Slashdot Mirror
Linux Making Inroads, But Not At Windows' Expense
zaphod123 writes "According to this article, the stories about Amazon (and others) switching to Linux have been misrepresented. The Linux install has replaced a proprietary Unix system, not a Microsoft Windows product. This is still "A Good Thing" for Linux, but not the downfall of Microsoft that some have foreseen."
>When asked whether the company would ever consider replacing its
>Windows machines with Linux, Busch said absolutely not, noting
>the lack of "robust office packages" on that platform.
I often think that this excuse really is more like "we can't get naive users to use it without being crippled". Linux distros need to test their software on non-Unix people more. Humans. Typical office people who, if you ask them if they have a Mac or a Windows box, say, "Yeah, I think so".
>And Busch threw another wrench into any mass Linux migration by
>noting that the overall cost of Linux and Windows 2000 is almost
>identical after you factor in support and maintenance.
in other words, after you get done with the hassles of Linux, and the hassles of Win2k, the hassles of Linux are a little bit more. time=money, so the cost of that extra hassle is the same as the cost of Windows & its apps.
So much for free-as-in-beer.
This hassle is invisible to the Linux developers cuz they know how to fix or work around glitches when they arise. So it seems "easy to use" for them.
Try it on grandma. then report back.
Marketing-driven companies end up over-marketing their products. Engineering-driven companies end up over-engineering
it still is good news in that they decided not to change to a m$ based solution. they went for linux.
-1: flamebait should really be -1: inciteful
Solaris, their operating system, has few advantages over Linux, nowadays. Frankly, without adding the GNU tools, Solaris is virtually unusable! (And, who's gonna pay $10k for their compiler when GCC does the job?)
Sun is about to hit a brick wall. Unless they change direction dramatically, Linux is going to gobble them up, just as SGI consumed Cray. Cray was meaningful for a long time, until the capabilities of "Minis" (as Supercomputer folk like to refer to UNIX machines) silently approched the power of super computers at a fraction of the cost.
The same is happening with Linux-Sun. For a small fraction of the cost, Linux on commodity hardware (Intel) is approching the power of Sun's products. It's inevitable, without some sigificant change.
As MS still holds a great deal of market share in server installs, this IS a blow to MS, as they failed to sell Amazon on their own product!
Besides that fact, it's still a VERY good thing for Linux, as Amazon is a HUGE online retail operation that serves as a model for many other businesses. That's how Linux is becoming successful - word of mouth and trial by fire. Linux proves itself in a very fast and competitive market, and more people jump on. Of course *NIX and BSD systems will be the first to be replaced, because the people who maintain them aren't as afraid to make the jump to Linux (they're already somewhat familiar with it). Give it time, though, and you'll see quite a few former MS boxen turning over to linux.
I mean, honestly, two years ago, did you ever think linux would have about 24% of the server market? No! So of course it seems impossible that it might steal an even bigger share - and thus there will always be those who doubt that it will ever happen. But slowly, it WILL happen. It's already happening.
I happen to know of one organization that has decided to convert almost all their desktop systems to Gnome+StarOffice, wherever possible. I think the plan is to have one or two Win* boxen, to act as conversion stations when having to send electronic documents to the outside world, but the overall plan is to dump Windows because of licensing cost issues.
Regardless of what that article says, the costs are very real and companies are definitely considering it. Perhaps one or two cases may have been misinterpreted, but by and large the case for converting to Linux has not been mispresented.
--jordan
Because of its robustness, modularity and stability, Linux is highly able to replace Solaris, HP-UX and AIX type licensed OS's in the enterprise. The people who buy these systems buy them to get the best technical solution to their problems and consider cost of ownership, which is high in any OS choice given the task, secondarily.
Trying to get Linux to beat Windows on the desktop is fighting yesterday's battle. Want to kill Microsoft? Sap it's growth, which is in server OS's and embedded systems (XBox, Pocket PC, etc.)
The amount of energy spent by the development community in trying to be the next Microsoft is astounding, but very few vocal developers seem to even focus on what Microsoft is trying to become.
To borrow a phrase from the Old West, "Cut 'em off at the pass" and focus on making an OS that runs devices better than Windows ever will, an OS that runs DB2 and Oracle better than any other and an OS that can be extended and integrated with server side applications at compile time with more ease.
If you take away Microsoft's revenue growth, you take away their stock price. Take away their stock price and you take away their monopoly.
Technology Marketing is what happens when people turn their hard work over to people paid to manipulate others.
We had come to the opinion that IT/IS departments that had gotten used to UNIX systems feel more comfortable about moving to Linux than IT/IS departments that had gotten comfortable with Windows. There still seems to be a strong feeling of uncertainty when it comes to planning for migration headaches (which are inevitable).
It's still awfully hard to penetrate into markets where the people involved are only aware of doing things a certain way. I can recall having a job in college where I became responsible for a file server running a quite old version of NetWare. I wasn't thrilled about it and the company that sold the box to my employer wasn't around anymore to support it. But it ran and I prayed that the box wouldn't conk out, because I feared having to convince my boss to migrate to another OS.
My sigs always suck.
I think I trust winformant to tell me about Linux about as much as I trust slashdot to tell me about Windows... :-)
Sure, this isn't a case of Linux replacing Windows, but it is a case of Free replacing Proprietary, and that's just as important, if not moreso. Microsoft's Ministry of FUD has been working overtime trying to scare people away from Free Software solutions, using "arguements" that are little more than "Free Software Is Communism!".
Free Software / Linux advocates should be glad that: 1) the best a multi-billion dollar corporation can do is mimic some of the very unoriginal trolls around here; and 2) companies are not being trolled.
Cause I coulda sworn I mentioned this before.
;-)
Where's my credit??
Good quote, too many chars. Seriously, the slashdot 120 char limit sucks!
Um, nope. They SAVED money because they already had the *nix expertise inhouse for running their web servers, and they already had their entire app written to run on *nix boxes. Switching from Windows to *nix is a 100% change in platform. That's VERY VERY expensive to do, which is why you'll find almost nobody doing it.
Linux (for now) competes on low end (mostly Intel) hardware. The biggest player on the low-end by far is Microsoft, so that's who's most affected. Users who switch from proprietary Unix to Linux do so because they see a cost benefit from switching to low-end hardware. If Linux weren't there then they would be forced to go to MS.
It is true that Linux has clobbered the main lown-end Unix: SCO. Good riddence :-).
One thing that does surprise me is that Windows is still so popular on basic file & print servers. These machines don't run any special software, so they should be simple to replace with Linux boxen. We just got a Cobalt cube in our office and it's really neat. Setup is fast (like 3 minutes!) and painless, and it does everything you need from a small office server. Why aren't these things more popular?
It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail. - Abraham Maslow
2005?! Like, in 3 years, right? This is said as if it's bad. Linux overtaking NT in the server market by 2005 sounds like one of the first realistic goals I've heard for the OS community.
At least it's much more realistic than the standard"Tonight? Tonight we take over the world" refrain.
Guvegrra?
> In the point of view of MS, anyone who's not using a MS product is concidered a failure.
Yep. The biggest problem with MSFT for the past few years is that MS has saturated the desktop and stalled out in its grab for server space. MS needs growth to keep those inflated share prices up.
Also, even though Linux is "competing" mostly with Sun these days, every time an organization adopts Linux for any reason Linux becomes more visible and more credible.
Those who wish to view the history of the universe as a struggle between Linux and Microsoft can still see this as Linux moving to consolidate its base by rallying the rebels and independents throughout {the galaxy, Middle Earth} before going over to the offensive against the strongholds of the {Empire, Dark Lord}.
Sheesh, evil *and* a jerk. -- Jade
I know you all read the article, but did anyone read the web address? http://www.wininformant.com. It's part of the Windows 2000 Magazine Network. Their motto is Windows news and information. Does anyone here see any potential bias when the website says that Windows will rule for the foreseeable future?
One fellow used to cast things in terms of products for the mass market, and products for the elite class of users. Sure, there's going to be a huge market for the 'computers grandma can use', just like 'billions & billions served' - but there's also going to be a small but vocal and powerful minority of very experienced users who just don't want a computer with the training wheels bolted on and whizzards to hold your hand thru all common tasts. In the democracy of 'market choice' it will become increasingly important to ensure that the rights of the minority users who know what they want and already know how to do it don't get trampled on.
Yes, I do it the difficult way because it's more educational and I want to know what's going on and be in control. Notice how every time your super-automatic wiz-bang box craps out *I* have to come over and fix it or figure it out for you??
try { do() || do_not(); } catch (JediException err) { yoda(err); }
Of course - most anyone skilled enough at general PC operation enough to use Linux is going to be aware of the massive ammount of software that assumes you are running in a Windows environment. Therefore, they are going to have some copy of Windows for the sake of convenience to be able to use that software if the need arises.
Note though that this only means that Linux owners are going to have SOME copy of Windows. Not necissarily the latest.
A. Training is VERY expensive.
B. Screw application. What about the CODE?? If Amazon was written in ASP with heavy use of COM objects, you'd have to do a multi-million dollar re-write to make it into a CGI/C application. Administrators are by-and-large button pushers. It's the application that's expensive, not the administration.
An interesting question is how the other *nix versions affect Linux. Is it better for "the community" to have several *nix variants out there competing against MS, or better to have just one, be it Linux or some other variant? Put aside your religiousish tendancies for your favorite OS, and lets discuss a wider benefit.
Having more than one version available gives more options to people and allows for several niche distros of *nix. It also presents several targets for MS instead of allowing them to focus their sights on one "problem".
With a single *nix front, we would be able to address concerns across more installs, and consolidate the knowledge from more sources to improve the overall product.
I'm not sure which way is best, and more than likely a hybrid will be the end result, and for the better. What's the feeling here about all of this?
THIS SPACE FOR RENT
> Switching from Windows to *nix is a 100% change in platform. That's VERY VERY expensive to do, which is why you'll find almost nobody doing it.
Yeah, and companies never have to retrain their staffs or rework their software when they upgrade Windows, do they?
As usual, the people you describe are pinching pennies for the short run and costing themselves dearly over the long run, by sticking with a system owned by a robber baron and changed at his whims, rather than moving to a system based on open standards.
Sheesh, evil *and* a jerk. -- Jade
Paul Thurrott, who's carer is clearly wrapped up in the success of MS products, pulls a nasty bait-and-switch in this story.
He talks about how Amazon and Intel switched some servers from $$IX to Linux, and says that the "anti-Microsoft" press has been mis-representing these moves.
Then he quotes an Intel executive saying that they haven't even considered switching their MS based systems to Linux. The implication being that NT is doing a great job in their back office. But the reason given for not making the switch is "lack of 'robust office packages'"!
So, the story, apparently, is that neither Amazon nor Intel dare run NT in the FIRST PLACE.
Or, to put my own bias on the shelf for a moment, Amazon and Intel see Linux a preferable alternative to NT/Win2k as a server platform.
How is this a win for MS?
-Peter
PS: This post was generated on a Linux desktop.
Microsoft is already a monopoly on the desktop, and all they are left with is clinging to that with challenges from all sides.
I have to admit-- Linux for community documentation, support, and features.
But-- FreeBSD is STABLE (check longest uptimes at Netcraft when you get a chance). If I could go for 4 years without rebooting with Linux... They have even dethroned Irix when it comes to stability.
So yes, they are a very practical alternative to Linux. It is really that Solaris and HP-UX are not so practical or cost effective in the small ISP market.
I actually now believe that Linux will form a shield which will allow BSD to grow into certain niche markets, such as high-availability web servers (currently MS and Sun).
LedgerSMB: Open source Accounting/ERP
Things are going pretty much the way I figured they would. Linux is making progress in the areas it shines in. If it keeps up, I see the following happening in 5-10 years:
The market will split into 3 basic genres. You'll still have the Apple/Macintosh vanguard, as those diehards won't disappear. Apple's done a good job of keeping that core audience, and they'll still have them. Microsoft will become less of a business solution and more of a home system. People still want an easy to set up system, and Microsoft gives them that. However, companies are already getting sick of MS licsensing and bugs. That leads to the major change, Linux will become the system of choice for businesses. Given 5-10 years, install and administration of Linux distros will be as simple as Microsoft's are now. Look at how far the last 5 years has brought Linux if you don't believe me. Businesses will go with the low cost implementation that Linux provides over the headaches that come with MS. Programs like StarOffice will make the transition of the business side less painful. Companies like Sun will find themselves having to shift priorities away from the OS in order to survive.
In short, Microsoft will stay a dominant player in the home PC field, with Apple being the secondary choice. However, businesses will tend to go with the cheaper and less bug prone Linux for their own installs. Of course, that's just my viewpoint on things. Your mileage may vary.
Electronic Frontier Foundation for online civil rights information
Linux will dominate the world with or without displacing existing Microsoft systems. Simply put, the potential future installed base of information systems is probably less than a percent of a percent today.
Microsoft will certainly be involved in many of the future ones, but Linux offers so many more advantages that its use will far exceed any benefits to be found in a Microsoft offering.
Won't win the desktop? Who cares! Why try to beat the McDonalds of the computing industry when there are plenty of kosher delis, sushi bars, trattorias, cafes, gyro places, hot dog stands, russian tea rooms, and so many other styles and qualities of restaurants that haven't been built yet?
That is why they don't even think of changing windows machines to Linux machines !
There is no internet browser that could be found as a decent one - and VERY far from good -, and about decent office apps only StarOffice could do the job badly if compared to Office2000/XP, etc...
Plus there is the problem of nothing is working on the desktop - end of the question! Everything is crippled, except KDE!
Everything is beta software when they release the "new/improved whatsoever" to make Linux users buy a new distro release; I speak from my own experience. They want (the distributions) to make money with the desktop, that's all.
I love and use Linux/FreeBSD only in text mode and for servers with grafical tools, NOT on day-to-day desktop/office computers. For me this is very sad, believe me on this one, because I feel Linux is superior but lacks a general strategy for the desktop, there is no master ideia, each Linux person/develloper/distribution have its own master ideia and its own standard, nobody is united by a common way of thinking about desktop usage. very sad like I said.
Like someone said above: "I am (was more on the past) a Linux desktop lover, not a windows hater", too.
The point of Linux isn't to wipe out Windows, or even to compete with it. Linux is about creating a free operating system that people will choose if it's the right tool for the job. Linux doesn't have to hurt Windows to be sucsesfull, it just has to keep improving. Just because Microsoft wasn't hurt by this doesn't mean that it isn't a victory for linux.
"Yeah, Microsoft is evil because they have taken away our choices, let's crush them so that the thing we like becomes our only choice" Why do we have to be for this camp or that camp? Why can't we all enjoy the best software available reguardless of who developes it?
So the question is this: how much effort should we devote to pushing Linux and BSD as alternatives to close-source OS's?
My answer to that question is "not much". We need to focus on our main adversary: Microsoft. John Q. Public and Buford T. Congressman are probably not going care much about which version of Un*x somebody ought to choose, but will care very much about whether to use Windows.
In short, "open source Un*x or close source Un*x?" is simply not on the political map, and doesn't need to be.
-Miko
Miko O'Sullivan
well, they switched from Unix to linux. it *could* have been Unix to windows. who lost? if you've taken any business/financial/economics classes, you know what opportunity cost means. linux might not have managed to erode the windows market (at least in these cases). but seems to me , has managed to stop windows from eroding Unixes market share further.
i mean, around the time when NT4 came out, everyone and their brother were replacing big iron (with unix) with multiple NT boxens. seems like we've managed to check that. it's only a matter of time before linux invades the NT/w2k/(whatever they're calling it this week) market.
you gotta stop their advance before you can make 'em retreat.
if i were a redmondien, i would not be happy because linux is merely replacing Unix. i would be extreamly unhappy that linux is replacing Unix. it could have been winNT/(whatever...) that was replacing unix. opportunity costs for MS. no new revenue streams. no new market shares.
gottsa love how MS and winformants can put a spin on things.
Remember that NT was originally marketed as a UNIX killer. Then Win2000 was marketed as the UNIX killer. The significance of these switches from UNIX to Linux is that Win2000 was not able to win these UNIX seats in these situations. With the ever shrinking overall growth in the computer industry, grabbing share from competitors is becoming quite important.
Yeah, but Novell is around (for the moment) and can support their products like gangbusters. Novell tech support has answers to just about anything that can happen to your machine - they have seen it all. We still run 4.11 just because it is so damn stable and you cannot replace ZENworks with anything - nothing even comes close.
I find it kind of amusing that everyone kind of assumed that Amazon switched FROM Windows to Linux, when the article never really said that. I thought that as well.
My beliefs do not require that you agree with them.
These guys just don't learn... Linux isn't about market share - it's about making a good OS! Linus and Alan Cox don't get up in the morning and think of ways to cut into Microsoft's pie, they try and improve the existing Linux system. Woz said it in his recent interview... he cut his hacker teeth building computers that competed against their previous versions, always improving.
As for Wininformant, yay well done. You caught the fact that a Linux win wasn't actually a Microsoft loss. Here's some more news for you: WE DON'T CARE.
shut up man
After all, Microsoft has been trying to take the
"enterprise" business from the unix vendors for
years. If linux replaces a traditional unix vendor,
you can be sure they at least considered, and rejected microsoft when considering Linux.
... if Linux didn't exist? I think these analyses ignore the loss of momentum that Linux has caused Windows.
Five years ago, as NT was replacing Netware in most enterprises, many predicted that Unix systems would be the next to fall under the Windows steamroller. However, in cases where simplicity and the availability of commodity hardware are more important than raw performance and scalability, people are turning to Linux to replace Unix systems, not Windows.
So while Linux may not have made major inroads in replacing existing Windows servers, it has prevented Microsoft's hegemony on the desktop to spread to the server side, and has given Unix (generically) a new lease on life.
I think that's a pretty major story.
There are no more "field" trips to a desktop. If it gets screwed up, you tell your management software to blow down a new image. That's the whole point of a corporate desktop. You don't bother remote admin because it's cheaper to send down a new image. Personal files on the desktop? Sorry, it's not yours, it's our machine and all business related files are on the network file server like our propriatary file management app makes you do. If you saved a file locally, tough, you shouldn't have been doing it.
Install their own software? They don't have the rights. If they do install their own software, their violating the AUP of the company. It's our desktop, not theirs.
When it comes down to the high TCO of a desktop, it's supporting the USERS with the APPS. That will remain whether it's NT, Linux, or Mac. Users are users.
Linux may make inroads into the corporate world (aside from small pockets of developers) when:
1. There are tools that plug into our management systems to adequately manage the desktop.
2. When we decide to stop spending millions of dollars developing our custom file management, accounting, billing, purchasing, instant messaging, telephone billing, office directory, HR, Benefits, the IE only Intranet, Remote Access, PKI, and the apps that integrate all the above, and start spending that money plus 100x more on hiring Linux coders, buying a duplicate server for each backend since they are mission critical apps we're not going to test the Linux clients hitting live servers, hire Network Admins to take care of the new test servers, hire trainers to train users how to use Linux and StarOffice, hire a slew of more technical support to handle the increased number of phone calls during the transition, hire a slew of people to handle all the document conversion issues that will inevitably come up.
Actually, it will NEVER happen because the first thing the CIO will ask is "Where's the ROI?" And when we show him the numbers, and say that converting to Linux on all the desktops will never pay off but we may break even in 10 years of not having to purchase Microsoft Office licenses for each desktop, the plan will get shitcanned. That's why we won't see Linux on the corporate desktop.
-- If god wanted me to have a sig, he'd have given me a sense of humor.
Not only do I admin and program on Solaris boxes, I'm also a GCC library maintainer. There're my qualifications.
"Frankly," you're utterly wrong. Not only is Solaris just fine and dandy, it has features for programmers which aren't anywhere near to showing up on Linux. For example:
Linux has none of these.
Severely uninformed statement, my friend. GCC doesn't generate SPARC code nearly as well as Sun's compiler. (Ask the GCC developers.) It's good but it's not there yet.
GCC cannot even generate a 64-bit binary yet. (Very close, but still some bugs.)
There are plenty of reasons to buy a SPARC, and to use Solaris, and to use Sun's software. It's all about the right tool for the right job, and Linux quite often isn't it. (I write this sitting on a Linux box.) Quit'cher karma whoring. :-)
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
I think part of the reason this "Win vs. Linux" cost of ownership battle rages on and on is because we're comparing apples to oranges.
Individual user workstations are rarely "mission critical". If they crash once in a while, productivity doesn't really diminish. (Sure, they have to spend a minute or two rebooting and logging back in, and sometimes they might lose the file they were last working on - but that's the extent of it.)
Servers, on the other hand, obviously pose much bigger productivity issues if they go down. Every user connected to one is cut off from what they were doing until it reboots.
Linux shines on servers for this reason. It's markedly more reliable than the average Windows-based server. If nothing else, it saves you from doing a lot of reboots when you reconfigure things. (Make a change to Apache or Samba configuration? Just stop and restart the daemon; not the whole machine.) Win2K and XP are better than ever about imitating that functionality, but they still ask you to "restart the machine for the changes to take effect" far too often to be convenient on a server.
On a workstation though, the rules change. The biggest factors become ease-of-use and training. Most employees come with a chunk of Windows knowledge in advance. Sure, some have no clue, but even temp. agencies requires experience with using the mouse, getting around Win '9x, and using MS Office apps. When you have hundreds or thousands of employees, it starts to look really good to use a lesser-quality operating system if it means most of your workers can already get around in it with no additional training.
This is something that only time will change (and then, only if people stick with Linux and keep making efforts to improve it over the years).
Did you read the article? Intel moved from and old application running on UNIX to a NEW APPLICATION running on Linux. There wouldn't have been any "migration" costs.
The fact is that NT and commercial UNIXes have some of the same downsides. License costs, license compliance costs, etc. Whereas all OSes have some maintenance and support costs. Surely you have to look at each situation, but the FUD that says "MS software has lower TCO in all situations" is patently false. Witness your own example, migrating from another UNIX. (BTW, NT is supposed to be a POSIX OS. Given this claim by MS it shouldn't be any harder to migrate from a commercial UNIX to NT than from a commercial UNIX to Linux.)
Anyway, how does cost explain away the author using a quote about the lack of an acceptable office suite to explain how Amazon and Intel not switching to NT for their servers isn't a bait-and-switch? And loss for MS? (Assuming that this is a zero-sum game, which it must be given the finite nature of hardware.)
The bottom line, I think, is that this guy is wrapped up in MS, and thinks that anything trumpeting Linux successes is somehow "anti-microsoft." I wouldn't have been so put off by this, but his article is basically accusing the "anti-Microsoft" press (InfoWorld!) of dishonestly pushing some anti-MS agenda, and backs up his argument with an intellectually dishonest argument.
Finally, let me say that I'm not "anti-microsoft." I am "pro" software that gives me more freedom. I am against MS illegal business practices. But I think you are trying to take the easy way out by just labeling me "anti-microsoft" rather than addressing my complaint.
-Peter
Linux is not a company, so we don't have to have a corporate strategy. Leave the strategizing to the folks at Red Hat, IBM, et. al. This is one of the biggest mistakes people make when they criticize Open Source. Open source is not a company, but there are open source companies with their own strategies.
What this means is that many different companies will have their own plans and these may include the desktop. When the desktop becomes a real option for most people, then these companies will be able to be there. Developers should continue to work on whatever products they find interesting.
This also means that we can deny market share to Microsoft in the short run by making companies like Amazon stwitch to Linux rather than Windows, and present a long-term threat to Windows.
LedgerSMB: Open source Accounting/ERP
Apple and Microsoft have the clout to do that. It's the benefit of a cathedral approach: you get to have a design pope. Gnome/GTK and KDE/Qt are trying to do that, too, but I doubt that they are going to have a lot of success creating consistency when they're already competing with each other. The proliferation of distros makes that job all the harder.
Do you really believe that Sun servers are noticeably more reliable than Intel-based servers from top manufacturers like Dell or Compaq?
I'd argue that they aren't. I can buy a refurbished Dell Poweredge server for under $9000 that includes a 3 year on-site warranty, and has plenty of hard drive space, CPU power and RAM to compete head-to-head with most servers I see people using from Sun with Solaris on them.
One problem I see with Sun hardware is that it's so pricy, people tend to hang onto it for a longer time before replacing it. That's not very sensible, because it leaves them behind sites on Intel platforms doing aregular 2-3 year upgrade cycle. (The Intel admins probably spend the same or less for 2 complete systems than was spent for one Sun server.)
If you have new systems every 2 or 3 years, you don't really need to be concerned if it's built well enough to run reliably for 7 or 8 years, now do you?
Now let's see, Microsoft is not growing on the desktop because they own 90% of that market so they are going after the server market which, at the mid to high end is running UNIX. Now Linux is getting in there and taking those BIG deals that Microsoft was gunning for and it's no big deal to Microsoft.
What bull. Microsoft have anything to do with this article? Sure sounds like spin to me.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
The lack of 'hot-swap anything' features make Linux every bit as unreliable, as DOS on any 'real hardware' you mentioned. There is a lot of 'low-end' hardware that supports those features, but not linux.
Does "Single User OS" mean anything to you? What about Cooperative multitasking. Running a server on Dos is like setting up a Solaris/CDE workstation and giving it to your grandma...
Oh yah, and the fact that Linux has to be rebuilt almost from scratch by the internal development teams working at those 'high-end hardware houses' doesn't mean anything to you guys, right?
Cool. YOu mean that Red Hat really doesn't do anything? Or IBM when we are looking at Linux?
And that is obviously why companies are really happy to move from Sun/Solaris to IBM/Linux?
LedgerSMB: Open source Accounting/ERP
Even if I presume that this is all true it:
1. Has nothing to do with the "anti-Microsoft press."
2. Doesn't address the hypocracy of an article that complains that the "anti-Microsoft press" is abusing its position by abusing his postion.
It seems like you both are having difficulty seeing that "server" and "desktop" are two different things.
Oh, and I forgot to mention before, get a fucking login.
-Peter
Yeah, but Novell is around (for the moment) and can support their products like gangbusters. Novell tech support has answers to just about anything that can happen to your machine - they have seen it all. We still run 4.11 just because it is so damn stable and you cannot replace ZENworks with anything - nothing even comes close.
Yeah... I seem to remember a story about one being walled in... I think it was even slashdotted...
Support is everything. I have been very impressed with the supportability of Linux over Windows and that may eventually be what moves market share...
LedgerSMB: Open source Accounting/ERP
If you have new systems every 2 or 3 years, you don't really need to be concerned if it's built well enough to run reliably for 7 or 8 years, now do you? For desktops and for servers you can afford to have down now and then, no. For mission-critical servers, yes. You want a MTBF of decades (if possible), so there is a very small chance of it going down in the 2 or 3 years before it's replaced.
That's the hardware requirement -- very good odds that nothing will break. Software requirement is, first, the only reason to _ever_ reboot is if you had to shut down to replace hardware, and second, a way to automatically switchover to the backup server if there is a hardware failure.If you see a Windows system in that sort of application, the person in charge is an idiot -- simply because rebooting to install a security patch is unacceptable.
I think that there is a benefit to having several core OSs. For example FreeBSD and Linux. For features and tons of community documentation, choose Linux. For stability, choose FreeBSD...
That heing said, more developer market share is better and provides a stronger base. I think that once proprietary OSs become beaten, I think we should look at the possiblity of making an Linux source code available for Free/OpenBSD developers. Obviously, this cannot happen until the Proprietary OS market is no longer viable. This would, however, help everyone out by allowing a greater degree of code sharing and good will...
LedgerSMB: Open source Accounting/ERP
If the highly paid IT staff is incapable of retraining *themselves* to a new OS and associated apps without a great deal of pain and suffering, then the IT staff is full of incompetent idiots that should be fired. Clearly such a staff has little, if any, native ability or talent.
I've never been 'trained'. It's always been assumed, in any job I held, that I would sink or swim on my own, and I'd better damn well get with the program if I wanted to keep the new job. I don't complain about this method; it separates the morons from the people who can actually do the job.
What all of these complaints about training tell me is that the IT staff in most organizations is just plain useless for anything other than a driver install or mucking about with the Control Panel. Christ. Talk about dead weight.
Max
My god carries a hammer. Your god died nailed to a tree. Any questions?
Ok, it's definately true that you don't buy a cheapo clone and use it for a mission critical server. But on real hardware (high-end Intel, RS/6000) Linux is every bit as reliable as commercial Unix. The only thing that's missing is "hot-swap anything" features that are only available on really high-end hardware.
Hmmm... Linux supports hot spares on Raid Arrays and will hot swap the SCSI disks if the bus supports it. So that point is moot. See the Software-RAID howto for details.
LedgerSMB: Open source Accounting/ERP
Irix used to hold all the highest uptimes for web servers on Netcraft. These are nearly all held by BSD these days. That is my measure of stability (on a web server, if it aint broke, don't fix it...)
LedgerSMB: Open source Accounting/ERP
I'm a Unix sysadmin with IBM Global Services (my opinions are my own, not those of IBM &c. &c. &c.). This is so terribly true that it's not even funny. What's more important is that every proprietary Unix is unusable without the GNU tools. They are each and every one complete and utter dog's dirt when compared to a Linux (or, I imagine, BSD) box. Kernel-wise, they are all superior--laughably so, I'm certain. But their userlands are unusable. Administering them is an exercise in masochistic abandon. E.g. AIX tries to be helpful with a Korn shell which offers--get ready to be blown away--vi command line editing! Yes, that's right kids, you get the joy of hitting Esc h to move left--naturally the bloody left arrow key doesn't work!
Don't get me started about the brain-dead qualities of such things as Solaris's umpteen-quatrillion version of standard utilities such as make, each subtly different from the rest. Or what about the fact that, when presented with the command ifconfig, both Solaris and HP-UX error out? Solaris will accept ifconfig -a to do what it is obvious you wanted in the first place. HP-UX, though, won't even do that. You get the joy of guessing what your interface are. Better fire up sam. Which, of course, uses $DISPLAY if available, unlike smit with its handy smitty TTY version (there may be an option buried somewhere, but I've not found it--not really looked, I'll admit).
You too can experience the joy of needing to perform hours' worth of manual install work, just to get a usable system. You too can run tar xzf and realise that it won't work, and you get to hunt down an unsupported binary package of gzip. If that fails, you can always hunt down the unsupported binary package of gcc--which you can then use to compile gcc, first, hoping that there are no undocumented trojans (wouldn't that suck), then try to get gzip to compile--of course, with any luck, some dead-simple thing which should be a standard, but isn't, because some half-wit of a marketroid wants to sell it to you for $$$$, is missing, and you get to find some bootleg version or, as in the case of none-broken /dev/randoms and /dev/urandoms, roll your own. Yee-frickin-hah.
Solaris, HP-UX and AIX are surviving solely because they are, for the moment, better OSes than Linux. Their deaths are inevitable because they are worse user environments than Linux. That, and the fact that every day Linux becomes a far better OS, while their half-hearted attempts at becoming better environments fail miserably (I love how gnome-cc crashes in AIX 5--that's an impressive feat!).
And before anyone tries to be clever, Microsoft's pap is neither a better OS (well, perhaps the NT kernel, beneath the layers of cruft, misdesign, maldesign and just plain folly, is--but no-one ever sees it, so it doesn't matter) nor a better environment. Unix makes sense. There's a learning curve. It's not user-friendly in some ways (rm being a chief example--users expect deletion to be undoable, and this is so achievable through several different methods), but it is extraordinarily user-friendly in others (e.g.: one app won't take down the entire machine; all basic utilites may be munged with stdin, stout and xargs; the learning curve is full of aha! moments, such as that glorious day that the full beauty of grep and, later, find is revealed in all its majesty).
Unix is great. The Unix culture is magnificent. Life in a Unix without the GNU utilities is the kind of hell I'd not wish on my worst enemy. It is an exercise in feeble-mindedness. Hello, Sun, IBM and HP: it's the third millenium--may I please have tab-completion now? While you're at it, can I have a make, a cc and a vi which are decent? Hell, can you just give me emacs and make both our lives more pleasant (mine, because I can get my work done, and yours, because I--and a thousand thousand other admins--won't be smacking you upside the head with the aluminium baseball bat quite so often)? Is it so difficult to master your bloody pride and admit that yes, a bunch of hackers turned out a better suite of utilities than your teams of engineers ever could?
Incidentally, the feeble-minded keybindings of AIX's smit are a direct result of engineers exposed to that most Unix-like of systems, VM. Hence the use of either function keys (which very often don't work across platforms) or ESC combinations (which take forever to type) instead of nice, simple, straightforward keys. Remember, PAUSE means scroll!
Sun, HP and IBM would do well to simply open-source their OSes and essentially abandon development. The Free Software community would do so much better with their code than they have ever done. Keep a dozen or so employees working on the code and co-ordinating development, and let things slowly merge into an OS which is already enjoyable to use.
You completely misunderstood the point of the previous poster. Solaris is a fine OS--much better in many ways than Linux. But its userland utilities are primitive beyond beliefs. Why can I not simply run route to get, say, the routing table? Why must it be netstat -rn? Why can Sun not ship a post-1984 shell? Why does ifconfig not do anything without a little -a tacked on the end?
There's no reason that Solaris (and HP-UX, and AIX) could not ship real tools--they simply do not want to. And their lack of wanting makes the admin's (and programmer's, and user's) job that much more unhappy. Not really harder; we can all memorise the incantations we need to force the system into behaving. Not even, really, more difficult (except when moving from an intelligent system to a brain-dead one, when we expect the fool to behave as the genius). But simply less happy and less joyful. The proprietary Unices suck the life out of one, misbegotten command after misdesigned `feature.'
Linux may be an awful OS (it's really not, of course), but it's an incredibly operating environment. I understand that the BSDs are much the same. Imagine, we let programmers do their thing and they write the systems they want to use!
I still believe that POSIX should be made a standard by the US Congress and support be required of all OSes sold. Just as we require support of the Real Honest True and Good Measurement system...
I'm a leaf on the wind. Watch how I soar.