Slashdot Mirror
IBM Desktop Linux Pledge, One Year Later
Blue writes "It's been more than a year since the bold announcement from IBM that they planned on dumping Windows for Linux throughout the company. InfoWorld is reporting that not all is well with IBM's desktop Linux push. What went wrong?"
I can honestly say that a lot would have to be done with their own internal applications to bring them to Linux. Domino client won't run. Neither is a Sametime client available. Both were in heavy use in IBM Global Services, at least.
I don't understand the unwillingness to port these two desktop pieces (both being on Linux would be handy where I am now), but between that and the web apps, they have a lot of work ahead if they want to fulfill a Linux desktop.
There are alternatives - Wine as depicted in the article. Crossover Office supports the Domino client. Meanwhile, the extension for Gaim, works okay as a Sametime integrator. Still, none of those solutions would lend themselves to correcting the internal issues at IBM. They have control of the apps - porting them natively is logical.
HBI's Law: Frequency of calling others Nazis is directly correlated with the likelihood of the accuser being Communist.
OMFG. The only thing they use is IE. That's the only thing their help desk is, uh, helpful with. I'm sure that's not where all their problems are coming from, but it speaks of an organization that isn't at all agile.
I love a lot of the things that IBM does and comes up with, but if your organization isn't flexible enough to work with more than one browser, you've got some serious problems.
Sounds like the Microsoft Lifetime Employment Program has deep roots at IBM.
--- Submission is feudal.
She almost sounds a little dismayed, perhaps even slightly frightened, by that fact.
This is really a shame, as it's so close to being good on the desktop - if someone with a clue and a small workgroup worked on it, within a few months it would be there.
The pay versions (e.g., Xandros) might be good enough, I haven't tried them. But why bother, when I know that Windows is quite good, and inexpensive?
Sigh.
The Domino web interface is ungainly and not standard. Furthermore, it's not Section 508 compliant which means US Government sites using it are being converted to other technologies rapidly.
You wouldn't tolerate it on your own web site, I suspect, and users never have liked it.
HBI's Law: Frequency of calling others Nazis is directly correlated with the likelihood of the accuser being Communist.
The average user doesn't want Linux. THey want the familiar Windows.
I work for IBM. Most of the people in my department who have a preference want Linux on our ThinkPads. Some of us dual boot anyway. I use cygwin on XP for now. Once there's a certified Linux C4EB, I'm switching.
the company's internal use of the open-source Windows operating system emulator did not translate into a ringing endorsement...
In other new, WINE is now an emulator dispite its name.
Have you even read the article?
Do you REALLY believe that a *corporate desktop* computer
needs to "easily install software" or "configure printers"?
I, for one, would cut off your fingers if i caught you installing
crap on one of my company's workstations.
Having seen more than one medium-sized company deploy desktops,
I'm positive that all computers are already configured to run
anything they need to run and print anywhere they need to print
*before* they are presented to the user.
No. The problem is (as you might have learned if you R the FA)
is at the application level. They are running into problem with
web-based applications that were geared towards Internet Explorer.
They are running applications on Wine (which they list as a
temporary workaround themselves).
So, you are right in that there are problems yet to be fixed,
but completely failed to put your finger on what it is that needs
fixing.
Interesting... I find myself exponentially more productive with Linux than I am with Windows. Unfortunately I'm forced to use Windows at work, and I'm just about always sshed back home so I can keep my sanity. I of course use Linux on all of the computers I own.
Game! - Where the stick is mightier than the sword!
Over a decade ago IBM was hyping its OS/2. However, IBM's desktop sales department refused to sell OS/2 preloaded and went with Windows.
Now IBM is hyping Linux, but IBM's support and web development groups are sticking with Windows.
IBM refused to make a choice with OS/2 and lost big times. It's going to have to make a choice with Linux. Let's hope the powers that be at IBM chose wisely.
If someone says he and his monkey have nothing to hide, they almost certainly do.
Huh?
I *always* read Slashdot with Firefox and it always looks fine to me....
Try .lwp. One of my biggest challenges being a Linux using IBMer is opening up documents created with Lotus Smartsuite. Open Office says it'll do .123 files, but in reality that only works less than half the time. It doesn't even have an lwp (Lotus WordPro) converter.
When I want to get WORK done, I boot XP.
:-) Others prefer the full Start Button, System Tray thing, good on em. If thats what you need to be productive, go for it.
:)
Depends what work you do, I guess.
I work for a living and use Linux all day, every day. I'm much more productive now than 3 years ago when I was forced to use Windows. (Unix sysadmin for 15 years)
I work in a very Windows oriented office, IIS based Intranet (ntlm auth reqd), Exchange, Windows shared directories etc, but there's NOTHING I cant do on my Linux box.
A GUI should be a personal choice. Personally I use a very minimal FluxBox, because it suits the way I work. (To me a GUI is a way to have lots of command line windows open at once
Oh, and when I want to get WORK done, I don't boot linux. It's alwaysi running. (Barring unfortunate UPS issues
The bank I work at is absolutely, crack-ho, addicted to IE as a platform.
It's like they went out of the way to be grautitously microsoft-locked-in.
If you ask about replacing the hundreds of crufty web servers, most of which are from Bombay BTW, with linux, then you obviously don't understand the extreme levels of blame-avoidance at any big company.
Because any project to replace a couple hundred goofy stupid poorly understood crufty web pages is bound to fail.
The really sad thing is that ActiveX has only been around about 10 years. It's not like this used to be a good idea that fell out of fashion, but then it was too late because they were trapped in a legacy -- it was always dumb, from day 1. This story isn't about Linux, it's about how IBM fucked themselves by not thinking. It's about how they didn't fire some idiot in time to prevent long-lasting damage.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
In 1994, I was working for a hospital trying to deploy OS/2. We'd had some success (most notably with the medical library), but we had challenges too.
In particular, there was a DOS-based package that we needed the workstations to access. OS/2 supposedly ran DOS apps as well as Windows, but this one froze up randomly. It was written in dBase or FoxPro, one of those database/language platforms.
The vendor (basically, the guy who wrote the code with a few people as a support staff) practically bent over backward trying to get it to work. He offered to give IBM the source code, if they would only sign a non-disclosure agreement. Remember, at this time nobody showed source for a commercial product; it was like giving away the crown jewels.
And IBM wouldn't do it.
That was the straw that broke the camel's back. OS/2 hung on for a while there, but the day the IBM rep called me and said they would make no further effort to get the package to work, its fate was sealed.
The problem was, while IBM promoted OS/2 publicly, there were all sorts of people there who knew Windows, liked Windows, and undercut OS/2 at every opportunity (in typical passive-aggressive fashion). Maybe they were Windows experts and didn't want to learn new things. Maybe they thought Windows looked better on their resume. Maybe they used OS/2 1.0 and never got over their initial negative reaction. But whatever the reason, corporate fiat couldn't win the hearts and minds of a lot of their employees. The same thing may be happening here. (They only support IE? WTF?!!)
I wish IBM well in their Linux effort. Maybe they'll eventually pull it off. But it's gonna take more than a decree from on high.
Garg
Garg
Alumnus, Xavier's School for Gifted Youngsters
I'm aware that IBM makes lotus. I have no idea the costs or time to port it to linux, but my point was IBM would have been better off if they looked at that more closely before trying to switch to linux.
A bonus, if successful, would be to sell that to other companies as part of a way for them to help in their linux migration. It would set a good example that their customers can trust the software works well because IBM themselves use it. Again, I don't want to assume anything and I don't want to say IBM made a poor decision. However, they themselves admit things didn't go smoothly. And from the reasons they gave, it looks like they can learn from these mistakes and hopefully migrate to linux after the applications they depend on are available to run well on linux.
Well, I spend a lot of quality time with Linux and a chuck of the IBM stack, so I tried to move over about a year or so ago.
Learned more about the innards of Linux than I ever cared to trying to get the OS to work on one of the higher end IBM thinkpads. SuSE SLES 8, more or less the standard for the WebSphere stack I work with, had a kernel that did not see the Ethernet port and the video was a mess. Due to a how-to and forum support on the Gentoo side, it was the first distro I got the xwindows and the wifi card working on! With a bruised forehead and a better understanding I went back to SuSE and got it to work as well on another HDD. Life was good. Problem was I suck as an installer, and getting the base to work (because I don't know the underlying details) was far worse (for me) than the development I was trying to do on it.
Eventually I tried the new SuSE Enterprise 9 (and desktop version) with the new 2.6 kernel. OS worked like a charm - many of the things I googled and dug through forums to figure out 'just worked'. Even Gentoo packaged up the hard bits to update. Unfortunately, it would seem that DB2 needed tweaking to get up and running, WebSphere was far from stable, and WSAD was a wreck. Same when I updated the Gentoo drive as well. With an extra six months, more config tricks, and a few helpful service packs it sort of works. This is my daily driver, however, so I reverted back to the older kernel.
So to sum up - it took about three months to get the hardware working, about the same to get the apps working, and a lot of work to do in between where I really should not have messed around with the system. My boss would die if he knew how much time I spent coding versus trying to just get the app server to install. I know the *nix gurus out there would laugh at my bonehead moves trying to get 1400x1050 to work (and then 3d acceleration), but I'm the type who had to hit the man pages to add users! So much easier today now that the hardware is a bit more mainstream. I'd say it was a year too early if they were gunning for the unwashed masses (like me).
+++ UGUCAUCGUAUUUCU
Along the same lines I was talking one day with my wife about switching from windows to Linux .... While she was working on the computer. No problem. She didn't want to do it, and then forbid me from changing what she had. So Along those lines she is still running a heavily updated Mandrake 9.0 ....
... when installing Open Office.org on a users box (especially if they are not a "power point ranger" type.). Never call it a "switch" Refer to it as upgrading their Office software. The reception of the new product will be much higher.
Go figure.......
I also did this in an office I worked in. I copied the "splash screen" from windows ME to a number of Win98 and NT boxes. I then spent the following week listening to people complain about how they didn't like ME as well as what they had and wanted to switch back. So I changed the splash screens again.
Finally
I'm sorry, I'm to tired to be witty at the moment so this message will have to do.
that keep big technical corporations from moving. Its the 1000s of little apps written by engineers and departments to do very local, very special little tasks. It takes years to move all of these because the corporate big wigs will never recognize the problem and realize that they need to send 80% of the transition funding to the people that wrote the invisible 80%+ of the applications. If they were a non-technical company where every geek didn't have their own set of apps that needed porting, the transition would actually be easier.
But a little bird tells me that IBM have created an OpenOffice filter which imports MSWord files absolutely perfectly. Without doubt IBM could set up a server to do the conversion of the .doc files everybody has on their PCs. The same bird tells that they are terrified of releasing the filter, because the full flight of Redmond leagles would come screaming out of their eyrie. One war at a time is sufficient to keep any organization on its toes.
"there are still a lot of legal questions surrounding Linux"
There are no legal questions surrounding Linux. There is a great deal of FUD however, and I have certainly seen bogus 'legal concerns' used to try and block internal work at IBM I have been involved with.
The facts are that Microsoft has been convicted of code theft and patent infringement and no Linux distributor or customer has. Period.
"Linux distros now cost MORE than Windows to license"
There are NO license costs for any Linux distro. Zero. None. They don't exist. The only sosts are support costs. Ask the folks at Earnie Ball what they pay for support. Funny, a small non-tech company that makes guitar strings can out tech-support IBM?
This is another classic in-house funny money game. One of the best examples I saw of this was a GIS group that was crying to have Suns for their work, but the in-house cost listed for Suns were higher than any premium one-off contact I had ever seen, so Windows 'looked' cheaper. What those poor folks had to go thru to try and get their work done. I personally went thru hell and high water trying to tell some folks who finally got permission to buy a Linux server NOT to get the RedHat Enterprise product for the two machines because of the absurd price. Even after buying it, the support stunk and it just didn't work. We wound up putting SuSE professional on them, and found this was what most of the Linux underground at the organization was doing anyway. $75 for those who wanted the printed manuals, and no cost to put on any number of machines you want from the media in the box BECAUSE THERE ARE NO LICENSE COSTS. It's GPL folks, and can be redistributed as you wish. Now, finally, you can download the DVD from the net for it as well, in addition to the long standing FTP server install.
Now, Microsoft does cost, per seat, and I have seen their 'enterprise support' at a 12K seat level and it is laughable. It makes the early 90's AIX support look charitable.
Bah, this makes me want to spit.
The fix is in the current gecko devel tree.
I heard that a few revs ago...
Someone set us up the bomb, so shine we are!
Some people say that the linux desktop will never happen. And to be honest, sometimes, when the latest version of Gnome or KDE comes about and screws everything up, I almost agree with them.
But just take a look at OSX. Unix, that Mac users can use!? Surely you're joking Mr Jobs. But he's not.
What's the difference here? Why is apple able to make a usable gui interface for unix, and yet, after years of development Gnome and KDE have not? Which is not to say that advances haven't been made. Far from it. Linux is more usable now than ever. But it still isn't as usable as it could be. Linus is not yet at the stage where developers can sit back and say "Ahhh! Now there's a usable system". Dispite all efforts Aunt Tillie still cannot use Linux! Why is this? What is the reason for Linux's failure and OSx's success. This question needs to be addressed.
If I had to guess, I'd say it's that Microsoft and Apple take a wholistic view of the OS and Gui, changing fundamental configurations in the OS layer to better facilitate GUI use and administration. Whereas linux window managers are just that. They must change themselves to fit the linux OS paradigms, which may not fit so well to the GUI paradigm. Gnome and KDE cannot change fundamental aspects of the OS and so must work around them, where as Microsoft and Apple can change one to fit the other. Well, that's my best guess anyway.
May the Maths Be with you!
Are you trolling or is this for real?
The idea of 'su' or the more modern 'sudo' is that you can let users run as 'users' and still have the ability to escalate to 'admin' rights to do priviliged operations, some users can escalate and others cannot, but all run as 'users' for normal operations. This is demonstrably the best way to implement user rights on the modern desktop, it prevents stuf flike viruses and spyware from being able to proliferate to non-user areas of the disk where they can affect other users. See Mac OS X, which has a well thought-out implementation, out-of-the-box Linux is not exemplary of what you can do with sudo.
As for 'too hard for the average user'...
I've converted two housemates to Linux, these are people who know nothing but email, chat, wordprocessing, and web. After a few days of occasional questions, they're total converts. Linux is NOT hard to use, it's hard to geek on, especially if you've been perverted by a Windows-only experience so far. Windows is really the ugly bastard child of operating systems as far as I'm concerned, it's still trying to reconcile it's past as a permissions free-for-all.
Dammit, Microsoft thinks 'Documents and Settings' is a more intuitive location for user profiles than 'Users' or 'home'! They don't offer a decent CLI shell and SSH-alike for people who need to admin their servers from cellular links on their PDAs! They ship a desktop OS that loads with defaults to prompt 'yes or no' to -execute unsigned binary code- right from a browser window. They ship this same OS with services for file serving, directory services, remote desktop, remote registry access, and lord-knows-what-else listening out of the fscking box, and they rectify it by enabling a software firewall that defaults to 'on' -three years later-.
Where are the file extensions? Good fscking question. Last time I looked Windows just chopped off the dot and the last three letters of the files and presented them that way, it also hides the entire root of your drive and the system folder. All the competitors use this really cool utility and library called 'file' that has the ability to type a file based on it's CONTENT, which is much safer and more sane. Some systems even use the filesystem to store -metadata- about a file, which is superior to both methods but not nearly as easy to work with or support from a developer's POV.
Alright. I've eaten your flamebait, and spat back burning embers. I await a response. I'll be asleep until 6 am EST.
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
I must first begin by saying that I am probably less experienced with Linux than most in this forum, I have been an on and off Linux user for about 4 years now. First Red Hat, then Mandrake, then Susue, then Ubuntu...But I am a business guy and partime geek. :)
From a business perspective, what I see mostly missing from this entire equation is a company, or team to step up to the plate and say: "Linux has potential...why don't we offer high quality linux hardware and software products?"
I am still not sure why this has not happened. Is there not enough of a market?
Why is there still not an easy way to install software like people are used to in windows? How can you expect hardware manufactuers to have their products come out with ALL of the distributions on a daily basis?
I installed Mepis on my sister's computer about 3 months ago and she has had relatively few problems, but when she calls me up to tell me she cant play a WMV file embedded in a webpage, and I have no answer for her, then there is still a ways to go.
When there is still no gui wireless roaming client for laptops, linux will still be a second class OS.
I say this not to bash Linux, but let's be honest, there is ALOT of work to be done in the Linux for this OS to be ready for everyday people.
Now, I am no Linux developer, but I would DEFINITELY be inetersted in seeing how many Linux developers understand the average user. Until I find a linux developer that actually cares about what the "average user" really needs and, instead of pretending that Linux is already a real alternative, is dedicated to bringing a usable alternative about, then I do not see the Linux desktop stepping into primetime.
Just my humble opinion...
Actually, children are so good at seeing through their parents' bullshit to the underlying behavior that at the same time they are most decidedly NOT learning to floss just because they are told that they should, they ARE internalizing this parenting "technique" to be passed on to their own progeny.
I know, I know, -1 Offtopic.
Last I checked, one program written for linux probably wouldnt install properly across all the various distros. That issue alone is stunting Linux's acceptance as a serious OS, but there are many more such issues.
And, of course, is the always-ignored (at least by the linux community) issue of a higher TCO, since you are then required to retrain each and every staff member on how to use their new computer, how to use their new programs, etc. 99% of people enter into corporate America knowing how to use Windows- retraining them to use something new is just, from a resource standpoint (both time and money), plain stupid.
at a presentation.... what struck me was that they used Windows left and right on the clients... I still had the announcement in mind...
There are millions of people running an old version of windows 98 or even windows 95. Their old machines are not upgradable to newer versions of windows.
Sell a small form factor box with Linux installed for $300 based on a mini-itx platform that has a 40GB drive. Setup this new computer next to the old computer. Plug it in. Connect the two computers together with a usb cable, or failing that, a serial port cable.
Install a software package on the old computer that will copy over the users preferences, bookmarks, background images, sound files, themes, data files, database files, old mail message, programs and just about everything that could be configured or installed or downloaded on the old machine. Filter out all the viruses and spyware so it is blocked.
Once the program says that it is done, power down both machines, disconnect the old machine and reconnect the shiney new machine. It boots up into a shiney new version of Linux with all the glitter. Except all the settings are configured. When you open your browser up, it has all the book marks and opens to the same start page. When you open your mail program up, there are all your mail folders and the mail servers are configured. If you want you can hit F7 to switch to an MS windows window that looks identical to their old destop.
Do this and you will get a million customers.
Palmisano's challenge to the corporation was to be technically capable to switch to all-Linux on the IBM corporate desktop, not to actually do it. That's a big difference.
I can tell you that as someone who is currently working within IBM Internal help that linux is not supported. I am actually taking calls as we speak (work from home) If I was to receive a call on linux the only thing I can suggest is to check out our internal linux info page. There is very basic support on this site. We did however receive an email asking for individuals that have linux experience to signup for a new Linux helpdesk. I have signed up but as of yet not word. This was a few weeks ago.
However, the past 10 years has seen M$ firmly implant itself in the corporate desktop suite and it would take the next ten years to dislodge it. Not just the M$ Office applications (REAL programmers don't use spreadsheets or even a word processor...), which for many users, there is no suitable substitute -- I'm looking at the parade upon parade of dorky, kludgy, awkward third party Windows applications that now have pervade the business environment, both in IT and general business users. Another strong irony is that a good bit of this stuff is now Java based, which was touted as "write-once, run anywhere" but totally dependent on Windows to run. Either via custom Windows desktop client software, or piggybacked on MSIE or through proprietary database requirements that alternative OS usage was never ever factored in by the vendor selling. Go stroll through the software suite of any large corporation (most all of which are IBM clients) and it's heavily laden with gooberish offerings totally reliant on the Windows platform. Even the server software will have frontends unusable without IE and/or Windows.
Even if the software and hardware fulfilled the bill of need for business usage, users would still resent and resist change from familiar work patterns. This will always occur, even if the change is an obvious beneficial move of immense proportions. To a business user, even those computer savvy, it's a learning challenge hoisted on top of an already filled worklog platter. A mandate has to come down from above, that a change has been blessed and sanctioned, and that there is no choice in the deal.
In my view, most firms would profit hugely from a switch, at least those entities not dependent upon special software not available in alternative OS (including Mac OS X along with Linux) -- more stable, less virus/malware/spyware concerns, less employee "goofing off" factor (most games are Windows only), etc.... ...but then, expecting a large company to behave in a cost sensible fashion is folly, as they'd rather pay someone else to guarantee the deal or take the blame when things go south... ...at the shop I presently work, I've heard the network and system support engineers (and their managers) bemoan the existence of Linux and FOSS at our company, that they'd much prefer it all was HP/IBM/MS stuff, so they could simply "open a ticket" to the vendor to fix a problem......and it fits in with the "let's move it to India" instead of hiring a few good people and letting them manage the systems... ...but then I've drifted into another rant here...
AZspot
I am a linux fan as much as the next guy but I think you are comparing apples to oranges in terms of drivers support. You have a brand new system built in late 2004 and are installing an operating system from 1999. Its no wonder there is poor driver support in Windows 2000. The system integrator provided the approriate drivers to make your system usable. Why Windows requires so many reboots is a different story.
I think a better comparision is a Suse distribution from the 1999 era.
- The goddamn backspace key. It seems to be impossible to tweak it satisfactorily so that my
.cshrc works across different distros and in every application.
- Copy and paste. I've no idea what's going on here. Different applications use completely independent cut/paste buffers. Simply copy-and-pasting from my web browser into a text window can be a headache requiring me to paste temporarily into an intermediate application.
- Shared libraries. I can run plenty of old Win 95 apps on a modern Windows XP box. Plenty of old Linux binaries will fail to run on a modern distribution. Downloading third party apps like RealPlayer is a real nightmare.
- Focus. Many applications pop up windows but they fail to get focus. Nothing is more annoying than doing a search in acrobat reader, say, and having to actually click on the search window to bring it into focus.
The fact is - all of these problems are soluble. But I'm no longer that single young kid who thought it was cool to spend all night hacking away to fix the most trivial problems. I now just want these things to work. They do under Windows, they do under MacOSX. No doubt some smart young Linux zealot wil tell be how to solve the above problems. But that's completely missing the point.Just so it's not all negative: it's a pleasure to have a working command line again. CMD.EXE is so, so, broken.
Doesn't it make you feel good to know that our freedoms are protected by politicans, lawyers and journalists.
Hardware support falls very much under the problem category of "negative reinforcement." Hardware vendors don't see a vital market still using Linux. Hardware vendors also do not know where to begin support on Linux due to the clustered (seen Heartbreak Ridge?) nature of Linux environments and stutter steps of incremental yet glaringly incompatable (backwards that is) changes. This limits users and developers, whom then think the solution is to make more changes that break more systems.
Note that what I am talking about here in reference to the changes is as follows. Lets use a generic system of "units" shall we.
In general terms, you have a system that is has a problem of 10,000 units magnitude. Primarily, lack of modern and semi-modern features including support of hardware... FULL support that is reliable and works with no fuss. Next, we see that in addition to various toy development (some very good theories on optimization) which may or may not include some degree of fixing the problem. Maybe hacked support drivers and user-space tools. Maybe retooling of any of the various layers/interfaces and messaging systems. This retooling is hoped by the most optimistic Linux developer to fix 50 units worth of problem but will break compatability with 150,000 units through direct and indirect means.
This requires more fuss on the part of the user, administrator, security team, and developers. This affects hardware vendors. Hardware vendors, looking at the pattern of Linux changes and direction, avoid supporting these changes for quite a long time for fear of investing so many resources into something that is so volatile and doomed to fail when chrome-011-56 is added soon after they could expect to release drivers and user-space tools. The hardware vendors overall lose more interest with each new change to the underlying core Linux system.
The result is dropped or extremely lessened support for Linux by that hardware vendor. The Linux community chooses to ignore the pattern and the lesson here and instead kicks their legs and flails their fists like a child who is denied that ice-cream trip because they poured their glass of milk on the TIVO after already it having been established that said child knows not to even bring food into the living room.
Wisdom does not mean you don't make mistakes (p.s. that implies that the wise are not perfect) but rather that you learn and adapt based upon those mistakes. Fools willfully refuse to accept reality and feel it more appropriate to cry victim than to solve the problem. The stupid are those who then lash out in attempt to destroy those that do adapt.
I am hoping that the maturity and professionalism found so much more in the BSD realm will hold steadfast against the inevitable influx of immaturity and stupidity as a result of more adoption of BSD to fill the gap of Linux.
There will always be Linux (or "a linux" rather) to fill the niche of folks who desire not a usable tool but rather a playground of chaos to tweak, hack, and play in. They care no more about the usability and stability of their toy than do they care about always being 2 steps (although that value grows as time passes) behind the expected level of functionality brought about by other operating systems.
To Microsoft, Linux is indeed a powerful explosive. However, I believe that MS has realized they can use it as ablative armor and thus turn it into a protective and strengthening tool. The burst of hype of Linux helps force improvements in Microsoft products, funny how competition does that. Meanwhile, Linux falls further behind in functionality and usability and further distance themselves from being a useful, no fuss, end prod