Alternative To Windows Desktops

Eric_Z writes "Ace's Hardware has got a article called "The Mad Hatter meets the MSCE" by Paul Murphy, about the TCO benefits of using UNIX(Lintel) instead of Wintel. According to the piece: 'The subject of this article looks at alternatives to the Windows desktop, which is a hot topic these days with IBM/SuSe scoring a highly public win in Munich with desktop Linux, and Sun aiming to build on StarOffice being the leading alternative to Microsoft Office with a software stack code-named Mad Hatter which Sun also plans to use extensively in-house. But companies depending on Microsoft Certified Engineers to adapt to Linux will carry over a number of problems, significantly increasing the chance of project failure. Paul considers the alternatives, the migration problems, and in seeking a more reliable alternative takes the opportunity to look at the business desktop from an entirely different angle, and propose a more radical solution.'"

Comments

Good good

[mrtroy]

Microsoft needs more competition!

The best way to counteract a fat monopoly like those microsoft whores is to put some good ol competition out there against them. Its tough to match those budgets and large scale operations, but more and more companies are fighting them from more and more directions...it can only lead to good things --- better products being produced by everyone.

Either that or more marketing.

Corporate directory services

[sphealey]

[...]With the exception of security management, essentially all of the practical skills associated with those functions will be invalidated. DHCP, WINS, SMB networking, Processor Affinity Management, Domain Administration, Registry hacking, and so on, are all technologies and ideas out of place in a well run Unix environment, though some pollution has crept in.

[...]can be, and therefore will be, perpetuated in the new environment despite having no natural role there.

I am not a fan of Active Directory. But if the author thinks that corporate directory services (preferabley Novell eDirectory, but Active Directory if you must) have no role in large-scale corporate networking, I have to question the rest of his conclusions a bit.

sPh

Desktop Corporate Linux... I tried

[oZZoZZ]

My company runs Windows 98 clients and NT4 server atm, and I figured it was time to upgrade. I looked into Microsoft, with Office/WinXP and server 2003, and the cost was about $40k. That seemed insane, so I decided to try Linux.

I've been running Linux at home now for a few years, and am quite competitent running it. My first step was to replace the slackware/wmaker combination that I was happy with on my laptop to Redhat/Gnome/Bluecurve, and I was immidetely impressed with how far linux has come on the desktop, I figured this wouldn't be a problem.

I showed the owners of my company Linux, and they said they were fine with it on every machine... now the tricky part, application compatability.

Under Wine I was able to get my payroll software and estimating software running, but the accounting software proved impossible. Using older style database clients and VBA, I was totally unable to get it working.

I came to the conclusion that while I can use Linux on the desktop, application support from large corporate vendors need to be there before Linux can run on the desktop. I also came up with: "in 3 years, if we want to run a different accounting/estimating/etc package, will linux work for us?".. That question is unanswered atm, and therefore using Linux in a corporate enviroment seems to be a gamble right now, a gamble that I am not willing to wager on for my company. Another issue is support from our existing vendors... they supported running their software on Windows and 2 of them *REQUIRED* PCAnywhere to be available whenever needed... this was not possible with Linux.
Linux on the home desktop seems more than ready, but enterprise/corporate enviroments seem to need better application support before it's possible... while I do belive that the application support will be there in 3 years, I don't think it's a risk work taking atm.

Re:Desktop Corporate Linux... I tried

[ip_vjl]

I don't understand this attitude that MSOffice is *required* if you run Windows. I see it a lot when comparing the costs of running the two OSes.

$ windows = OS cost + MSOffice cost
$ linux = free OS + free office app

There's nothing preventing you from running free (beer/speech) software on Windows.

If you need Windows to run legacy apps, why not do it in stages. In your case, upgrade your boxes from 98 to XP - but don't do the MSOffice route. Use Openoffice.org (assuming it will work for you since you were going to do a full linux switch anyway) and other open source software when applicable. (Mozilla Firebird instead of IE, etc.)

This way, you don't abandon your legacy apps ... and in a few years (at next upgrade time) there will either be a feasible open source solution, or maybe Wine will have advanced enough to run what you need.

If you can do a full transition, good for you. But to compare costs the way you did isn't a real comparison.

Re:Desktop Corporate Linux... I tried

[FreeLinux]

I disagree. I would suggest that the application support is there today but, not the way people are trying to accomplish it. By and large most evaluations like you have done are trying to use Linux as a drop in replacement for Windows. This will probably never happen. While there are many applications that can be used as drop in replacements to Windows applications there are even more that are not. And Windows applications, for the most part, don't run on Linux.

Bit, how is this different than the likes of Windows 2003. There are countless applications, even Microsoft applications such as Exchange 2000, that will not run on Windows 2003. For some people this will mean that they will not implement Windows 2003 but, as time wears on most if not all will move to Windows 2003 and upgrade or replace their existing applications to ones that do run on Windows 2003. They will buy Windows 2003 and they will also buy Exchange 2003.

So, rather than looking for a seamless drop in replacement to Windows in Linux, why not look at it from an upgrade/migration point of view? There are numerous accounting applications that do run natively on Linux. The specialty apps that are written in VB will need to be rewritten for Linux. But why not? Chances are that those same VB apps are right now being examined for a rewrite in C#.NET. They'll have to be for the sake of Windows 2003.

The point is that people seem unwilling to rewrite or migrate their apps for a Linux environment but, for some reason, they think nothing of doing this for their Windows environment. The thing that they fail to take into account is that in the Linux environment this will almost certainly be a one time affair. But, in the Windows environment it will be a recurring theme every few years because that is what Microsoft wants and has to do in order to keep selling the same companies more software.

All too often people say that it is not cost effective or it is too difficult to make the switch but they seem to disregard these same issues as they run on Microsoft's treadmill.

BTW, have you repatched your Microsoft RCP service?

Re:Desktop Corporate Linux... I tried

[Zathrus]

Good suggestions, but not always doable in practice.

Open data standards for payroll and accounting data? I'm sure there are some... they're probably old as dirt and about as fun to utilize nowadays too (yes, I've just spent a couple weeks learning the horrors of X.12 in the shipping industry -- it's used all over but it's archaic, has over 3 decades of different revisions, and an utter PITA to actually use). You can roll your own format (we did... we're in a position to) and make it reasonably open (again, we did... at least to our customers), but the odds of getting someone else to write to your format is low, especially for things like payroll/accounting. You could also reverse engineer their data files (a coworker did so for a flat file database at a former company, producing a real time importer for Sybase/Oracle), but that takes some pretty serious skill and money.

Don't think that it's just MS producing "proprietary" data. Virtually everyone does. And it's not the big, obvious formats that are a problem -- those have enough people looking at them to crack the nut eventually -- it's the small, uncommon formats that will keep you locked in. And it's equally unlikely that you'll easily find replacements that are low cost and open format. Companies have an incentive to lock you in... the counterbalancing force to this is that in a competitive market place they also have incentive to read other people's formats, which will either lead to a common format or to everyone figuring out how to import everyone else's data.

In general, without government mandates, it tends toward the latter rather than the former.

It's the apps, silly

[sphealey]

The author missing one minor point. The core of business information management for small and medium sized business is Win32-based client/server applications. These are the products that you see advertised and discussed in Manufacturing Systems and CFO Magazine. In the middle to late 1980s they were available on several platforms and usually had a Mac version, but by the mid 1990s they had migrated almost exclusively to the client/server model on the Win32 platform.

These midrange apps are the bread-and-butter of corporate computing. They do not run on the Mac and do not run under Linux. Some are starting to move toward a web browser based model, but not all and not necessarily quickly.

Until Linux equivalents exist for these midrange apps, the Linux desktop will not be used in midsized organizations.

sPh

Troll

[poptones]

Better be careful, you're gonna get modded down as a troll.

I just stuck a fresh install of RH9 on a laptop. It installed amazingly well - in fact, it installed better OOTB than win2k.

But "better" lasted only until it came time to actually do stuff with it. Sure, samba seems to work well and it has no problem browsing shares on MS boxes. But try to play a video file... oops, no media codec installed in the RH9 default distro. Hmmm... well, try to play an MP3 then. Ooops, no can do - cannot play an MP3 file from a file in a samba share. Try copying the file to this machine and perhaps it can be played then...

I really want linux to live up to the promise. Really. And I'm looking forward to working with the new media structure in gonome, and hoping to do my part. But I'm honestly beginning to wonder if linux can ever catch up - much less take the lead - on a user friendly desktop.

Re:Troll

[Richard W.M. Jones]

Please. Red Hat chooses to leave certain feature [patented video codecs] out of their desktop, and it reflects on Linux desktops in general?

Red Hat would, I'm sure, love to include these in their distribution. Were it not for the fact that larger companies have made sure it's illegal for them to do so.

Rich.

Microsoft will be their worst enemy

[Bob-o-Matic!]

By making it more and more difficult for users to run unlicensed copies of Windows OS (XP was a great start, they'll do better next time for certain), the home user who wants to upgrade will find themselves "upgrading" to something else entirely if they want to keep the price the same. No one wants to pay for a software "dongle" to make other software they (may) have paid for work. People buy computers to surf the web, send email, play games. They don't feel they need to pay just to be able to move files around.

I am hoping that the kind folks at OpenAL and OpenGL make a compelling replacement for DirectX so that games will run natively on Linux. When you get the gamers, you will have won. MS has the gamers right now. When those gamers come to Linux, they'll learn the OS and show their friends. Windows will lose its ubiquity on the desktop because no one wants to pay to upgrade their copy of windows, or even pay for an original license when building a machine.

It is only a matter of time.

Re:A few problems with this

[ichimunki]

1) Most office workers barely know how to use the software they have. The transition will require training them to not know how to use a whole different set of software. Oh, wait, it won't because no one needs to be trained to not know stuff.

2) Which is?

3) Benchmarks are where?

4) Not nearly enough said. Again, benchmarks are where? And why are they "Linux" FSs in #3, but now we're talking Solaris? Which is it?

5) People had to learn to use both Word and Excel as they migrated from packages like WordPerfect and Lotus 1-2-3. Not to mention the changes from version to version of just the MS software. I think your users will survive.

Re:Whats new?

[Liselle]

What we need to do is be developing newer, fresher ideas which keep microsoft on their toes

This isn't really all that true. You can't just dump a UI/functionality change on the average user and expect them to embrace it. Many have trouble with the interface that we've had since Win95.

These people have the right idea. Ease folks into it. Otherwise you will have a response similar to what happened with WinXP, where the interface was made more intuitive and easier, but casual computer users still complained (and rightly so, I think) because the things that took them so long to learn got turned topsy-turvy.

The author is in a happy dreamworld

[reemul]

Several of the comments made in the article seem to indicate that the author is living in a happy dream world, where clever users are oppressed into mere drones by MCSE's and MS software. He acknowledges that it is a best practice in the Wintel world to lock down machines as much as possible to minimize support costs, yet seems to think that Unix will "empower users" (from a sidebar) without causing any problems at all.

Is he crazy? The reasons that machines are locked down is that the endusers are stupid. They know nothing about computers, and ideally they shouldn't have to - they are just tools to do their real jobs. Any extra capabilities will just allow them to break more things. Sun can only support so many users per admin by locking systems tighter than most MSCEs could dream of - the answers to what is wrong are so easy because there are no other options. The users aren't empowered, they are chained down as much as possible. All to the good; but believing you can take the same idiot endusers from a windows shop, give them magic Lintel boxen and some responsibility and rights to manage their own systems, and get *fewer* support calls is just delusional.

And thinking that it's the OS that is driving all those fast upgrades to physical machines is also absurd. A huge portion of all business desktop and laptop upgrades is driven by vanity, not need. Good luck thinking that a rational OS decision based on security and TCO will quickly stop "mine's bigger" purchasing. You think execs sending email, looking at excel spreadsheets, and playing solitaire need those multi-thousand dollar laptops? You think that running linux they'll stop buying them?

I liked the approach of the author, to look at the practices that will be reflexive to existing support staff and the effect they will have on a Linux implementation. But his take on the reflexive approaches of the *users* is completely unrealistic, and renders his article mostly useless. Face it, most of the people here on Slashdot have dealt with those endusers - you think the majority will agree that they will miraculously become wise if just given a chance? Or will the /. crew decide that the author is living a dream?

Re:Whats new?

[Liselle]

Market research and focus groups led Microsoft to implement those lovely menus that auto-hide. Nifty idea in theory, however in practice the "play around with it" aspect of users using a program was lost, because they never saw stuff that they didn't use regularly. Sometimes users don't know what's good for them. Focus groups are not the answer to all of life's problems, unfortunately.

Re:But I like Windows desktops

[chill]

Because, in larger business environments, they suck.

The argument presented in the paper is more of a thin-client vs client-server/desktop approach.

With the software properly installed and managed on a central server, not individually on each PC, there are significantly less problems.

Whole industries have been built around the Windows PC that aren't necessary from a corporate standpoint. I speak of client-side firewalls, anti-virus and disk imaging software.

No need to "push" an image when the PC gets corrupt. No need to reboot the PC. No need to run and license individual anti-virus applications. No need to scan for spyware, etc on each PC.

"PC Empowerment" is a BS phrase. The only thing most PC's empowered the coporate user to do was send worms, catch viruses and play games. Applications like a word processor, spreadsheet, presentation program, CAD, project management, e-mail and other business software can just as easily be run via a central server. Administration is tons easier.

And with full-duplex, fast ethernet to the clients and gigbit or bonded channels to the servers, load and run times can often be faster than off of cheap PCs with hard drives.

I think the Mac is the only answer for now...

[The Lynxpro]

Speaking as someone who works for a very large government institution, I think the only way to get off the Microsoft train is to go with Macs. And I'm not being a troll when I say this. #1, the computer and the OS come from the same supplier. This is where Linux fails because none of the reputable hardware companies will offer real support for the OS if you run into troubles; nor will they indemnify the institution from the frivolous claims by the likes of SCO. If you buy the HP line, what are you going to have to do, install Mandrake on your own? Won't you still be paying the Microsoft tax unless you buy the PCs from Mitec, a mom-and-pop whitebox store, or purchase a Dell line with that DOS'ish OS on a bundled disc? Or, if you want support, you have to pay extra to Red Hat, IBM, or Sun? #2 the Microsoft apps won't run natively on Linux. You have to run Wine or Codeweavers software, and I'm sure if a government agency does that, Microsoft will be on the phone with the various elected officials to start investigations on software purchases as well as EULA violations (and a BSA audit wouldn't be too far down the road too). I've been thinking about all of this because we run Win2K on Dell P3 800mhz machines, and its time to start upgrading. But each of these concerns is enough to kill any suggestion for switching to Linux, especially when everyone who has a hand in deciding IT issues has MCSE certification to justify their jobs. Whereas if an agency becomes a Mac OS X shop, you have the Microsoft Office apps, but the hardware upfront costs more. Granted, you can shave off 1/3rd of your IT staff if you go Mac, but the political party that would be most interested in saving government monies in such a manner (through layoffs and eliminating redundancies) would probably not be inclined to help Apple out since Jobs and others are left-of-center in their political affiliations, not to mention Al Gore is on their board of directors even if it is in a ceremonial position... And the taxpayer suffers, not to mention us employees that have to use this *poodoo*...

It doesn't matter! MSOffice == Windows

[gosand]

If you need Windows to run legacy apps, why not do it in stages. In your case, upgrade your boxes from 98 to XP - but don't do the MSOffice route. Use Openoffice.org (assuming it will work for you since you were going to do a full linux switch anyway) and other open source software when applicable. (Mozilla Firebird instead of IE, etc.)

To many corporate people, it would be just as hard to migrate the Office software as it would the OS. MSOffice is so ingrained in the corporate culture it is pathetic. I have to send my status report to my manager in a Word doc. Everything is stored in freaking Word docs around here. Want to show some people some pictures? Put them all in a Word doc, that way you can email one huge .doc file. I once complained to a guy because he was attaching screenshots to a bug report like this. I explained "do you realize that for someone to see these, they would have to use MSWord. They are just images". His response? "Everyone here has Word installed, that isn't a problem."

As for the others, you won't see IE go away as long as MS is the OS. Hell, our internal website won't work with Opera, the browser I use. I am actually surprised that my boss lets me run it. Gotta conform and everything.

Our department gets its MSWord licenses from Corporate, so it doesn't cost our department anything. That is what the managers are most concerned with, their budgets. As long as it doesn't cost them anything out of their budget, who cares? If we all have to upgrade to OfficeXP (which we are doing) from Office2K, then Corporate will take care of it.

It doesn't matter how compatable it is, if it looks like Office, acts like Office, is better than Office, or is 100% free. If it ain't MSOffice, a lot of places won't use it. Companies sign deals for their OS/Office licenses, so many times you can't split up the OS/Office software. Oh, and you have to upgrade every 3 or 4 years.

So while I appreciate your idea, in companies where MS has them by the short hairs, it doesn't fly. It is also one of those things that makes me yearn for a better economy, so I can quit this cubicle wasteland and go work for a small company again. The "corporate atmosphere" is slowly killing me. It is killing everyone else too, they just don't realize it.

Kee-rist, sounds like somebody has a case of the Mondays.

Oh, that again...

[Nurgled]

This is quicky becoming the "All your base" of 2003.

Re:MCSE?

[boaworm]

"A Microsoft Certified Systems Engineer is to computing what a McDonalds Certified Food Expert is to fine cuisine"

My 2 cents...