some Insightful, Some Not So Insightful

[killjoe]

"I may have some techie cred in our office,"

Wow, i'd hate to see what the rest of your office is like. Rdesktop IS a terminal services client. You can also check out what nomachine offers.

Re:Some Insightful, Some Not So Insightful

[lphuberdeau]

Article Myth: Linux doesn't do P2P networking. Fact: Linux just doesn't have a Net Neighbourhood/Places GUI. There is nothing that requires Linux (or BSD) to have to have a domain controller. In the past week, I've provided support in online forums where the problem is stated that on Windows they can't see the other Windows box - because they are using Network Places, which relies on NetBIOS and can take up to 45 min for a computer to show up in. This is the reality of the userbase - GUI.

Well, I have it KDE 3.3... and I think it has been there for a while.

Myth: Printing sucks Fact: No argument - it sucks. No central tie-in into the system so all programs use the same printing config. I shouldn't have to setup CUPS, and then setup each and every program I want to use to use CUPS.

Once CUPS is configured, all my apps use it. I had some hard times getting it to work under Gentoo, but it was out of the box using Fedora (it even has a friendly interface to modify configurations)

CUPS is acting like a server. There should be no problem using it remotely...

Myth: Laptop support is non-existant Fact: There's sites dedicated to it; as long as the hardware is available, for the most part there is no trouble booting linux on a laptop. Rather, the article says that there's just not enough wifi support in laptops...

It really depends on the laptop. If you buy a super-cheap compaq laptop, you WILL have hardware problems and incompatibilities.

As for wireless, it's a vendor problem. It's not easy to guess how their hardware works and make drivers for them. They should be the ones distributing the drivers anyway. Linux has everything needed to get wireless to work on supported cards and it's very well integrated in most desktop environments.

Re:Some Insightful, Some Not So Insightful

[evil_tandem]

mandrake 10.1 does this out of the box. i was amazed when i found it, and it worked exactly as advertised... with no configuration...

Re:Some Insightful, Some Not So Insightful

[Anonymous Coward]

the above post should have been modded up the parent was talking from distant memories certainly where suse is concerned if you were talking about 8.x then yes its a fair comment, but 9.x addresses issues heck there is a network nieghborhood just like in windows.

unless your familiar with present day versions you shouldnt be commenting as if you are- the grand parent is a troll.

WiFi - Debian - NC8000

[nathan s]

I recently installed Debian on my NC8000 laptop and spent the better part of a couple of hours last night doing the relevant research to get my built-in IPW2100 adapter going.

It honestly wasn't THAT difficult, but I say this with the reservation that I'm a fairly advanced user despite the fact that I primarily use Windows. I can see how someone with limited *nix experience and who lacks familiarity with a CLI might find it nigh-impossible.

The sad thing is that I have to do it all over again, since I only allocated 1.5GB to / on Debian and now it's choking every ten minutes on me (since I have gnome running and it insists on having 10,000 useless packages or else it will go away and sulk). :-( You know what they say about hindsight...

Re:Just keep using Windows

[jimicus]

1. Windows Network Neighborhood visibility and UNIX/Linux visibility in the same panel.

Check. It's called Samba.

2. Active Directory password management which includes single sign-on and password expiration policies.

Check. It's called Samba with Winbind. Though it could do with being better integrated with most distributions.

3. Interoperability with Exchange 5.5 and Exchange 2000.

http://www.novell.com/products/connector/

4. Font compatibility with Microsoft Office and Openoffice.org and/or StarOffice.

TrueType fonts work fine for me. Though again, a well-designed installation program would be nice.

5. Windows Terminal Server clients using RDP out of the box for home grown applications and special Windows applications.

http://www.whitepost.pwp.blueyonder.co.uk/rdp.png

6. Ability to click on a file in a Windows or Samba share and initiate the associated application.

Have they used Konqueror lately?

http://www.whitepost.pwp.blueyonder.co.uk/useprog. png

7. Device management for hardware compatibility.

One already exists, it just doesn't (yet) integrate to the point whereby it can install drivers automatically.

http://www.whitepost.pwp.blueyonder.co.uk/infocent er.png

8. Compatible Windows Media player Codecs.

Which ones? Xine supports most:

http://xinehq.de/index.php/features

Wuh?

[Wylfing]

Peer-to-peer networking, functional printing

This is a good list of criticisms....for me to poop on!

Really, I don't get these problems. Getting networked printing to work with Windows leaves you with red marks on your forehead from banging it against the desk. And P2P networks? Half the time Windows machines don't see each other in the "workgroup" or refuse to exchange information or doggedly insist on logging you in as the wrong user. In both cases, printing and networking, if something doesn't work correctly there's often nothing you can do to fix it besided rebooting and seeing if that helps (which, bizzarely, often does).

On the other hand, networked printing in Linux amounts to selecting a CUPS or Samba printer and clicking OK. Oh, and you might have to specify that it's an HP G85. How is that not functional? I think I took one step to set up my OfficeJet as a shared CUPS printer, which was "apt-get install hpoj". P2P networking, uh, come on you must be kidding me. We had this nailed before Micros~1 even knew what a network was. And with the interfaces now available in Gnome and KDE, traversing networks is almost transparent. "Sharing" is even very Windows-like in KDE (right-click and choose share). How's that not functional again?

Now as for the others -- AD support? That's rich. Not exactly parallel, but where is, for example, the support in Windows for ReiserFS, ext3, and JFS? I say that makes Windows "not ready for the corporate desktop" because it can't read non-MS filesystems. A clever driver-getter would be spiffy, but in Windows it's merely an agreement with hardware manufacturers to bundle/offer their drivers. This would be a reality for Linux if hardware vendors had open source drivers available, so really it's not a Linux shortcoming at all but a hardware vendor problem.

Re:Wuh?

[barrkel]

apt-get install hpoj

That's too much information right there. AD obviates any nead for this.

AD support? That's rich.

Do you know what AD is? What comparable enterprise-wide technology comes with linux and windows support and is extant in the wild?

Re:Wuh?

[yoz]

Half the time Windows machines don't see each other in the "workgroup" or refuse to exchange information or doggedly insist on logging you in as the wrong user. In both cases, printing and networking, if something doesn't work correctly there's often nothing you can do to fix it besided rebooting and seeing if that helps (which, bizzarely, often does).

You're right, Windows Workgroup networking utterly sucks. I ccontinually wonder how something so successful and so mature can still be so hideously unreliable.

However, your criticism is irrelevant, because almost all large enterprise installations of Windows use Domains rather than Workgroups, which work a hell of a lot better.

Not exactly parallel, but where is, for example, the support in Windows for ReiserFS, ext3, and JFS? I say that makes Windows "not ready for the corporate desktop" because it can't read non-MS filesystems.

Oh, please. "I'm not driving on the wrong side of the road! Everyone else is!"

Besides, anyone who's going to access those filesystems from Windows in a corporate environment is going to do so over the network, so filesystem drivers are irrelevant. You either want Samba on the UNIX side or NFS on the Windows side.

Re:Wuh?

[eV_x]

You've never actually run a business have you?

Companies do want to lower costs, but all things require some level of skilled workers. So simply eliminating licensing costs does not mean something has a lower TCO. This is a very common misconception by people here on Slashdot - thinking that no licensing costs means no cost.

That's only part of the picture and here's why. You have support costs and contracts - that's why distributions like RH have enterprise versions. These are not free, but are generally considered a requirement for any large business. Second, you have resources to support training, installtion, and management of what you deploy. Imagine simply rolling out FC3 to 10,000 people in a business without training. These are people who have, like it or not, learned the Windows way of doing things, including the OS and Office. OpenOffice is not nearly the same, and neither is KDE.

So, first we have to train 10,000 people - well that involves training instructors or paying consultants to train. That also involves the cost of taking people to training and not working on things they normally could do. It's a loss in productivity, so you stage it over a year or so to reduce the impact to the business, possibly.

Now, we haven't even deployed yet - that involves physically touching 10,000 desktops and completely migrating them over to a new environment, and ensuring you haven't lost user data. If you've ever been part of one of these, you'll know that this is also timeconsuming and costly.

Last, then we support the releases, put in infrastructure to push new patches out (yes, Linux has plenty of them), and then managing those desktops as easily or easier than Windows. This can be expensive and the easier it is to manage bulk loads of desktops remotely, the better. Yes, you have this with Windows, but now you've got to either re-tool and existing staff that was probably Windows based before in server management or hire new people - depending on your geographic area, you may have less people skilled in those areas than you would with Windows simply because Microsoft is so prevalent.

The point is nothing is free because these are not single user instances at a person's home. Thinking that licensing costs are the major portion of desktop management is a fallacy. They may be when you're a small to medium sized company, or when you're starting from scratch, but the costs of migrating can be very problematic and require a lot of time.

Re:Wuh?

[iabervon]

I think Linux printing is ready, but just recently. That means that things don't come preconfigured to use CUPS yet, which means there's significant setup effort.

The way things go with Linux is that things start out unsupported. Then they get flawed support. After a bunch of development, the right solution is made, but it requires a lot of configuration to set everything up. Then it comes preconfigured and everything just works.

(When I started using Linux, in '96, in order to get X working, you had to write a mode line with the timings you wanted to get things just right. Then X started coming with mode lines for all the nice modes. Now you don't need mode lines at all; the server will come up with the right information itself. Imagine my surprised when my new X server, with nothing in the config file other than my monitor's capabilities (old monitor; new monitors report their own capabilities), instead of coming up in 1280x1024, came up in 2048x1536 because that's what it could do.)

Today you have to tell CUPS what your printer is. But tomorrow, you won't because the software will read /sys/bus/usb/drivers/usblp/*/../product, and look up the right info. Then it will look in /sys/class/printer.

The article is thinking in the microsoft way about getting drivers. Why should you have to click on an unsupported device in order to get a driver for it? Just try to use it and it should fetch (or build, or just load) a driver. If it doesn't know what the device is, it should use a cddb-like system to report the lack of support, and let users who get it working report what they did.

Re:Just keep using Windows

[PoprocksCk]

Well, when looking at the above list, I can't help but be frustrated. The majority of those things are already available. Let's go down the list item by item:

Windows Network Neighborhood visibility and UNIX/Linux visibility in the same panel.

Huh? What are these people using, FVWM? With Samba it's easy to set up a Windows network on a Linux box that can be viewed on both GNOME and KDE. In the same place as Windows shares. GNOME (and probably KDE, not sure) can even display different manual networks, such as FTP servers in its network place.

Active Directory password management which includes single sign-on and password expiration policies.

Can't comment on this, I'm not familiar with Active Directory.

Interoperability with Exchange 5.5 and Exchange 2000.

Am I completely crazy, or can't Ximian Connector & Evolution already do this?

Font compatibility with Microsoft Office and Openoffice.org and/or StarOffice.

Again, I ask the same question -- "huh?" -- if you want to use the Microsoft core fonts, install them! It's not that hard. It's not a fault of OpenOffice.org or StarOffice, it's just a case of the fonts that come on a Linux distro by default -- there's not Arial, Times New Roman, etc. because those are Microsoft fonts and Linux distributors can't distribute them. Might I ask a daring question: why don't Windows users install the Bitstream Vera fonts? I find it annoying that "Microsoft Office" doesn't have compatibility with "OpenOffice.org" (even though the office suites are not the problem in the first place).

Windows Terminal Server clients using RDP out of the box for home grown applications and special Windows applications.

Again, excuse my ignorance, but ... what's wrong with VNC? Why not switch to an open solution?

Ability to click on a file in a Windows or Samba share and initiate the associated application.

I don't agree that that's the problem: KDE (and GNOME maybe, I'm not sure though) can open the desired application just like normal but it does it in an undesirable way, IMHO -- it doesn't open the file from where it is, it copies it to your home directory and opens it from there. I think that that should be improved.

Device management for hardware compatibility.

That's very vague. Do they mean a GUI? If so, what's wrong with distro-specific hardware GUIs such as YaST (which is very good IMHO). A universal distro-independent solution is not a good idea, as is exemplified by LinuxConf. If you want a GUI for hardware management, pick a distro that has one.

Compatible Windows Media player Codecs.

That's the dumbest one yet, and the answer's right here: http://www.mplayerhq.hu/

Re:Just keep using Windows

[Anonymous Coward]

Is your LDAP kerberized? (in other words, can you tie Kerberos security to permissions on the retrieval and setting of LDAP attributes?)

Or is it just another Linux mish-mash of pieces and parts that could be tied together with string, bubble gum and a paper clip?

All my corporate customers prefer the bubble gum and paper clip approach.

What about HP laptops?

[drgreg911]

...no one offered Linux preloaded on laptops.

A while back HP started offering SUSE pre-loaded on one of their business-oriented laptops. Is that still going on?

Re:What about HP laptops?

[Jaegar]

Yes, you can order the nx5000 with Suse 9.1 on it. I believe it was also supposed to come with OpenOffice loaded on it.

However, when you try to configure a system like that, you'll notice in nice bold letters at the top: "HP recommends Microsoft Windows XP Professional".

roaming profiles

[cliffyqs]

that would be nifty... we have no such thing. Outlook gets mail and stores it on the local machine, removing it from the server. this may have been done because of server space in the past, but now it is a pain if a workstation dies. spares are a pain. users exist on the server (for email) and on local machine, everything on the local machine. not many computers to support here, but enough that I would like it to be easier.

don't be like us, plan ahead for time & cost of support.

Re:roaming profiles

[YodaToo]

If you move to IMAP, it will take care of the mail portion of your PST crash-and-burn risk. Doesn't help with the "other" data though. I use Evolution and store my data in my NIS /home/users directory. I have backups of /home/users exclude FireFox caches and the IMAP cache to keep unnecessary junk out of the backups and reduce the media required to store.

"Linux desktop could fail" WTF?

[Anonymous Coward]

What is up with these suits talking about "failure" if Linux doesn't meet their requirements. This is yet another instance of applying a corporate mentality to a non corporate entity. He doesn't define fail because its a meaningless term. The majority of the OS users in this world are home users, not corporate users. These latter users want a stable system that doesn't force them into unnecessary purchases. There is no fucking way Linux is going to fail with this crowd. If the corporate stooges don't need it, that's fine. It doesn't have jack to do with failure of the Linux desktop.
But here's where the failure comes in. If your company's IT budget eats up your marketing or manufacturing resources and your competitor saves that money and beats your ass in the market then guess what Mr Tidy Tie ---you fucking failed.

is the man an idiot?

[Lumpy]

peer to peer networking? that is a frigging nightmare to any enterprise It department.

why do you think that we lock out the ability for you to share folders on your machine in the domain wide settings??

Gee, that's all we need is something that makes it easier for a Virus to spread.

the author has some interesting points but most are examples of someone writing an article about something he knows nothing about.

I admit that some things are slightly laking, but some of what he talks about exist (sorry, but single login to a active directory? that's a windows product why not use the unix solution like this that has existed for decades?)

These are the types of articles that simply Fuel the PHBs into thinking they now know more about a subject.

Re:is the man an idiot?

[bluekanoodle]

"sorry, but single login to a active directory? that's a windows product why not use the unix solution like this that has existed for decades?"

I have to diagree with you. Getting linux in the the business is all about conversion. You can't change peoples minds overnight. You can't expect a company to throw out their entire infrastructure just to save a few bucks on desktops, especially when that infrastructure is already paid for, and it's working for their needs.

These are the types of statements that simply fool the "i'm a geek, and everybody else is stupid" crowd into thinking they now know more about a subject then they really do.

Re:is the man an idiot?

[Anonymous Coward]

Sorry but the Cray supercomputers, Silicon Graphic workstations, and OSX macs can not connect to incompatable software products either.

It's a stupid statement like, "this BMW oil filter will not work on my Dodge! So BMW's must suck!

It's called using the right tool for the right job. You want linux on the frontend? Then you HAVE to have linux/unix on the backend.

These are the types of statements that simply fool the "i'm a PHB, and everybody else is stupid" crowd into thinking they now know more about a subject then they really do.

get a clue and when you become an integrator and know what yoo are talking about, please come back and tell us.

Many of us HAVE been in the trenches and doing this, lumpy included from what I can gather from his past posts, so I think someone that is actually doing this would certianly know more about it that say someone like you.

Re:is the man an idiot?

[bluekanoodle]

I never said that a linux front-end wouldn't work better wuth a linux backend, or that linux sucks. You really should learn to read.

I'm pointing out the stupidity of the statement, that companies should get rid of an infrastructure that's working for them in order to save a few bucks in desktop licensing.

That's like saying "I made this Oil filter using cheaper technology, and it's doesn't work with my Dodge, so I'll go buy a new car that it does work with."

The fact is, that if you want to have your product make inroads in a company, you MUST give them a compelling reason, and you need to make sure it fit's in with their overall scheme.

Don't like it? Fine, then don't complain if the company chooses not to use your product.

I like how you attacked my professional skills without knowing the first thing about me or what I do. That just shows the pure zealotry of your argument, that you have to resort to assumptions, mistruths, and plain old lies to make your point.

I have been in the trenches, hell, I've dug the trenches! I'd match my resume against an AC on /. anyday of the week.

Re:is the man an idiot?

[dbIII]

"sorry, but single login to a active directory? that's a windows product why not use the unix solution like this that has existed for decades?"

MS Active Directory is a subset of LDAP, a unix solution that has existed for some time.

Re:is the man an idiot?

[bluekanoodle]

I know, I was quoting the OP

It's a *nix wide problem though?

[gilesjuk]

Other systems are just as archaic in some ways when it comes to drivers and hardware support.

Some even require everything built into the kernel. Commercial Unix operating systems are sometimes inferior to Linux (dare I mention SCO's offering).

Re:Just keep using Windows

[Anonymous Coward]

Dump Active Directory and use something that's a bit more cross-platform. There's plenty of LDAP-compatible stuff out there, and Novell will sell you a drop-in solution for single signon. If you do it right, you get single signon across Windows, Linux, Solaris and HPUX.

Care to provide some basis for your assertion that Active Directory is not cross-platform? It is fully supported by MIT Kerberos, can be integrated into PAM, and is pretty damn secure.

LDAP solutions are NOT secure, as they pass credentials in CLEARTEXT. Yes, you can use certificates but now you've introduced the thorny issues of key distribution.

Microsoft's Active Directory has smartly tied Kerberos and LDAP together, so LDAP queries can be encrypted with Kerberos... so no certificate distribution problems and secure from sniffing.

Put that in your pipe and smoke it.

(Yes I'm posting anonymous because this post actually has something other than FUD about Microsoft, and I'd get killed by this crowd for telling the truth about Microsoft).

Device Manager

[RogerWilco]

I can realy adhere to his call for a uniform Device Manager/Driver (and therefore uniform driver API)
In my specific case I do not want to talk to /dev/ttyS0
I want to talk to a Modem and tell it to Dial 123-456 at 9600 Baud 8N1
using GSM network transperency. I want the driver to know how the modem should accomplish this. He also mentions printing, it's a similar issue AFAIK.
It's just an example, but a stable uniform interface and API on a higher level would make life much easier for a lot of independent software providers and hardware vendors to support Linux. This in turn would make Linux more viable.
The big question is:
Who would define such an interface and get the GUI and the Kernel people on board?

I know it's all about the age-old discussion between a fully free open easy to debug any modify system on the one hand, and a predictable stable consistent user friendly less open and free system on the onther hand.
It's between Ideal and reality, between "we do not need to replace MS" and "we should replace MS", between "the optimal" solution and a stable environment, between open source and closed propretairy drivers. It's about backward compatibility and ease of use.

I bought Civ:CTP for Linux from Loki in 99/00, I haven't had time yet to try it on my SuSE 9.1, can anyone tell me if it will work?

A 1995+ program for Windows will most likely work on Windows XP.

Re:Device Manager

[Adam Avangelist]

There is sysfs and libsysfs with namedev parts of the udev/hotplug system. Desktops are (or should be) building in support for HAL and D-BUS. libsysfs provides a generic interface querying library for sysfs, which has information on devices. Namedev allows you to independently name devices in correlation to a unique serial id, and base device file creation off that with udev. Hotplug acts as a metigator for event notification with the kernel allowing for device detection. HAL and D-BUS are freedesktop projects, HAL a backend for allowing a more uniform presentation of hardware to graphical enviroments, and D-BUS for interprocess communication. The udev system is already real technology, integration of HAL and D-BUS into your favorite Window Manager/DE, assuming you use one should take several months or longer.

Re:Device Manager

[RogerWilco]

Thank you I will check those out.

Re:Device Manager

[PipsqueakOnAP133]

I totally agree. The saddest part is that Linus himself doesn't want to stabilize the kernel module API, supposedly because it'll allow for more binary drivers.

I'm not a Linux guru by any means, but have craploads of experience across the board, and yet it still took me over 2 months to attempt get MythTV set up on hardware "known" to work..... and still have it fail. (yes, 2 months, with approximately 3 hours per day spent working on it.) (For those interested, it's a EPIA board with a PVR250. Eventually returned it for a Sempron/VIA-KT/Geforce.)

Everybody talks about "oh, it's easy, use apt-get ivtv, or emerge freevo (for a different project)" or whatever. But do you really think this is easy? Especially if freevo is masked out of Gentoo's tree, and ivtv appears to need stuff compiled into the kernel?

What if i'm working on an older machine with 1 gig of hd space? If my whole point is to avoid the bloat of Windows and I end up spending all my disk space on kernel source, the kernel, and swap...... all to just compile nvidia drivers?

Look at XP. Sure it sucks ass. Sure it has so many vulnerabilities that I now recommend Win98SE to all my friends because all the surviving viruses can't infect it. But it can load (sometimes) old drivers. When it comes time for some clueless newb to install a piece of hardware, a driver is a driver, and it'll install. Unlike this, "your kernel needs version 2.4.6-subtype2-redhat-fc2-9847-test1337-wtfdoesthi sdo" even when the only difference between that and what you're running is that you accidentally deleted a dash in the version line in the makefile.

Having a uniform driver architecture will make installs so much easier. Distros can share driver databases. You no longer will have dependency on specific how-to's for each distro. And finally, there can be a declaration of a uniform friendly config system for all the hardware devices. (really, am I going to take the time to memorize that I need a module flag of "tuner=5 cardtype=6" for bttv..... and must grep the bttv source to even figure this out?) Something as simple as having a XML document listing options and values and a simple UI generator like menuconfig ought to do it.(with then a GUI version with more data) If it requires more detail, like color calibration for example, then make sure the text version gets everything necessary to start up a GUI even if blue and red are switched and you're locked into 640x480 or whatever video funkiness exists.

You know, most likely the only way to get around this is to get the kernel developers to all agree to it, or fork the kernel entirely just to develop this with some wrapper for compatibility.

Re:Device Manager

[dbIII]

I bought Civ:CTP for Linux from Loki in 99/00, I haven't had time yet to try it on my SuSE 9.1, can anyone tell me if it will work?

Probably, it works on fedora for amd_64, which is a stretch the programmers wouldn't have thought of. If it doesn't work, then you just need a copy of the libraries that it complains it cannot find. Linux doesn't suffer from DLL hell, libraries have version numbers, and you can have dozens of copies of the same library with different versions if different programs need it. That is how I got Neverwinter Nights running - I got the rpms for the 32bit version of fedora for the right libraries and installed them - NOT with the upgrade option - you still want to keep your existing libraries. With other situations in the past (xjig and old versions of Staroffice) it was a mater of compiling and installing the version of the library that the application complains that it cannot find.

The downside however, is I haven't spent much time configuring the rest of the machine after getting Civ:CTP to work.

A 1995+ program for Windows will most likely work on Windows XP.

If that was true it would be very nice, but I have come across several that will not. There is a good reason that you still see a lot of NT4 boxes around, despite the utter pain of installing it on new hard drives.

no they don't

[flynt]

The Linux desktop could fail if companies continue to pilot programs and conclude that it's less trouble to buy Microsoft. Everyone loses in that scenario.

Microsoft certainly doesn't lose. And how do the companies lose? They just did a pilot study on cost effectiveness and determined Microsoft was the answer. If Linux was cheaper and better for them as a company, they certainly would have switched.

What you can do

[PigeonGB]

Quit being a whiny little bitch and contribute some code, documentation, consultation, or just shut the hell up.

Actually, shutting the hell up isn't going to help anyone. Speak up. Don't like how a program works? Let the developers know what you want. Feature requests are important. Found a bug? Speak up.

Shutting up only prevents the knowledge from getting to who needs it.

I understand the point of the previous post, but having a dialogue with developers is important. Mailing lists, IRC channels, etc all exist to help contribute to software which is made by community rather than marketing/legal.

Re:What you can do

[Ucklak]

RTFA. There is truth in it.

"If it doesn't work out of the box, you won't win the business. "

Having just experienced this, corporations don't care what it can do, they want something that will do. And you'd better forsee their expectations before delivery date.

Developers are considered tech support after the sale and that's what they see.

THEN, once the sale is made, feature requests become important but not as important as getting the sale in the first place.

Re:What you can do

[PigeonGB]

Still, just shutting up isn't going to help make the situation any better, right?

Re:What you can do

[Ucklak]

Yeah on that. My point was directed to the fact that corporate customers want a finished product that plugs into their existing system with the 3 month transition being the training for the employees.

Re:What you can do

[the_mad_poster]

Well, while that's true, the point is that these people on Slashdot seem to spend most of their time whining like a bunch of crybaby zealots because they're too lazy/incompetent to do anything useful. In that context, it probably would be better if they did just shut up.

Re:What you can do

[PigeonGB]

And I wasn't arguing against it. I was simply taking a single sentence out of context and commenting on what people can actually do that would help more than simply shutting up. It may have helped if I just put "Slightly OT" in the subject line.

I do agree. If a product isn't going to meet with requirements, no amount of whining will make it better. Even RMS had to use closed source Unix while making the Gnu system. If doubt that anyone would have just created stubs compilers and stub text editors and then said, Well we should use these nowhere-near-finished "programs" instead of the closed versions available because it is the right thing to do.

Similarly, free/open solutions exists, but if they don't do what the user wants, they won't get used. This is why I think Gnucash isn't being picked up by most people. Besides not being available for Win32 platforms (AFAIK), they don't support closing accounts directly. I love Gnucash, but that is a glaringly missed feature.

Re:What you can do

[SCHecklerX]

I agree. For at least one of the OSS projects I use on a regular basis (rox filer), I did not have the time to re-learn C in order to contribute, but I *did* make suggestions to the development team (and did contribute a couple of perl scripts). They implemented pretty much everything I suggested, and later implemented even more of it.

So, yes, if you have an opinion about how something should work, by all means let the developers know. It definitely makes the software better.

Re:What you can do

[chthon]

Comment on your .sig : I have 3984.58 bogomips.

Re:Just keep using Windows

[jmulvey]

Again, excuse my ignorance, but ... what's wrong with VNC? Why not switch to an open solution?

VNC is designed solely for accessing a console. Windows Remote Desktop allows multiple users to have sessions with a Windows Server (or a Windows XP client, although only 1 user on the console or Remote desktop at a time), while the console is locked.

WHY?

[Libor Vanek]

8. Compatible Windows Media player Codecs.

WHY???? Show me ONE big corporation which needs to play movies on the users desktops!

Re:WHY?

[bujoojoo]

Show me ONE big corporation which needs to play movies on the users desktops!

Simulation and visualization (read: play movies on the desktop) play a big role in manufacturing, engineering, medicine, construction (architecture and design), entertainment (advertising), et. al.

I don't think movies in the popular sense are what is being referred to here...

Re:WHY?

[rmayes100]

Actually the company I work for puts a lot of it's corporate training material (for things like ethics training, safety etc) on internal websites using media player. They could just as easily use a different codec but it's still a windows world for the most part except for a few developers...

Re:WHY?

[lavezza]

The company I work for (big, multinational, etc.) has training courses in Windows Media format. A lot of the new hire stuff is now done on the desktop after the employee is hired and not in instructor taught courses. Also, once or twice a month there is another Ethics/Legal type training video that EVERYONE has to watch.

Sure, these could be converted into another format. But that is just another line item that needs to be done to switch to Linux. Once the "Things we need to spend money on because they don't work in Linux" list gets to a certain number, migration just isn't going to happen.

Re:WHY?

[coachvince]

Just a guess; Sony, Time-Warner, advertising firms.

Come on, really; many big corporations work with video. Even if they use Linux/OSS, when they distribute the video to Windows users who do not have the ability to install their own players, then WMP is the built-in option. So, it would be an advantage (though small, in my opinion) to be able to play WMP files.
However, being unknowledgeable in that area, I don't know if that capability is legally available without paying a MS tax.

Re:WHY?

[phiala]

WHY???? Show me ONE big corporation which needs to play movies on the users desktops!

Is the US government big enough? I realize it's not a corporation, but much of our mandatory training has gone to online sources, most of which use some audio and video. Often if not always in media player formats.

Another example, probably also found in the corporate world - conferences with all the hot-shots that they want us to know about, but don't want to pay for us to attend. Either streaming video or archived video (or both) used so that we can be just as bored, I mean enlightened, as the attendees, but from our own desks. Again, media player formats.

Re:WHY?

[bluekanoodle]

Let's see. Here are some things that Corporatations need movies for. Corporate training videos Online training Advertising Review Promotional materials As far as answering your question, we made exactly these kind of materials at one of my previous jobs. Our Clients included: Intel HP Sharp GM Do I need to go on?

Re:WHY?

[brufleth]

Corporate webcasts. We usually get a couple a month. Some are just "our company is great" talks but others are "what is going on" talks and are worth watching. And my employer is the DEFINITION of "ONE big corporation."

Re:WHY?

[/dev/trash]

AOL Time-Warner?

Re:WHY?

[Anonymous Coward]

I too worked for a big multi-national corporation and we used windows media and quicktime for training videos, corporate webcasts, etc.

Re:WHY?

[Anonymous Coward]

• Intel HP Sharp GM

Some horrible, nightmare conglomerate?!

More than a smattering of IBM pages

[the_rev_matt]

Several thousand, actually, most of which are technical instructions on how to get [x] running on a particular Thinkpad model. Remarkably detailed, I've never had any trouble running linux on Thinkpads.

See also IBM Products Certified for use with Linux: http://www-307.ibm.com/pc/support/site.wss/MIGR-48 NT8D.html.

Re:Just keep using Windows

[C0vardeAn0nim0]

repeat with me:

backward compatibility.

again:

backward compatibility.
backward compatibility.
backward compatibility.

got it ? it may not be important to you, but some big companies have _decades_ of data stored in their systems, some of this data only accessible through aged proprietary apps written in clipper, cobol, VB 3.0, whatever (some of those only exists in binary form. sources are long gone)... heck, once i went to a stock brokerage office and they had an access 2.0 running under OS/2 (by M$ recomendation) because access 2.0 was the only thing their PBX supported, and they had by force of law to record every phone call, internal or external.

it's easy for me or you to ditch windoze from our home machines because we don't have such worries. most of our valuable data are stored in open formats or easy-to-break proprietary ones and in small volume. now try to imagine GE. GM. Siemens. Toyota. Citibank. US gov.

i'm old enough to remember the reluctance of compnies in migrating from DOS to windows 3.0, or moving away from wordperfect. it only happend when M$ word/excell became stable enough, with reasonably good WP/Lotus 123 converters. that was between 10-12 years ago.

now that linux is starting to mature as a desktop environment, companies can start evaluating it. but since IT people in big enterprises abhors sudden and traumatic changes (it can cost them mora than millions, it can cost billions if something goes terribly wrong), they'll firts demmand a high level of compatibility. then as old applications are phased out, compatibility becomes a seccondary issue.

a friend o'mine recently said me he was stuck with windows in his small company (he's owner and only empoyee) because of some old clipper apps. then i showed him flagship and sugested that he could run the DOS binaries in dosEMU while adapting them to compile under flagship. he did that and is pretty happy. he knew about linux desktop but delayed the move because of 10 yr old clipper apps. and he's only one. now imagine GE's 300.000 employees...

Interface

[cliffyqs]

which will bring us to: "Computer... Tea, Earl Grey, hot."

This post is a troll

[Anonymous Coward]

"Is that L-Y-N-I-C-S?"

You can carry on insisting on pronouncing it incorrectly, but give other people a break when they don't have the slightest fucking clue what you're talking about. Linux is pronounced linux.

OS X Meets Criteria, But Not Solution

[Spencerian]

Mac OS X meets almost all of the criteria that the article suggests for Linux compatibility... ...except that Mac OS X is not Linux. (That, and the Windows codecs, although the popular VLC application does the trick in all but the stickiest non-QuickTime codec.

So, taking a page from both Apple and Microsoft's business handbook, what can the Linux community "steal" from Microsoft and Apple to make Linux a stronger enterprise player?

Getting things from the Apple side isn't very hard since its resources come from the FreeBSD world, which is open source. Samba works great in OS X, which means stronger integration in Linux is needed to match OS X's performance, which I suspect does nothing particularly special.

Same is true for AD authentication. Mac OS X uses a plug-in its Directory Services that understands this LDAP-variant...surely this is something that would work in Linux, or does it lack a refined mechanism for handling multiple directory services as OS X?

Ximian already provides Exchange compatibility in its mail product, and Exchange 2000 works with IMAP provided that Outlook Web Access (WebDAV) is running. Special features of Exchange (and its Outlook client) may be missing, but Mac users are still missing features from Entourage, the successor to the Outlook client on Mac OS X, so this is not quite the biggie. Linux/Intel users can run VMware (as Mac users would run Virtual PC) to use the actual Outlook client if needed.

The Microsoft Office component is a toughie. Mac OS users have a genuine Office client. Microsoft knows that holding back creation of a Linux client would sap power from its enterprise drive.

No easy answers in this, really. I think, however, that Linux could use a central business owner, although I know its nature makes that impossible. But wait--isn't that what Apple's doing with OS X by licensing or using BSD components?

What if a company licensed a Linux distro and took the reins to make a Linux-compatible OS with the same functionality, but also the "one-click" simplicity, application strength, and security that Mac OS X enjoys in its Mach/BSD fusion?

Of course, we know that this appears to have been done, with Red Hat, et al. But has it really been done well?

Re:OS X Meets Criteria, But Not Solution

[duhgeek]

There's always Wine (or Crossover Office) available if there's no other way to avoid the M$-Office applications. As opposed to VMWare, you avoid the Windoze tax while still using the "killer" app(s).

Re:The cult of UNIX must die!

[Anonymous Coward]

"GUI good, CLI bad"
Sure. Now go away and find a directory of 10,000+ files and find all containing a capital M or a sequence of three digits in the filename. Then select any that were created between April and August of this year, and have not been modified since September. Now move them into another directory. Use a GUI filemanager to do this. Go on, do it now.

Database access

[Durango_44]

Huh. No mention of databases let alone the ubiquitous Access97 variety. Look, I didn't pick the database the company I work for uses--Access97 was here long before me--but I am stuck supporting it. And for better or worse, many small businesses have homebrewed their data management using Access.

When I have documented the business case to move off windows to Linux, we always run into the lack of a comparable application within the Linux/OSS community. Staroffice had it on its previous version, but that is gone now. The OpenOffice folks seem to be working on it, but it is not yet ready. The Boss looks at my suggestion of MySQL and sees lottsa money and time spent converting and training. The use of various JDBC and ODBC drivers make a conversation technically feasible, but I suspect that many in the small and medium sized corporate world need a one-to-one application capable of natively sucking in those .mdb files and running with them. If that was there, we'd start converting to a Linux desktop this afternoon.

It is surprising that the Consultingtimes ( article literally does not mention databases.

Re:Database access

[99BottlesOfBeerInMyF]

a one-to-one application capable of natively sucking in those .mdb files and running with them.

Google gives me about 15 different mysql front ends, capable of opening .mdb files. Commercially packaged, professional grade setups with support, run as low as $45 a license for linux versions, and have windows and MacOSX versions for cross-platform support and gradual migrations. Maybe all of the solutions out there are crap. I don't know. But maybe this is a non-issue.

Re:Database access

[PitaBred]

Access can use an SQL backend. Start the move there. Let Access be the frontend, have the forms, all that. Stop it from storing the data in it's awful file format.

Re:Database access

[ThousandStars]

I agree, and I already wrote a response to a similar post. The short version: Even though I don't use Windows on a day-to-day basis anymore, I still need Access. And I just don't see one.

The biggest threat

[codepunk]

The biggest threat are all of these wannna be linux admins that are deploying linux like you would in a windows environment. Linux is not windows and should not be deployed as such. The windows deployment model is to load the os on every machine to maximize sales profits.

Linux on the other hand thrives in the thin client environment. When you deploy thin client all of these arguments about patch management, usability, control, software installation all melt away. I run 200 desktops from a single server and spend about 5 minutes a week maximum doing any sort of admin work for these guys.

New rollouts I can use a modified copy of a slackware cd that boots straight to a thin client
environment meaning I can deploy most medium sized companies in a single day including the time to load the server. Not only that but the risk is non existant since I don't touch the original drives.

Good points, but......

[Anonymous Coward]

Once again, the little guys are ignored. I know of atleast 5 small businesses (that have corp. capabilities if necessary) that can deliver servers, desktops, AND notebooks with most flavors of linux preinstalled, and are more than willing to assist and/or completely configure the network for the customer/client. I myself have done this for 5 medium size businesses for pilot or replacement programs.

Articles like this are forgetting the smaller companies that in MANY cases have more competant technicians on hand to handle things of this nature, and thus the smaller companies don't win the business. They don't win because they are pushed under the radar of someone interested. They often have better hardware, better support (rarely needed since their hardware doesn't break), and fair prices for the assurances you gain. Granted there are exceptions to this since there are many smaller shops that have no business in the field, but if you do a little checking you'll find places like mine can handle your needs and have been around for 13 years.

Long story short, change the mindset. Pay a little more for hardware, and you'll get what you WANT not what someone things you should have. And you'll end up with equipment you won't have problems with. I haven't had a single peice of hardware returned to me in 2 years for faulty parts. Compare that to Dell's 10% failure rate on desktops and 27% failure on notebooks. Same goes with Inet services. Is there something wrong with a personal touch?

Ooooooh!, OSX

[Anonymous Coward]

I think OSX has has lots of OOO factor with all the eye candy.

Re:Just keep using Windows

[nine-times]

2. Active Directory password management which includes single sign-on and password expiration policies.

Novell Evolution embraces mail, calendar and address book standards to ease data sharing.

I think part of the issue is also the actual logging-on to the computer. You know, having your computer log-on authenticate to active directory. Yeah, I know you can do it, but it never turns out as easy as telling it, in the setup process, to authenticate to a Windows server. Frankly, even when I've gotten it working, it still ends up being quirky, but maybe that's just me.

Plus, since you brought up Evolution, it might be nice if the evolution login/password (when connecting to exchange) would automatically pull the Active directory login/password you used to sign in.

I can tell you that this is one of the explicit issues that has kept me from being able to talk my bosses into trying linux out on a few desktops. However, the biggest issue, right now, seems to be a generic, "What if....?"

What I mean is, it used to be that there were a lot of explicit problems with running Linux on the desktop. The interface was immature, or you can't run [insert application here] and there's not a comparable Linux application. Lately, though, that seems to be getting less and less the case.

What has kept companies that I've worked for from switching to Linux is the big, "Well, just what if some unnamed thing comes up that we can only do on Windows?" That's what I mean by a generic "What if...?"

It's not a real, current issue, but more a recognition that it's still a Windows world, and a lot of commercial developers make Windows-only software, and a lot of the companies we deal with use windows, etc. The result is, you get a generic uneasy feeling that something is bound to come up which will require Windows, and after you've spent all this time/effort/money to migrate to Linux, you'll just need to switch back.

I hope this will go away as time passes, and Linux, little by little, gains a greater user-base. Hopefully, there will be more successful cross-platform apps (a la Firefox, and to a lesser extent, OpenOffice) that will make the transition seem much less traumatic.

One thing to note-- what will make the transition less traumatic, I think, is not simply making a Linux-replacement for Windows functions. It will be making cross-platform replacements for Windows functions. Firefox is replacing IE on a lot of Windows user machines. We want OpenOffice to replace MS Office on Windows user machines.

I think that'll be the key to getting people comfortable using/selling Linux. If Windows users start using open-source products on Windows, then you can say, "I can switch your operating system to something better, and you'll barely notice the difference because you'll be using all the same applications." it be much easier to sell it.

This is why I'd abolutely *love* to see Evolution, for example, ported to native Windows/Mac versions. Evolution handles Exchange servers much better than Entourage (Microsoft's own Mac product), and pretty much on par with Outlook. There are a lot of businesses out there where half the users or more only use e-mail, a web browser, and office suite. You get them over to Evolution, Firefox, and OpenOffice on Windows, and you'll have a hell of a lot easier a time convincing them to move to Linux.

Marketing

[Anonymous Coward]

is the most important factor in pushing Linux, as with any other products.
The fact that people at Dell and many others don't know what it is is a clear proof of marketing failure.
I work in tech support myself and 80% of people I deal with are so ignorant that don't know what a power plug looks like and don't care of finding that out b/c they expect you to think for them since they pay you.
Why doesn't the Linux OS community get it that the MS's success with Windows was mostly due to their marketing efforts in putting the MS logo on every f---ing ad, box, paper, etc.
People need to be consistently brainwashed by every possible and impossible means that:
1. Linux exists
2. is the best
3. is free
4. there are lots of games and other programs for it
5. is secure
6. everybody embraces it (add company names etc)

Here's a BIG pitfall

[Anonymous Coward]

http://www.feyrer.de/g4u/g4l.html

Now that you know. What are you going to do?

Re:WiFi - Debian - OT

[Anonymous Coward]

An off-topic Debian tip:
Install as little as possible when installing desktops. For GNOME,

apt-get install gnome-core

And, debfoster is your friend when removing packages you no longer want.

"Is that L-Y-N-I-C-S?"

[Rheagar]

Dell is based out of Round Rock, Texas -- I picture their phone operators to look like Peter's neighbor Lawrence from Office Space.

"Well you don't need a million dollars to do nothing, man. Just take a look at my cousin, he's broke, don't do shit."

I wouldn't hold it against them for not being able to spell linux.

Clearcase

[Anonymous Coward]

Clearcase is big. Clearcase is nifty, and clearcase is widely deployed.

And it really needs an open network share to work (in a common configuration, anyway - it may be possible to configure it otherwise - but our ISD dept have not done so, and must have a good reason to have actively blocked the 'close all open shares' script.)

Are 26 letters in the alphabet too much?

[kardar]

Exactly. Or take a directory of files named file001 through file999 delete file001 and file002 and move the rest of all of the filenames down two numbers - do that with your mouse!

People who don't understand the CLI don't realize that it's actually a programming language file, but you only get to see one line at a time (default file size is 500 lines) - you can change that, you can search, edit, switch to multi-line mode, write little scripts, change the copyright year notices on all of your webpages on Jan. 1 with a simple one-line command, etc...

Extending the argument that GUI is better, what you're actually saying is more like questioning why we should use C or C++ or Java or basic or any programming language that has to be typed in, let's just point and click our way to writing software programs!

Furthermore, why should we even use the English alphabet or the keyboard? We can do everything we need to do with a mouse - that way, you get carpal tunnel sooner, which means you can quit your job and get paid more than half of what you were earning for having people do surgery on you!! Great idea, huh?

English is a language, just like Perl is a language, just like C is a language, just like any other programming language. And you know what? Bash, the Linux default shell, it is also a programming language. That's the whole point. You can type your business letters with your keyboard or you can type them with your mouse...

On top of everything else, if you make it so darned easy for anyone to do anything on a computer, you will outsource yourself to someone making minimum wage. Since when is being stupid a desired thing? I think we need to get over the idea that computers have to be "easy" to use - it's a skill - take 6 months to learn how to use a REAL operating system the right way - what is life expectancy - 65+ and growing? What is 6 months out of that to learn how to use a tool that you are going to use for the rest of your life?

Setting stupidity as a goal is counterproductive. I think it would be better if the goal were that computer operators learn the skills necessary to use something like Linux or BSD. The basics are something you only need to learn once - they are more like concepts - so if the computer operators learn how to use the computers properly, we will have a more realistic computing environment - because no matter how "dumb" you make something or how "dumb" you make a job description, it's always someone else's job to come and fix it when it doesn't work anyway. What I don't understand is why intelligence and skills aren't the focus - they are in almost anything else? You don't hear someone say - "They ought to make theoretical physics easier... it's too hard!", or "They ought to make Partial Differential Equations easier... they're too hard!". C'mon, people... get with it already.

Re:Are 26 letters in the alphabet too much?

[couch_warrior]

Oh, my, Gosh, this is incredible*** "Setting stupidity as a goal is counterproductive. I think it would be better if the goal were that computer operators learn the skills necessary to use something like Linux or BSD. The basics are something you only need to learn once - they are more like concepts - so if the computer operators learn how to use the computers properly, we will have a more realistic computing environment - because no matter how dumb you make something or how dumb you make a job description, ... They ought to make theoretical physics easier... it's too hard!, or They ought to make Partial Differential Equations easier... they're too hard!. C'mon, people... get with it already." It's no wonder linux is going nowhere. You UNIX nuts are incorrigible. You really do WANT computers to be hard to use. Hey, why not replace those "lazy" electric starters on cars with hand cranks, that'll sell big. And food processors - lets go back to hand cranked meat grinders, the way it oughta be. Give me a friggin break. Computers are a consumer appliance. They CAN be made easy to use, and only the ones that ARE easy to use will sell. Layoffs in the IT industry went up 60% last year - because of rigid control-freak basket cases who won't update their technical skills. UNIX is DEAD, and if linux doesn't unhook from the corpse of UNIX and become more user-friendly, it's going to end up in the junkheap of history with IBM OS/360 and UNIVAC EXEC-8.

Re:Are 26 letters in the alphabet too much?

[Shotgun]

Exactly. Or take a directory of files named file001 through file999 delete file001 and file002 and move the rest of all of the filenames down two numbers - do that with your mouse!

How often do you do that? And why? If I spent the 20 minutes to search the net for the one line script to accomplish this daring feat, would I have done anything except waste 20 minutes? (ie, is there any possible way I would remember it 6months from now when I wanted to do it again).

People who don't understand the CLI don't realize that it's actually a programming language file

And of those who do understand the CLI, most don't CARE that it is a programming language file. Yes, I could spend months memorizing strange cryptograms to move files around, but I could rarely do anything useful with it. NOTE: Rearranging an MP3 collection for the umpteenth time does not qualify as useful.

What is 6 months out of that to learn how to use a tool that you are going to use for the rest of your life?

A serious amount of effort that could be much better spent building my airplane, that's what.

I wanted to print some pictures of the Thunderbirds that my son took at the "Wings Over Wayne" airshow this past weekend. I first kicked up the Gimp. The first attempt gave me a big brown splotch of the upper corner of my picture with large square pixels. So I grabbed the documentation and was introduced to a lot of trivia on aspect ratios and conversion formulas. While all mildly interesting, I didn't really give a shit.

I transferred the pics to my wife's laptop running XP (which she HAS to have for her real estate business). Pulled up the directory with the pics, clicked on "Print this picture", and followed the prompts to print a page of wallet sized pics.

I doubt that I will print ANYTHING in color in the next year. I could have spent an hour with the Gimp documentation, but instead I got the picture printed and still had 30 minutes to work on a wing rib. I suggest you go outside now and then, possibly get a hobby that doesn't involve computers, so maybe you can discover why people want computers to be easy.

Re:Are 26 letters in the alphabet too much?

[kardar]

1. The files thing - it's not THAT particular thing, it's the concept - performing a series of operations, any operations, on a series of objects, in this case, filesystem objects. It's the CONCEPT - you can't do that with a mouse in Windows. Sure, it's easy to do some things in Windows, but it's EASIER to do much more complex stuff in Linux if you know what you are doing. A blender might be easy, but an industrial blender is EASIER for a particular job, even though using the machine might be more complicated. It's not the difficulty of using the tool, it's the difficulty of the job that the tool has to do.

2. You obviously don't understand the CLI. If you understand the CLI, you know it's a programming language. A simple one, but a programming language nonetheless. How could you not care about that if you knew it? It's the case of the sour honey - the bee's nest was too high up in the tree, the bear couldn't get to it - so he gave up and consoled himself by becoming convinced the honey was sour anyway. It's not that hard, trust me, you just have to take one step at a time. Bash is your friend.

3. I guess what I see is that perhaps many people feel that it's "not worth it" to learn GIMP. Your attitudes towards GIMP are the cause of your difficulties. Many individuals, for instance, would gladly take lots of time to learn Photoshop, because their attitudes towards Photoshop are more favorable and Photoshop has the "image" and the "coolness" factor. Truth be told, if you are a graphic artist, you need to know both. Comparing GIMP to some application that allows you print out wallet-sized photos isn't fair... GIMP is an incredibly complex program with intense capabilities - you're comparing apples and oranges. What YOU want is called the "Gnome Photo Printer". Google for that and you will have your wallet-sized photos.

I guess it just boils down to whether or not the whole point of the thing is to sell massive quantities of computers to massive quantities of people, or to produce a quality product that appeals to certain types of people, regardless of the quantities of people.

Seriously... if you like Windows so much, use it - but if the reason you don't like Unix or Linux is because OTHER people like the advanced capabilities it offers, or because "everyone who stands up for Unix never goes outside" then I would have to say it's a failure to understand what a computer is. A better analogy, perhaps is a stick shift, not a hand crank. Got a problem with my stick shift? C'mon. Your opinion counts just as much as anyone elses - so it's always best to respect yourself and make up your OWN mind - if you don't like Unix because it's got a "command line" - if that "command line" which you never have to use really just bugs you that much, and that's how you really feel - why of course you are entitled to that opinion. Don't let others' appreciation and respect for the command line drive you away from a superior product.

Re:Are 26 letters in the alphabet too much?

[kardar]

Ok. Don't use Linux for two years. Just as an experiment. Then, just try it again. Not good enough for you yet? Wait another two years. Then try it again.

The best thing about Linux is the speed at which the development occurs. Open source is a different development model - it allows something to "not die", something, which if it was proprietary, quite possibly might have died. Open source, more specifically, the GPL, ensures that Linux will continue to thrive just so long as there are at least a couple hundred people on the face of the planet that would like it to. It's totally different if you would like to keep working on and improving on that proprietary operating system or software program, but the company went out of business and the source code is locked up in a vault somewhere... this doesn't happen to open source, so open source won't "die". Besides, technically, it's NOT "Linux", it's "GNU/Linux". And GNU is a recursive acronymn that stands for "Gnu's Not Unix".

Linux is winning. You can realize that now, or you can realize that later. Meanwhile, Linux is going to continue to grow and get better at an incredibly fast rate of speed. So if you don't like it now, wait a year or two... by then, it'll be so easy to use you will have wondered what you missed - it's a moving target, a very fast moving target, so what you don't like today will be history next month.

Re:WiFi - Debian - OT

[Anonymous Coward]

Lovely tip I was also going to mention. deborphan is very useful too to find unused libraries.

Secondly, why not just resize your partition? SysRescCD is your friend.

More complex than that

[einhverfr]

Nearly every business I have consulted or worked for has depended strongly on line of business tools developed on Windows. There is no guarantee that these will work under Linux.

To deal with this problem I add af ew more steps between 3 and 4 above:

3.1 Review LOB tools to see which ones are most likely to run on Linux w/Wine, Crossover, etc.

3.2 Set up a testing system in the department which is expecting the fewest problems. Work on getting the existing LOB tools working here. Some don't work? Either pay someone to make them work under WINE or look at moving them to Python/GTK in the next version.

3.3 Once these are dealt with, roll out the tools to a selected team in the department. Get feedback and once all major issues are dealt with, roll out to other similar teams in the department. Wash, rinse, repeat until entire department is running Linux.

3.4 Repeat these steps for all other departments.

Understand that even with this process, it is likely to be expensive to do a total move to Linux, and that some legacy workstations may be left running Windows for some time. This is not a problem. It just takes some time.

My business offers technical support and consulting for all types of businesses and I specialize in migrating businesses to open source software. You can find more information at http://www.metatrontech.com

Look, it is neither inexpensive nor easy, but when done properly, you get a *tremendous* return on investment.

Re:More complex than that

[killjoe]

"Nearly every business I have consulted or worked for has depended strongly on line of business tools developed on Windows. There is no guarantee that these will work under Linux."

There may be people who are so locked into windows they can never migrate. There is a reason why the phrase "vendor lock" exists. If the company has made a concious decision to lock themselves into a vendor by using non cross platform tools then they may very well be screwed, they will never migrate from windows and MS will jack them around like puppet. It sucks when you are at the mercy of your vendor.

Having said that perhaps they can rewrite their application in java, C#/mono or a any scripting language. It may prove to be too costly in the short them but it might be better in the long term.

Re:More complex than that

[chthon]

I worked for a bank, and they had to migrate from WANG VS (talk about vendor lock-in) using ISAM files/COBOL, to HP/UX/Oracle/Forms/COBOL. I helped with the migration.

I think that there was even more disparity between WANG VS and Unix, than currently between Windows and Linux.

We got help from some consultants who had seen that there was money to be made by migrating people away from WANG VS to more open Unix platforms, and they had a tool which analysed the WANG VS COBOL applications and created new skeleton programs from them, and they had an API which simulated some necessary aspects of the original platform.

I think that people who know both Win32 and GNOME|KDE|GTK|Qt have a business opportunity here : to develop tools which make it easier for developers and software companies to migrate their programs to Linux.

You will never have a complete 1/1 translation from Win32 to the other platforms, but if all requirements and test specifications are written well, then probably 90% of the work can be done.

I think there are more similarities between the Win32 API and Linux graphical and system toolkits, than between WANG VS and any Unix system (especially since most people on WANG programmed in COBOL).

The rest is compilation and editing, testing and editing and hope that nobody wants to put in extra features before the application runs on the new platform.

Without MS Access-like functionality...

[fyngyrz]

...Linux bars itself from business. Linux DB dweebs consistantly underrate the power of the Access interface, mistakenly focusing on the limitations of the underlying DB engine - limitations that are irrelevant to the majority of businesses, which are small businesses.

Linux has awesome DB engines readily available - unquestionably - but that power is not accessible to your average office cube dweller. That is the genius of Access; simple DB applications are easy, while amazingly complex ones are still possible, given patience and time. And that is how many of the more complex Access apps are developed; more functionality is added over time, as needs change and applications a tested against daily experience. This is easily done, because - Access is easy. Get that right, and a whole new class of businesses could come over to Linux. Without it, I think trying to sell Linux into the small business venue is just pissing into the wind.

One other area where Linux falls down is input methods for other languages. For instance, try entering Korean in a Linux system set up for English, using Open Office. Good luck trying. Ami (the app that is supposed to enable Korean input) doesn't even begin to work. You end up having to hand-insert each character from a font table, which is numbingly slow. It is awfully hard to share Linux in this direction or that when you can't get the thing out of its English entry state. I have not had occassion to try to enter Chinese yet, but I don't look forward to it based on my experiences with Korean. Windows, on the other hand, "just works."

Don't get me wrong. I'm a huge Linux fan, but these things have been brick-wall problems for my companies (three of them.) I think other business owners have very likely run into the same issues.

Re:Without MS Access-like functionality...

[Dunkirk]

That is the genius of Access; simple DB applications are easy, while amazingly complex ones are still possible, given patience and time. And that is how many of the more complex Access apps are developed; more functionality is added over time, as needs change and applications a tested against daily experience. This is easily done, because - Access is easy.

And this is the problem. It's easy to get something going with Access, but then you have people setting up databases who have ZERO concept of how

• Software development works
• Code ought to be commented
• How software scales
• What networks do to database traffic
• What file contention of a single .mdb means
• What a good user interface actually looks like

In short, its ease of use creates nightmares for those of us who have to come in behind some knows-just-enough-to-be-dangerous hack who now has gotten 18 months of the shop's time and attendance information locked up in some screwy schema which will only be straightened out with a complete rewrite. Preferably in a proper web application.

Gah!

Re:Without MS Access-like functionality...

[ThousandStars]

The parent post is exactly right about Access. Part of the reason I bought an OS X machine is becaues of the availability of VPC, which in turn allows me to install Windows, which in turn allows me to use Access. Any OS without Access is a deal-breaker because of the business in which I work. Without it, I simply can't get the work done. Some of the Access databases I use were created ten years ago.

The best part of Access is how easy it is use. Until Linux has a database program that combines Access' ease and power, I think most small businesses will continue to use Windows/Office, or sometimes OS X/Filemaker.

Re:Without MS Access-like functionality...

[fyngyrz]

Fine. All of that will be true to some degree in many situations. None of it affects the fact that small businesses now expect to be offered easy database handling with an office suite. If they don't get it, they will, quite reasonably, decline to use that suite.

You may not like having to come after sloppy work, or trivial work, or even broken work. However, the bottom line isn't what you like. It is what the small business owner/operator likes.

I would also point out that making/keeping things difficult offers abhsolutely no insurance that things will be created beautifully, well documented, well interfaced, cleverly network aware, or with contention addressed. And so forth. I rather think the opposite is true. And, if Access has bugs, file contention problems and so on, there is no reason that a clonelike Access endeavor under Linux would have to bring over those problems.

Finally - if something is created that needs fixing, you have (or someone has) a job opportunity, don't you? Right now, very little is being created, because nothing can be created easily. Surely that's a far worse situation.

Re:Without MS Access-like functionality...

[Anonymous Coward]

It's easy to get something going with Access, but then you have people setting up databases who have ZERO concept ...

Damn straight! I'm a programmer at a small company of oil engineers who frequently "try their hand" at creating "applications" in Access and Excel. It's copy-and-past all the way to the finish line. Who needs the notion of abstraction when your ever-ready friends Ctl-C and Ctl-V can do all the hard work for you? Ctl-C is like the left leg, Ctl-V is like the right leg. Apart they are useless. Together they make a deadly team.

The stuff they churn out is usually visually replendant, but in functional terms are complete pieces of crap.

Honestly, I feel like punching someone.

Re:Without MS Access-like functionality...

[mikefe]

Yes, FileMaker.

My company used to be a Mac shop when I first came in (most of the desktops were PPC601 based PowerMac 7200s).

We've since switched to windows, but FileMaker has followed us there because of the company president's preferences who keeps tight control over the databases. One thing I'll give to filemaker is that it has great RAD capabilities, which overcome the core DB limitations in our case. Finally in version 7 it is getting multiple table support in one file (a sorely needed feature).

FileMaker/Access is one of the applications that has no replacement on Linux that makes a windows terminal server a requirement for a Linux migration here.

Re:Without MS Access-like functionality...

[empty]

I was never able to get ami working, but I found nabi and that works similar to Windows IME and is quite easy to use for Korean.

Re:Without MS Access-like functionality...

[fyngyrz]

Getting it now; thanks for the tip.

Re:Without MS Access-like functionality...

[fyngyrz]

I got nabi to work with some applications (eg, "gedit" and in fact I can write in gedit and then copy->paste into Open Office and this is so much more efficient than inserting characters it actually feels good. :)

However, it would be nice to be able to enter hangul and hanja directly into OO; any ideas?

Thanks again for the tip - I'm way further along than I was yesterday. Tehdanhi kamsa hamnida!

Re:Without MS Access-like functionality...

[dbIII]

knows-just-enough-to-be-dangerous hack who now has gotten 18 months of the shop's time and attendance information locked up in some screwy schema

How about most of the client contact information for a 200 person company in a single persons MS Outlook? It "just worked" so long as you were sitting at that PC but it was a disaster waiting to happen. Just as well I got it out of that *.pst file (which had no backups of course) before anything nasty happened.

With web front ends to most databases now it shouldn't matter what desktop you have, and with the right setup WAP or iMode can also work.

Re:Without MS Access-like functionality...

[fyngyrz]

Got it going. Had to create an ".i18n" file in home and tell xwindows about the input method. Works fine now. Awesome.

Re:Without MS Access-like functionality...

[Lord Omlette]

This month's DDJ had an article on Firebird, a DBMS that works on both Windows & Linux... maybe that'll help?

Re:Without MS Access-like functionality...

[mikefe]

Are you kidding me?

Businesses want a front end where your average tech dummy (think sales person or PHB) can change a form, move fields around to whatever order they want, and able to set the margins for printing on dead trees.

Firebird, MySQL, PostgreSQL are all back end database systems. They don't provide the front end that the end user is going to be starting on their desktop for pretty menus.

Re:Without MS Access-like functionality...

[Lord Omlette]

D'oh, I misread the article. ^^;; The "ease of use" comes from its integration w/ Borland Delphi...

Re:The cult of UNIX must die!

[couch_warrior]

Uhhh, have we ever seen a little tool called "Search" under Windows - you can search based on substring names, date of modification, and preselect the directories to search in, WHne the results come up, you can highlight some or all of the files, then cut and paste them into an arbitrary directory. This stuf has been around for almost 10 years (since Win 95). You need to get out more...

Sigh...

[mikers]

Linux and mainstream, linux and desktop, linux and corporate desktop... not ready? Of course.

Who are the people who like and use linux? Hackers, tinkerers, new adopters (some), those actually interested in programming and configuration, those who enjoy the challenge of technology, hobbists...

Who are mainstream users? People who are relatively new to computers and who don't want to fiddle, but just want to use it for email, the web, printing stuff, school work, Word, Excel and sometimes powerpoint. Users who would be happy with a maintenance free, dumbed down interface.

The gulf between is the 'adoption' problem. Just as some people will never fix their own cars, let alone maintain them, so will some people never care to tinker with linux. Not until linux is as easy to use and install as say OSX or Windows XP.

MS has been working hard the last few years to make Windows as user friendly as possible for the average person (their success can be debated but...) XP didn't really fix a lot of problems with services, or even security... BUt it sure added a lot of crap, handholding (paperclips, little dogs, need I go on?) and eye candy.

As a tinkerer, XP broke my heart.. I started feeling like there were big training wheels attached to the side of my computer. This might be comforting for the average user, but not for me. It was no longer fun, it was frustrating and irritating. I of course love linux, and would like to tell everyone I know about it, but it isn't for everyone. Not everyone likes to tinker.

Either linux dumb down and perhaps lose the community who grew it (they will move on to whatever is the next big tinkering thing of course) or it stay in the margins as the alternative for alienated serious tinkerers, and those looking for something rougher than the slick veneer of the northwest US.

I would almost rather linux stay in the margins and stay a well kept secret. I don't want to see it dumbed down into useless eye candy, hand holding and stupid paperclips you can't get rid of.

This whole article and discussion smacks of post-modernism/post-colonialism and the marginalized wanting to be in the center. I say there is no need to want to be the big dog in this case: Linux works better without the mainstream involved. SCO would not be interested, and Microsoft wouldn't want to kill it (or care about it) -- how can this be bad? Does every joe, grandma and PHB really need linux?

OK,

[Run4yourlives]

now all you have to do is explain how to do that to a small business that just wants to share a printer...

Freedom, free and free.

[khasim]

Tech know nothing PHBs know something you don't: if it's going to cost 2000 man hours of work at a $30 an hour average to redesign internal systems, templates, and procedures to work on a non-Microsoft system, that more than wipes out the cost of licensing the desktop systems.

That is the "migration cost". The high cost of migration is usually due to "vendor lock-in".

In other words, it is in Microsoft's best interest to make their systems as proprietary as possible because this will make any migration extremely expensive.

Even if your TCO is lower with the new system, you still have to pay the one-time migration cost.

That doesn't include the cost of the lead up in which you have to test, deploy, and integrate all of your servers and desktops, plus the lost productivity from people needing to be retrained or retraining themselves on the shortcuts to use Linux.

Actually, the end user won't have much to un-learn/re-learn. There won't be much lost productivity and it will be quickly re-captured by having a more stable system. The people I've seen trying to learn WinXP after Win2K have more problems than with Knoppix (until XP was made to look like Win2K).

It's not that simple when you have internal customer service apps to migrate.

And that is the single biggest expense of any migration like this. There are far too many undocumented (and often un-known) apps containing "critical" business processes/knowledge that have to work 100% after a migration. That just isn't possible on a short schedule migration.

It's very easy to handle if you have a migration strategy and will be handling this over the course of 3-5 years.

But most of the articles written do not take that approach. They look for whatever flaws they can find in the whole system and use those as an excuse to not migrate.

If your point-of-view is "find the problems so justify killing the migration", that's easy to do.

If your point-of-view is "we're going to migrate 100% of our systems in the next 5 years and we'll need to work to overcome any obstacles", then you'll finish your migration.

From the original article:

We followed change after change in requirements for approximately nine months and the customer refused to pay us anything. They said they were helping us make a better product. Additionally, the customer demanded that we meet the changes or lose the business. Try servicing a zillion dollar investment bank with all your developers while you attempt to live off revenue.

This is a customer looking for a reason to kill that migration. The requirements keep changing, even after they are met.

The first step is knowing the political agendas of the individuals involved. Politics will always win. It's not as simple as meeting technological requirements and a price point.

Easier (technical) solution

[Run4yourlives]

Talk to Apple about porting osx to the pc platform under an open source project...

and who says consumers are always losers?

Re:Easier (technical) solution

[Spencerian]

That's a thought offered by many, but not likely to happen, at least for the main OS X product. The OS X core is an open source project.

Apple gets its profits from hardware. It just uses its software to leverage the purchase of its Macs (Mac OS X) and iPods (iTunes Music Store).

For Apple to license Mac OS X would mean its eventual demise as it would not make much money. Not even Microsoft relies on a simple software sell model, diversifying into peripheral hardware and things like the XBox.

That doesn't mean someone couldn't take Darwin and meld that with Linux components...

100% identical? Good luck.

[mgkimsal2]

I just ordered 100 lattitude D800 laptops. every one of them will be 100% identical

Wow - we ordered 2 *on the same day* and they both arrived the same day from the same location. They have different wireless chips inside. One person has wireless under linux, one doesn't.

Here's hoping all *100* of yours are 100% identical down to the internals.

NIS+NFS vs AD? Not a chance

[buchanmilne]

NIS+NFS more secure than Kerberised everything and well-secured LDAP implementation using signed and encryped CIFS?

Sorry, there's no way.

Now, maybe it's more resistant to spyware and virii, but it's not more secure.

Just run:
$ ypcat passwd|jack
to find out how insecure you are!

If you were running Kerberos, OpenLDAP and NFSv4 .... maybe, if your setup was good and your ACLs on your OpenLDAP server were reasonable, it might be more secure.

billing

[Anonymous Coward]

Sales team shoud have been more focussed on billing, engineers on development.

from a sales perspective

a. Bill for out of box solution
b. quote for customisation work
c. assume people will try screw u
d. smile
e. don't do work unless you can bill for it
f. bend over
g. make sure you have some hold over client / lockin
h. smile

keep it seperate, agree a deal *then* do the work. Your paying sales to put the deal together.

Re:Just keep using Windows

[Anonymous Coward]

Interoperability with Exchange 5.5 and Exchange 2000. Am I completely crazy, or can't Ximian Connector & Evolution already do this?

Not well they can't. Go look up all the Exchange/Outlook features you lose by using that turd.

Windows Terminal Server clients using RDP out of the box for home grown applications and special Windows applications. Again, excuse my ignorance, but ... what's wrong with VNC? Why not switch to an open solution?

You're right, you are ignorant. Its not so easy to "just switch" to anything. Not to mention you can't lock VNC down enough that users only have access to a single program. Not to mention VNC protocol is unsecure and easily intercepted. Not to mention I might actually want to use my local LPT ports, COM ports, etc. through my session.

And if you think its a better idea to have a bunch of different hardware config tools instead of a common one, well then I don't know what to tell ya...

Re:WiFi - Debian - OT

[Anonymous Coward]

Also, if you do decide to reinstall, save your kernel deb, /etc directory, and wifi drivers and it shouldn't be much work to get things going again.

HP offers linux preloaded on a laptop

[Turtle Master]

Acidus writes "I called around today to the big OEMs (Gateway, Dell, HP, IBM) ... While no one offered Linux preloaded on laptops...

Dunno who you talked to dude, but HP has been selling a linux-preloaded laptop (the nx5000 for quite a while now. It's gotten quite a few nice reviews as well:

http://www.msnbc.msn.com/id/5831949/
http://reviews-zdnet.com.com/HP_Compaq_Business_No tebook_nx5000___Linux_configurable_/4505-3121_16-3 0816347.html

Aw heck, just do a google search yourself...

Re:HP offers linux preloaded on a laptop

[santiag0]

sorry, must've posted just before me..., didn't see your post.

and, yes, the reviews have been good.

I'm a PHB

[Anonymous Coward]

I am a PHB, and I would like to run linux. However, it's my techies that don't. Only half of them enjoy running linux and the other half think everything that spews out of Microsoft is golden.

I would love to save on the licensing costs of Microsoft. I think Microsoft is a blood sucking company that creates virus-ware (software that requires other software made by Microsoft to work effectively) that I hate having to proliferate accross the company I work for (1200 workstations). Many on Slashdot like to blame a lack of adoption of linux on management, but that's not the case here.

So, how do I convince the other half that Linux should be in our future? Firing them is not an option. No Linux migration would be practical without their complete support and "decreeing" that we will move to Linux will be a failure if my main tech folks don't support the move.

Re:I'm a PHB

[jmulvey]

Assign the half that don't like Linux to a "Special Projects" team. Give them jobs like sorting power cables by color.

Re:I'm a PHB

[mpe]

I am a PHB, and I would like to run linux. However, it's my techies that don't. Only half of them enjoy running linux and the other half think everything that spews out of Microsoft is golden.

Obvious Windows Evangelicals are not just found in management. Possibly it's more common within management because there arn't that many managers who are cynical enough when it comes to software salesmen.

So, how do I convince the other half that Linux should be in our future? Firing them is not an option. No Linux migration would be practical without their complete support and "decreeing" that we will move to Linux will be a failure if my main tech folks don't support the move.

Are there any other differences between these two groups of people?

Re:I'm a PHB

[Anonymous Coward]

Not really. They are both very intelligent groups. I think those that would support Linux are a little more practical, but that might just be my bias showing through.

The only other thing I can think of is that the windows supporters generally do not have any college/university education, while the Linux guys do.

Re:Just keep using Windows

[sewagemaster]

4. Font compatibility with Microsoft Office and Openoffice.org and/or StarOffice.

Crossover

I use Crossover on my linux desktop for this purpose. But it still isn't perfect - especially with visio. You just can't export .eps properly. I need an environment for me to do diagrams for my papers/reports/thesis. Unfortunately, neither dia and kivio are even close to what visio has under native windows.

samba cupsd hotplug

[Anonymous Coward]

Peer-to-peer networking, functional printing, laptop support, single sign-on to Active Directory and a better Device Manager (with a driver-get mechanism) are among the things companies are asking for.

sorry I use gentoo, this is pretty simple:
# emerge samba
# emerge cups
# emerge hotplug

"While no one offered Linux preloaded on laptops,"

[santiag0]

HP? How much research went into this article? Keerist, this was all over the web including slashdot.

Here is HP's nx5000 notebook preloaded with SuSE 9.1:
nx5000

Re:Single sign-on to what ? - Mods this is OT

[ecliptik]

This is a completely OT question for Walrusss so please don't mod me down, thanks

Walrusss,

I was looking for some tux stickers last nite online and I found your site on google, but it is giving an error:

Bad Gateway

The proxy server received an invalid response from an upstream server.
Apache/2.0.52 (Fedora) Server at tuxstickers.ptaff.ca Port 80

I'd like to see what stickers you have and purchase some if they are too my liking. Please either reply or email me at the address on my website, thanks.

Re:Just keep using Windows

[SillyNickName4me]

> I have no problems using TTF fonts should I want to.

I think there is a small legal issue with ttf bytecode that amkes that hintign doesn't really work unless you specifically compile it in (and get the appropriate license or want to run the risk of patent infringement and its possible consequences)

Specifically, this concerns the following patents registered by Apple

US05155805
US05159668
US05325479

Network Neighbourhood for Linux

[smartfart]

Fact: Linux just doesn't have a Net Neighbourhood/Places GUI.

Actually, KDE's Konqueror (using the lisa daemon) works wonderfully with windows, and also with any *nix servers it finds on the LAN, via FISH. Boatloads of other stuff work too.

Re:Network Neighbourhood for Linux

[Tenareth]

P2P networking in large networks is a horrible hack that needs to really get killed, it is a worm superhighway. If a PC could only talk to the server networks, and not to each other you would almost instantly reduce the possibility of worms wiping out your network in a matter of minutes.

Tiny sites it's fine, but people don't really understand that they just opened up their C: drive (or /) to the entire world because they just clicked a few buttons, and now Joe Schmoe can see their files.

Re:Network Neighbourhood for Linux

[Anonymous Coward]

How the fuck does the workstation know if \\SPIKE is a "server" or another desktop?

If you really want this, you can implement it on a network level without crippling the clients.

Re:Freedom, free and free.

[the_mad_poster]

No. Your first step, as with any business decision, is to justify the cost of the process. If you can't justify the one time cost vs. the ongoing costs, you don't do it. Businesses aren't generally interested in throwing money down the toilet in the interest of their IT department's idealogical bents, so if the cost justification doesn't exist, it doesn't happen. If the biggest problem you face in your quest for change is management not wanting to save money/increase productivity, your company has much deeper problems than their software platform. Blaming non-migrations on "stupid PHBs" is disingenuous. If they think that doing it will save money or increase productivity, they're not going to stand there and say "hmmm... something I could take at least some credit for - nope, I don't I'll further my career today".

As far as your comparison between Windows and Knoppix, your anecdotal evidence is irrelevant. OpenOffice does not function like Microsoft Office, like it or lump it. Nor does Evolution function like Outlook. Each of these also lack features of Office that some users will have difficulty getting over. On top of that, a GUI'ed Linux system is about as stable as a tower of Jell-O. The X Window System should've been scrapped and rebuilt as a real windowing system more than a decade ago. Amusingly enough, when Windows Explorer takes a shit, it respawns itself. When Gnome or KDE go belly up, they don't. They either throw a kernel panic (and good luck training Betsy the Bimbo Secretary the concept of Magic SysRq) or crash to a prompt. Linux is only stable when you don't add all the external cruft, and nobody wants to type memos in ed and pipe them to sendmail.

Novell Linux

[OnlineAlias]

I agree, the Novell Linux Desktop is a mess too. In order to get it to work in a Novell environment, one must modify the eDirectory schema, be on a specific version of Groupwise, use NFS to get to Novell shares, change login script methodology, etc etc. Novell really BLEW it. I wanted it to at least support Novell products and installed base! Instead it is Suse with a bunch of N's all over it....

Re:Freedom, free and free.

[mpe]

Businesses aren't generally interested in throwing money down the toilet in the interest of their IT department's idealogical bents

They aren't interested throwing money "down the toilet" to support Microsofts idealogical bents either. By Microsoft has enough spare cash to persuade many people to do so and get the suckers to pay for it.

Re:Just keep using Windows

[carrierbagman]

The Vncserver on linux defaults to multi-user, multi-session rather than displaying the console. I don't know if you can get the windows version to behave this way though.

pick two

[cliffyqs]

You may only have two:
Fast, cheap (in $), good quality.

in more words:
Fast + Cheap won't be good;
Good + cheap won't be fast;
Fast + good won't be cheap.

Visio support???

[ughhgu6]

Where's my Visio support? Same thing for OS X! As a network designer/implementer I absolutely require Visio support, and converting to 2003 compliant XML files doesn't count.

money

[hkht]

fantastically easy reliable, efficient business apps must be developed for linux. maybe it should not be an open-source effort where the apps are created for the love of it with no guarantees. people need to be fully employed and directed and fired if necessary in efforts to create top quality tools of business and computing. with that said i hope i'm not the first person lynched by open-source programmers. funny thing after making comments about quality software all i can in see in my mind is the blue screen of iptraf clicking off the packages coming in and out. who am i kidding! who cares if anyone adopts linux, it's not our loss it's their's!

Dunno about the big corps as a whole...

[carlmenezes]

...but I bet the HR depts would love it :)

Be like Windows?

[HangingChad]

Why? One of the things I like about Linux is that it's NOT like Windows. At home I have Windows and Linux boxes on the same network and they see each other just fine. I can configure a printer in less time than it takes in Windows. Not seeing a problem.

I realize if Linux did this and that the PHB's might pop out of their prairie dog hole and look at it, but if they don't what do I f'ing care? The PHB's will come around in their own time, or not. If they're too short-sighted to see the advantages, tough luck.

But if I'm setting up a small to medium size business these days, there won't be a Windows machine or piece of MSFT crapware anywhere near it.

I'm just sick and f'ing tired of people coming along and telling the Linux community what they have to do to gain commercial acceptance. Like my brother. Going on about Linux just isn't there yet. Linux isn't the problem, he is.

Re:Be like Windows?

[Anonymous Coward]

Sounds like your brother has some insight into how the real world works, whereas you have none.

Its fools like you (who just want to close their eyes and pretend all the problems with Linux don't exist) that hurt the Linux community the most.

Telling customers that its "My way or the highway" while spending their money to do it doesn't cut it in the real world junior.

Hardware Failure

[copponex]

Replacing an x86 motherboard: $70
Replacing an x86 processor: $100-200
Competition in the x86 component market: priceless

Replacing a logic board: $200-500
Replacing a slow as shit G4 processor: $200-500
Having shiny buttons: goddamn expensive

Linux is where the two shall meet. Open platform hardware running open source software. In a few years, for FREE, I'm sure at least one distro will have it down. Fedore is damn close already. And at the price of a CD-R or DVD-R, I'm sure there will be a lot of takers.

And look out for Apple to start kludging up their OS again once some people with bad ideas get stuck in their hierarchy. Pretty soon it will be obvious that no large business can compete with the meritocratic programming method.

Let's just hope Apple and Microsoft get themselves a DRM system. Their market share will drop by 30% within two years if people have to actually pay for their products.

It's more than money.

[khasim]

No. Your first step, as with any business decision, is to justify the cost of the process. If you can't justify the one time cost vs. the ongoing costs, you don't do it.

You've never been in a company where the new VP pushes a migration from a working to system to the "new" stuff. I have. I've seen working NetWare/GroupWise systems ripped out and replaced with Win2K/Exchange systems, even though more servers were required and more money was spent on the software than was spent in the previous 5 years on those systems.

Politics is a major factor and the numbers can be managed to show any results that you want.

Businesses aren't generally interested in throwing money down the toilet in the interest of their IT department's idealogical bents, so if the cost justification doesn't exist, it doesn't happen.

Yet in case after case, that exact situation has happened. Again, the numbers can be managed to show whatever someone wants them to show.

Being the new CIO or VP and doing nothing except maintaining the status quo is not going to look good on your resume. That's where the politics come into play. If you aren't already on the most popular system, lots of "problems" will be "found" that can only be "fixed" by migrating to the popular system. If you're on the most popular system, then most managers will not risk their career by championing a migration to a less popular system. Instead, they'll focus on centralizing that which is decentralized and decentralizing that which is centralized.

Blaming non-migrations on "stupid PHBs" is disingenuous.

And I did not do that.

If they think that doing it will save money or increase productivity, they're not going to stand there and say "hmmm... something I could take at least some credit for - nope, I don't I'll further my career today".

Incorrect. The actual thought process is more "hmmm... something that might save money, but might fail and cause me to lose my job - nope, I'm not risking my career".

OpenOffice does not function like Microsoft Office, like it or lump it.

Yes it does. I can sit someone down and they can type and print from OpenOffice the same as from MSOffice.

Each of these also lack features of Office that some users will have difficulty getting over.

Only if the users at the company in question are part of the "some users" group that you mentioned. If they aren't, then there won't be problems.

On top of that, a GUI'ed Linux system is about as stable as a tower of Jell-O.

Whatever. Lots of people use it and is seems to work for them. I'll leave out the rest of your ill-informed rant.

In business, it's about politics. That's the fact. The sooner you learn that, the sooner you'll be able to move beyond tech.

What a load of tosh.

[jotaeleemeese]

- Training is a continous process. People keep being trained no matter what their desktop is, because applications keep changing (otherwise people would be happily stuck with the oldest version of a given software). So if you are investing in training anyway, it does not stop you to do so in Linux based applications.

-Deployment: please make my day, tell me that it did not cost you anything to migrate from W98 to W2K or WXP. Well, that cost would be in the same ball park if you do a Linux migration (knowing what you are doing of course). The big difference would be that you may have more power in your hands to decide when to migrate once you move to Linux. If your Linux provider begins to pester you with upgrading it very well would be cheaper to move to another Linux provider (since you are keeping your data and applications properly segragated and in standard formats, which are much easier to port in Linux thatn in Windows).

-Support: dude, it takes 2 to 3 weeks to train a competent system administrator to RHCT or even RHCE (or any other accolade you may want to obtain for your SAs). If your Windows guys can't grasp a different computer system then I would bring into question the health of your IT support anyway. A professional SA can easily and confidently translate the skills learned in one platform to another. And since your company surely is providing regular, timely training anyway, you replace the WXP update with a RHCE and you are set. If you really need to you may hire one or 2 consultants to baby sit a migration, but once people are up to speed, what would be the problem?

Licensing costs are not as important, but neither are the points you tried to drag out into the open.

All boils down to a very simple matter: do you want to leave your IT infrastructure in the hands of a single company (which is not a trustworthy busines partner)?

I don't, if you do that is your problem, many people are paying far too much money for the privilege of receiving little or nothing in exchange.

Re:What a load of tosh.

[eV_x]

- Training is a continous process. People keep being trained no matter what their desktop is, because applications keep changing (otherwise people would be happily stuck with the oldest version of a given software). So if you are investing in training anyway, it does not stop you to do so in Linux based applications.

99% of the workforce is already substantially trained on Microsoft products. You're ignoring that - many people have Microsoft Office on their resumes and training on these systems is almost *always* bypassed unless needed.

-Deployment: please make my day, tell me that it did not cost you anything to migrate from W98 to W2K or WXP. Well, that cost would be in the same ball park if you do a Linux migration (knowing what you are doing of course). The big difference would be that you may have more power in your hands to decide when to migrate once you move to Linux. If your Linux provider begins to pester you with upgrading it very well would be cheaper to move to another Linux provider (since you are keeping your data and applications properly segragated and in standard formats, which are much easier to port in Linux thatn in Windows).

Most companies skip every other microsoft release and only then do they upgrade matching versions of Office. This is well known and documented and shows the enormous cost associated with migrating. Saying it would be in the same ballpark of cost ignores the cost of resources and the availability - in some areas it's not an issue, in others it's not. Don't BS me and act like that's not a fact.

-Support: dude, it takes 2 to 3 weeks to train a competent system administrator to RHCT or even RHCE (or any other accolade you may want to obtain for your SAs). If your Windows guys can't grasp a different computer system then I would bring into question the health of your IT support anyway. A professional SA can easily and confidently translate the skills learned in one platform to another. And since your company surely is providing regular, timely training anyway, you replace the WXP update with a RHCE and you are set. If you really need to you may hire one or 2 consultants to baby sit a migration, but once people are up to speed, what would be the problem?

Again, you ignore available resources, cost of training, and already existing skillsets. Yes, people can learn, but do you not comprehend that it takes time and money to train someone for 3 weeks? You seem to have no grasp that 3 weeks on 50 resources is not a minor thing.

Licensing costs are not as important, but neither are the points you tried to drag out into the open.

You have not "blown away" anything I've said. Instead, you've continued to persist the mentality that migrating 10,000 desktops is as simple as waving your hand. That's BS and you're either working for a small to medium sized business or have yet to ever be part of a big desktop upgrade/software deployment.

All boils down to a very simple matter: do you want to leave your IT infrastructure in the hands of a single company (which is not a trustworthy busines partner)?

I'm not arguing for using Microsoft, but this doesn't make people buy Linux either (and by "buy" I mean put their money where the resources are). That's a pretty tired argument to sell to a CIO when over the past few years IT spending was substantially cut due to the economy across the board. More IT projects were put on hold or cancelled and are just now starting to kick back in. This is the real world, not "yeah but we're not evil". Whatever.

And why do you need Windows media for that?

[jotaeleemeese]

You are locking yourself in when there are wonderful open formats out there.

The lock in mentality is so entrenched that peopleforget what is the objective (training videos, conferences, etc as you said) and defend the too impossed on them by the manufacturer as an end on itself.

You have to hanlde it to MS, their biggest success has been to make people forget what the priorities are equiating legitimate busineess needs with MS solutions whihc are expensive and force a company into sclerosis since they stop looking for the best alternatives.

Bullshit.

[jotaeleemeese]

After having to deal with underperforming Access "databases" by dudes that thought they knew better, my company (and another one I worked for) barred the damn thing from desktops and put DB development where it belongs: with dedicated teams.

These teams, knowing their stuff, would not touch Access with a 10 metre pole.

As for small companies, they are carving their own obsolescence: I used to porvide support for dentists. While the Access solutions they had normally gave uncountable headhaches, Linus or UNIX solutions kept working silently, the trusty powerhorses that ensured the dentist could do his work and not wait in frustration thanks to the latest virus or BSOD.

Re:Bullshit.

[fyngyrz]

If your company can afford a "dedicated team", your company by definition isn't part of the audience. Small businesses need small, easy solutions. You aren't talking about small solutions.

For the record, we've been using Access for many years here at my first company and the databases are still working fine. I wrote them; I do actually know what I'm doing to some degree (I write custom PostgreSQL and MySQL applications under Linux, in fact) and perhaps that has a little something to do with it. I have personal experience here, and I can assure you that your blanket condemnation of Access is flat out wrong.

Re:Just keep using Windows

[Spoing]

1. 1. 3. Interoperability with Exchange 5.5 and Exchange 2000.

   http://www.novell.com/products/connector/ [novell.com]

Exchange 5.5 and earlier was not supported by Connector last year. Has that changed?

Linux on Laptops

[Anonymous Coward]

Linux on Laptops -- Pick your brand...

Re:Just keep using Windows

[Spoing]

1. got it ? it may not be important to you, but some big companies have _decades_ of data stored in their systems, some of this data only accessible through aged proprietary apps written in clipper, cobol, VB 3.0, whatever (some of those only exists in binary form. sources are long gone)... heck, once i went to a stock brokerage office and they had an access 2.0 running under OS/2 (by M$ recomendation) because access 2.0 was the only thing their PBX supported, and they had by force of law to record every phone call, internal or external.

To add to your Clipper data story, that's exactly the reason to use open data types if not open source itself; you can get at the data in 5...10...50 years. I have word processing documents from college that I can no longer access. Using a propriatory, single vendor, closed format, application to store data or business logic is the core problem.

It's conservative to move away from these traps and pick a better way of storing business rules and data. Open source, formats, and protocols provided by different groups (commercial and/or non-commercial) are the best way to do that.

Windows will always be easier but not cheaper

[Anonymous Coward]

Tom's point that Linux may be more hassle than it's worth will certainly discourage some techs from migrating but if they're discouraged that easily they wouldn't make the migration to any new platform no matter how good. I recently returned from a discussion with a fortune 1000 company that was looking at migrating to Linux and the number one driver for them is cost. Annual MS licensing fees costs them a fortune and most of their users use only three applications. They have a fairly homogeneous hardware environment and Linux runs well on most of their hardware.

Tom's other points like Windows peer to peer networking and ADS sign-on are absolutely critical in heterogeneous environments with Windows and Linux systems and vendors like Xandros that have made this functionality as easy as in Windows (Suse is making progress in this area as well.

The biggest impediment right now is finding large scale deployments where Linux makes sense. You need users who only have a few application requirements and those applications have to be available on Linux. Lack of applications is the number 1 impediment to adoption of Linux on the Desktop.

Another issue that comes up is how to manage those desktops and again, the same two players (Xandros and Suse) have decent solutions. Suse has Red Carpet (rebranded "Zenworks for Linux" I think) and Xandros recently announced xDMS which deploys and manages large numbers of desktops.

MOD PARENT DOWN

[alizard]

It took me 3 working days to get my Lexmark Z605 printer running under Linux.

It took me a full month to find Linux backup software that can actually be installed and made to work by a non l33t uberh4xx0r, and all I was trying to do was clone a drive, make incremental backups, and archive to DVD-R. I wound up having to rewrite a rsync script I found to do the incremental backup. Interesting experience, since I haven't had to do script stuff since I had a DOS desktop.

I got the how to back up Linux workstations article done 30 minutes before the deadline. The reason was that it took DAR several hours to make a DVD-burnable backup and I wasn't going to explain how to use it until after I knew it worked, since I'd installed other packages and couldn't get them working.

To claim that the problems are solved is total bullshit. This is stuff that I shouldn't have been able to sell articles about because THIS STUFF SHOULD HAVE BEEN WORKING OUT OF THE BOX (FC2, in my case)

I shouldn't need to run a Windows emulation (Win4Lin) to get my work done, but I do, and I'm finding myself having to install MORE Windows software because dia suxx0rs (I'm using Visual Thought) and it looks like my search for Linux project management software as good even as the original MacProject running on the MacPlus of 10+ years ago has been in vain.

ALL THE PROBLEMS ARE SOLVABLE, BUT THEY WILL NOT BE SOLVED AS LONG AS PEOPLE DENY THE PROBLEMS ARE REAL.

Fix the problems and we can push MS into the tarpits. Telling us that anyone who doesn't know that Linux desktops are ready for prime time is technically clueless only serves the interests of Microsoft.

Your Milage May Vary

[Rysc]

I work as a computer technician for the local school system. My big complaint in the still largely Windows 98 world drivers are a pain. You throw an image on a machine and... oops! Wrong image. The drivers aren't right! Do I spend the time to grab the right drivers, or spend the time to reimage with the right image? Unless the image I had happened to have the right NIC drivers I just have to reimage.

With WinXP it's slightly better, but still a pain in a lot of cases,

You know what I do when a user says "My sound isn't working!" and it's because the driver isn't installed correctly? I pop in my liveCD, today it was SimplyMEPIS, and it detects the hardware. I can then go to KDEs device manager or (more likely) just lspci to find the make/model of the card. I can then go and grab the right windows driver, reboot and install it.

Figuring out which driver to use for a sound card which as listed as (say) "Intel AC '97" is nearly impossble. It has ALWAYS been easier to just let Linux figure it out.

Linux live CDs are my "It works everywhere!" item. BartsPE doesn't do it for me since half the time it wont detect the NIC and do DHCP. With my liveCD I can easily say "Yes, the card is bad," or "You need the driver for Weird Chip 81627, not Weird Chip 81630" and I can do it quickly.

I don't like Windows. The driver support just isn't there.

Active Directory

[Anonymous Coward]

Whether or not it's a windows solution doesn't always matter to big companies trying to adopt linux. However there is a solution available (although not free): VAS

History

[Sun Rider]

It was the same before, most corporations bought only IBM, or at most CDC, Burroughs or Univac mainframes. Then a few individuals and small companies started using pcs. It will be the same with Linux, Mozilla, Openoffice, etc. Small companies, some foreign countries, and a few individuals in big corporations will start using it. Gradual build up until sudden landslide.

Getting a desktop linux out there

[solune]

If you take a look at the small business crowd you'll see people more willing to try, and stick with, Linux. We're tinkers, and risk takers, individualists that have a unique vision. That unique vision, however large and small, is the exact same type of vision that caused Linus to develop the Linux kernel in the first place.

A few of the things Linux programmers can do to capture this market are pretty simple:

1)Templates! I'm not talkin' just the minimal crap I get with OO.o, I mean a decent sample of polished templates like what I can get with Word Perfect, or other proprietary wares (or warez,for the l33ts).

2)Decent, simple tutorial maintenance programs: One of the best programs I ever used was one by Clear and Simple software for OS/2 (I forget the name). Besides getting you through many onerous tasks that everyone should do, it went a long way in teaching where configuration files lie, what most of the lines meant, and (Best of all) how to tweak it.

Many small business owners aren't gonna shell out for new hardware for everything they need to do, so knowing how to maintain and tweak is pretty important.

What it boils down to is this: if you want to see more linux desktops, it has to be able to run a business out of the box, complete w/ web design, word proc, accounting with the ability to output industry standard files. (a lot of DIY business ppl design their own literature [or would like to] only to be thwarted by finished product in proprietary formats that have to be translated @ extra fee.)

Finally, what all software developers are seeming to miss, is we want accountability. As a contractor, there are certain vulnerabilities in my work that absolutely would not be tolarated, legally or professionally. Conversely, it seems software comes with terms that amount to "Use it, but if you get screwed it's not our fault." Damn, I wish I could say that!

Rather than develop for the "Enterprise market," which would only help corporations gain power through control of the software --creative-- market, why not usurp their power by building social networks *around* them? The Net routes around damage, right?

I guess what I'm trying to say is: "Are there any home software contractors out there?" For this, Linux is ideally suited. This metaphor should also, imho, help validate the value of open source software. Imagine how much I would have to charge YOU for a new bathroom if I had to license the technologies for "fastening wood segments with cylindrical ferrous alloyed metals." Oh, and it might mitigate some of that outsourcing damage too.

Mandrake and SuSE play tag for device support

[leonbrooks]

Each supports stuff better than the other, both of them support stuff better OOTB than the article implies. Both of them support stuff OOTB (like network-based document scanning including automated remote scanner detection) which MS-Windows simply cannot do. Can't speak to Red Hat and the others, but Fedora can't be too far behind (desktop land, remember?) and Debian-land have Xandros and the like for slick installs.

Practically all of them, even the single-CD distributions, offer a wealth of applications completely unparalleled in the MS-Windows world by even the most generous OEM. Start totting up the cost of extras to do serious office work, graphics work, virus scanning, PDF creation and display, email service, database service, file service (and pile on the "seats") and the like, and suddenly MS-Windows looks awfully expensive in comparison. Add in the cost of having to find, install and maintain all of these from disparate sources vs a single regularly updated RPM/DEB/TGZ repository and the comparable TCO/ROI figures the consultants are bandying about start looking kind of... tilted.

Re:Mandrake and SuSE play tag for device support

[jdray]

Well, right. That's been the big attraction of Linux the whole time. The nut, though, is that the base operating system (including window manager) just isn't as well polished as Windows, nor as well integrated with the apps. This fact isn't lost on distro companies, and they're working furiously to resolve the problem, but they're not there yet. Simple things like a consistent look and feel for control panels, consistent button placement or an easy-to-navigate file system (okay, that last isn't so simple, and isn't something Windows has either) are the pieces that create roadblocks to widespread adoption.

Take all that and add in that software in general, and OpenOffice in particular, takes a long time to launch, and you have a system that's rough around the edges and feels slow whether or not it actually is. You can cry all you want about how Microsoft has deep hooks into their operating system that allows "bloated" applications like Word (mind you, I use a lot of Word's features regularly, so I don't consider it bloated) to launch quickly, but the fact of the matter is, that's the sort of integration that makes their solution (stem to stern) so popular. You would think that, with all the vaunted power of the open source movement, and the example set by Microsoft's success, that similar and better functionality could be created in the free world.

But it's just not there yet. I will continue to hold my breath.

True story

[leonbrooks]

Is it simple enough for a PHB to understand and use?

I recently had to deal with a PHB who edited up a document on how badly his OpenOffice-on-Linux office system was working.

He wrote it up on OpenOffice on Linux, completely forgetting that this was not MS-Word running on MS-Windows. It All Just Worked. Any more irony and you could pick it up with a magnet. He was very embarrassed when he woke up to what he'd just done (he couldn't find the MS-Outlook icon to send what he'd written).

It seems fairly clear from this and other similar experiences that often the core issues have nothing to do with performance or capabilities and everything to do with perceptions.

As well as making our software better, we need to make our software seem better. The software company with the desktop lock is big on promotion. Perhaps stuff like the NYT Firefox ad are a better idea than many SlashDotters seem to think. I don't think competing head-to-head is going to be viable, but certainly some more professional promotion of FOSS as a poster-child and not an also-ran might-do-the-job afterthought is a Good Thing.

I take care to promote FireFox, ThunderBird and OpenOffice to my MS-Windows-addicted clients a being better for practically everything than the corresponding MS product. This gets them used to the idea of FOSS being higher quality and safer, which also prepares them for bigger steps later.

You used the wrong tool for the job

[leonbrooks]

The GIMP is not about collating and printing photo albums, and neither is PhotoShop. They can both do it, to be sure, but you should be using something like KAlbum or FLPhoto instead. Both of these and a few others sit in Multimedia, Graphics on my Mandrake 10.1 desktop.

Single sign-on vs. centralized authentication

[Etyenne]

I am not sure exactly what the guy meant by "single sign-on to Active Directory". People often confuse centralized authentication with single sign-on. Centralized authentication mean the authentication data and process is managed centrally, by a "domain controller", for example. Single sign-on mean you provide your credentials only once (at login time, most likely) and don't need to provide them again to access a set of services (files, email, etc). The principal benefit of the former is to ease management of credentials (ie change your password in a single place) while the benefit of the later is to let you provide your credentials only once per session (instead of once to log on your workstation, once to access your email, another time to access the intranet, etc).

Centralized authentication from Linux to Active Directory is available here and now, and work flawlessly. It's all possible with the magic of nsswitch and PAM. At most, you will need to install MS Service for Unix or AD4Unix on your ADC to extend the AD LDAP schema so it include attribute required for Unix account information (login shell, home directory, etc). It take about 5 minutes to setup on Linux. I do it all the time, it's boilerplate stuff. The password expiration stuff have nothing to do with the client, it's all a function of the server (password expire, authentication fail).

No Linux distribution do single sign-on to Active Directory (Kerberos, actually) out-of-the-box AFAIK. Technically, it is entirely possible though. You would just need to Kerberize all the applications your desktop use and the service you provide. On the desktop applications front, it is not granted that everything support Kerberos application at this point, that would be worth checking. On the service front, I know all the daemons that use the Cyrus SASL library (OpenLDAP, Cyrus imapd, etc) are by definition Kerberized, and so is Samba. The biggy would be to figure out how to Kerberize Web application, although I seem to recall Mozilla have recently gotten some level of Kerberos support.

That being said, it is worth noting that there is no such thing as a single sign-on under Windows even. I guess "AD-aware" network service can authenticate via Kerberos, but if you must provide your password more than once a day, you are not doing single sign-on.

In short : centralized password management against AD is here and fully functionnal now, single sign-on is within the grasp of a determined organisation.

Re:MOD PARENT DOWN

[MmmDee]

I agree with poster. I RTFA and while I disagreed with a few minor points, the emphasis is right-on. Too many Linux advocates are trying to make the round peg fit the square hole. I suspect they seriously underestimate how computer illiterate the business/home world really is. I'll bet that 50% of the computer-using public can't tell you correctly what a "cursor" is. In the business desktop environment, as well as in the home, the OS has to work "out of the box". Give MS their grief, but with Windows XP and even going back to 2000 and to some minor extent Windows 98, their software automatically identified hardware and tried to setup the appropriate drivers, and it came with wizards to set up basic file/print sharing. MS software (written by them or 3-rd party) generally had no dependencies. Binaries that work for Windows 95 most often than not, also work on Windows XP. Most DOS programs also continue to work.

Single sign on needs to be made easier

[Anonymous Coward]

I can set up a mail server, web server, file server, dns server, ntp server, and most other servers easily enough to get things done. What I've found extremely difficult, and haven't solved yet, is to be able to sign on with the same identity on a network. LDAP, or whatever other authentication/identity/whatever methods are out there need to be made easier for small business.

Right now, the only way around this for the small businesses I know that are dipping their toes in the Linux waters, is for a main server to run everything, and the clients to login to the main server through forwarding X. They are basically doing the same thing as they used to do with Netware. With Netware, they popped a floppy in (actually, never removed it, the boot floppies have been in the floppy drives for over a decade without being removed), and booted off the network. So they are familiar and comfortable with doing this.

So setups with installing Linux to each client hard drive, and then booting up, hitting the up arrow for bash history to the correct ssh or telnet command for forwarding X, and they are logged into the server. But this isn't really the way to do it properly. For a network where more than one server exists, a single identity is needed, and this is where LDAP or OpenLDAP comes in. I haven't seen Suse since 8.2, but in 8.0 and 7.3, NIS was used. Hopefully SUSE has a frontend for LDAP as well, but in Debian, LDAP ain't easy to put it lightly.

I'm not familiar with Active Directory, and neither are the small businesses I'm familiar with, since they are all using Netware and older versions of Windows, and plan to keep using them until the hardware dies and they run out of parts from the cannabilized computers in the back rooms. But I'm assuming that Active Directory does what LDAP does.

Is there an easy front end to OpenLDAP for Debian?

As far as Linux on the desktop, LDAP seems to be the hurdle that needs to be lowered from a beginner's standpoint. Once that hurdle is tackled, then application servers would be the next jump. And one major application, or setup to give a lot more attention to would be running something like OpenMosix or LTSP type of applications where old hardware could be put back to use. For small businesses, putting old hardware back into service (or actually taking the old hardware which is still running an old version of windows or dos or netware and putting it to use on a Linux cluster would be more accurate) would be something many small businesses would be interested in, and outside the US this would be appreciated as well, so the need for more efforts in this area exist.

Windows Networking and Linux

[greggman]

I plug 2 to 10 windows boxes on a net and they network with each other. I put a linux box (FC2) on it and doesn't.

Assuming NO DNS SERVER and NO WINS SERVER, this is a home setup. Can someone point me to how to get Samba to work and work well? What 12 obscure files and options to I have to edit on the FC2 linux box to get it to share correctly. So far I've had no luck.

All I want to do is give the FC2 box a name like I do in Windows but setting the name (which appears to be burried in network->eth0->settings doesn't make it start working.

Even after reading though all hours of samba docks and getting it to basically share there are often delays of 5 to 20 seconds where there are no delays between windows boxes.

Can anyone point me to a faq, document or otherwise that covers HOME windows networking and linux/samba/fc2?

Re:Just keep using Windows

[C0vardeAn0nim0]

read my post again. that's the whole of it.

problem is: you can't move away from proprietary formats over a weekend. sometimes it requires replacing expensive hardware that's still doing it's job fine.

in the mean time, linux' better support the proprietary crap (or at least a decent ammount of it) or it's chances in the corparate desktop will be smaller, way smaller, than it should.

and on the open vs. closed data types: my impression is that corporations actually _DO_ like closed data types. or at least open data types that alows them to keep the actual data inside secret. most of them base their business models in the assumption that any leaked information may come back to bite their market share.

i believe that's the argument IBM, Unisys and other supliers used during the late 60's, 70's and early 80's. "use our . since we have full controll of it and it's kept under all kinds of secrets, we can ensure your data will be safe from unauthorized eyes, and since we're we're not going away anytime soon, so you'll allways have support for ".

managers are trained to analize options and make the decision that will:

- bring more functionality;
- be more secure;
- have continued support;
- give the greatest return for every $ invested;
- be easiest to use;
- cost less;

not neccessary in that order. sometimes they make compromises or mistakes in their analisis, but choosing a proprietary data type over an open one is (in their point of view) not neccessarily one.

Re:Just keep using Windows

[Spoing]

1. in the mean time, linux' better support the proprietary crap (or at least a decent ammount of it) or it's chances in the corparate desktop will be smaller, way smaller, than it should.

Fully agree. In addition to Linux, though, any replacement system must also support the previous proprietary crap.

One machine shop I helped out -- we're talking 50+ ton computerized hydrolic presses -- was not only stuck using DOS for the CAD system they chose years before, they had to use specific video cards.

That means no DOS boxes.

No hardware emulation.

There are no programs on any platform that support the data files.

When the hardware fails and can no longer be upgraded, they will have to redo all of the templates that the business is based on.

That's my concern...and I see similar things all the time on my larger corporate sites. MS Access/VB is a poor choice (though not as serious) for the same reasons.

I think you'll agree that it's not enough that Linux applications chase propriatory formats. The propriatory formats are the problem.

Re:Just keep using Windows

[dbIII]

now that linux is starting to mature as a desktop environment

It's worth mentioning here that in most cases of comparing MS to linux, MS is the new kid on the block. Linux can run a lot of *nix applications that date back before MSDOS, plus being a *nix clone there are features that have developed over time. Linux had a decent framework and had learned from years of mistakes from day one. Go use NT3.51 and then try coming back and telling me that MS windows has always been a rock solid multiuser system.

That said, a lot of office work could really be done on a playstation with a keyboard and web browser. Most of the time it just needs to be a typewriter - or if you need professionally typeset documents an old Mac does it better.

Re:The cult of UNIX must die!

[dbIII]

Uhhh, have we ever seen a little tool called "Search" under Windows

You've missed the point somewhat and don't know what is possible. Thankfully "grep", "awk" and "sed" are also ported to windows, which makes shell scripting usable far beyond the limitations of the search tool. To go beyond that, there is also cygwin.