Slashdot Mirror
Linux Is Going Down
villoks writes "Doug Miller, Microsoft's group product manager for competitive strategies is trying desperately to find arguments against Linux." Many really good points, and many other equally bad ones.
← Back to Stories (view on slashdot.org)
I install networks for small business (10 - 100) users. I do install NT for those customers that demand it.
Let's look at some problems with your reasoning:
NT admins cost (at least in South Florida) more than $30,000 a year. You can probably get a very good help-desk person for this salary though. With an MCSE and a CIS degree, $40K is closer to what you'll pay. Add in the cost of the site license, SMS, your mail server and so on, and this figure quickly eclipses a Linux solution. I won't bother with the extra hardware resources needed for a workable NT solution versus a Linux solution.
A Linux admin for $60K? Sure, if she also knows Sun, IBM and HP, and she has a couple years experience. True, unix admins get more money on average, but they also generally take care of more users. But it's deceptive to claim that a Linux admin is $60 and an NT admin is $30K..
As for the remote user machines, I had not even added *client* licenses into the above. Factor that in and your costs go way up. But what difference would it make for the server? A Linux install would be transparent to remote users. They wouldn't have to relearn anything. On the same token, I can't imagine anyone preferring to remotely administer a Windows box versus a Linux box.
I guess it does come down to what the users and company will tolerate. Does a 20 user company want to hire an NT admin or would they prefer to install a Linux machine once and forget about it? Or are they under the delusion that, because it's NT, they can take someone part-time to service the box.
If they want support, I'll sell them a service contract for $15,000 a year (potential savings for a 30 user company is close to $100,000 vs an NT solution). For NT it's more because I need to send someone on-site because even a driver update requires a reboot.
Problem one: when all the Linux companies go "Tits up", hardware companies will no longer feel a need to release a few drivers for their products. They are only catering to the niche right now because they think that niche is growing.
Almost none of the drivers I have are written by companies. The best companies ever do is release register information, and sadly, a lot of the time it's reverse engineering work that gets things running. I don't buy this arguement for a second, and you should have a poke around in the kernel source sometime. Companies that are friendly to linux now have largely ALWAYS been friendly to linux, and I don't see any reason to think that anything more than goodwill is the cause - I doubt linux support affects huge percentages of their general revenue (linux users excepted).
Problem three: if we lose these companies, we will be losing many of Linux's best programmers. Reasoning: while some of the better ones are hobbists, a lot of the best coders work for money.
I work for money; I even make a pretty good wage. Even to the point of developing on microsoft APIs and microsoft platforms! I can say that some of the best code I've ever written has been for free or with an academic interest, and had very little to do with money. Most of the code I write for work is pretty mundane. This arguement doesn't hold weight. Back-in-tha-day, there was no commercial incentive, and there was still plenty of development. Although, XFree and Linux sucked more then. The suck less now, and will suck even less in the future! I love it!
They are coming to Linux not only because they see a development challenge, but a monetary opportunity through companies.
I laugh at you loudly. There are few if ANY jobs out there developing linux software. Mail me some information. They don't exist (relative to the opportunities doing embedded work, windows work, or generic network code, which I guess could be done in linux, but not exclusively for linux). Don't underestimate the 100-million-plus seat windows market, that's why you don't see games for Windows; Nice or not, we're not statistically relevant in that game.
Problem three: You seem to have many hours a day where you can code programs to give away for free. I don't. Most of us don't. Right now I'm going to college, but even now I'm swamped with work and expenses necessary to keep food on my table. I can only imagine it getting harder when I leave. That's why I can't help you in your idealistic ways.
I have very little time to work on free projects; That's why I hand stuff around - maybe someone else can do something with the little tidbit I wrote. Much of what I do is of little interest to anyone but myself - playing with genetic algorithms and 3D, for example. I do it for the love of the art, not the money. Linux was written by people that did it for the love of the art, and would do so regardless. There are very few things in this world I have any natural aptitude for, and coding is one of them. Why waste that gift?
That's why I can't help you in your idealistic ways.
Maybe I'm idealistic, but it doesn't change my original statement that if all the linux commercial involvement went tits up, very little of what I do would be affected in Linux. Linux exists outside of the traditional commercial world, and I see no problem with that. It will continue to evolve, and improve, with out without Redhat, VA Linux, etc. That's why we'll win (Eventually).
..don't panic
Except, of course, to the recipient.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
"a static growth rate"
I actually laughed at that one. Not a flat market share. A static *growth* rate.
e.g. Linux is growing at 5% a year; i.e. exponential growth
That's a sure sign of impending doom if ever I saw one. (NOT!)
-WolfWithoutAClause
"Gravity is only a theory, not a fact!"I just went and tried to swap the ram in my win2k/nt box since that is something they said Linux doesn't do - I assume they must do it.
Well, once I popped that CPU out (and burnt my hands) the machine kinda died. Hmmm....guess NT couldn't deal with it either.
I had a similar problem with trying to hot-swap the ram.
:)
""There really isn't much value in free," said Miller..."
...as opposed to paying $100 for the latest and greatest bug fixes?
Or maybe Doogie was referring to the value in paying hundreds of dollars per machine for a halfway stable OS (Win 2000).
Light a fire for a man and he'll be warm for a day. Light a man on fire and he'll be warm for the rest of his life.
Yes, but business is about the money. And with the exception of individuals who're independently wealthy or have someone paying all their expenses, most of us have to work in some sort of business.
As the viability of Linux in a business environment increases, so does my ability to deploy it where I work. The more Linux boxes and less Windows boxes I have to worry about supporting, the more my job becomes less "work" and more "fun". It's true that there'll always be a degree of work involved, but to get paid to do stuff you enjoy doing on your own time is a lot better than getting paid to do stuff you don't enjoy at all.
I'm a fan (and occasional practicioner) of the noble art of trolling, and this one is amazing.
The second paragraph is the best:
These are three key Linux trends to watch for in 2001: a static growth rate, lessening mainstream interest in the open source operating system, and a sharp decline in Linux-based companies' stock value, said Doug Miller, Microsoft's group product manager for competitive strategies.
This is beautiful. It's irrefutable.
Obviously linux growth will slow, you can only grow at an exponential rate for so long before you run out of servers and people to run them.
Similarly with "reduced mainstream press." At linux ceases to be a novelty, the mainstream press will start giving it normal coverage.
Finally, the bankruptcy of linux companies will be a side effect of the venture capital spending spree having caused some linux companies to get funding without a solid buisness plan. With the bursting of the internet bubble, they'll have trouble making that second round of financing.
All three trends are clearly in evidence and obvious.
The clever thing is to use them as proof that linux is doomed.
Doug Miller, I salute you. You have a gift for inciteful comments that appear logically sound at first glance.
If you ever want to start trolling slashdot, let me know. We can hook you up with a low user id account with plenty of karma.
--Shoeboy
The DNS servers where not running Linux before the crash. Why they are running it now is because Microsoft outsourced the DNS handling to Akamai (you know, the distributed content serving network) to prevent this mistake from happening again. Akamai is one of VA Linux's biggest customers and run virtually all of their servers on Linux.
The problem is that "real" enterprise servers don't ever go down, even when they change the hardware. Linux doesn't support many of the features that have been in more traditional UNIX servers for a decade already. Likewise, there are several companies making PCs that have hot swapable PCI cards, CPUs, etc. Does Linux support any of those? Nope. Does Linux support disconnection and reconnection of SCA hard drives? Does it even have a completed journaling file system?
Don't kid yourselves, guys.
Linux is awesome for hobbyists, good for workstations, and debatable for enterprise servers.
OpenBSD doesn't even support SMP, so don't feel all bad. Nobody has every feature. It's just a matter of priorities. Linus hasn't put enterprise features as his number one priority (yet?). Maybe in the future, Linux will compete better.
Get out into the "real world" and see what a real server can do before you start talking about Linux taking over.
(not a troll) (--- that is how you can tell it's not a troll) (would I lie to you?)
Seriously, I'm just trying to inject a little reality into the blind advocacy. I've run Linux for some 7 or 8 years now, so I obviously like it well enough.
Some nits to pick...
"The problem is that things are getting more and more complicated - very soon, things like SMTP will be obsolete, and only groupware like Exchange will be viable - simply because it's more productive for a company to have groupware."
Exchange is not groupware O.K. I'm biased, as and ex Lotus Domino/Notes programmer. But Exchange does not, in itself, contain enough to be groupware (add on VB and MS .Net, then it gets somewhere.)But this is similar too the O'Reilly book on groupware. In said book the author shows how to use various open source applications towards of the goal of creating groupware.
"There isn't the money in open source to be able to afford to produce things like this - because there's no revenue in giving things away, companies can't afford the programmers to produce the complicated products of the future."
IBM, amonst others would disagree with you. MS's revolution was showing that one can make a lot of money off selling software, regardless of platform. I think Linux will show one can make a decent living off services supporting software. For example, IBM will happily set up Linux and an Apache web server for you. Integrate that with a legacy database, and show you how to develop applications (preferably with Webshpere of course.) Now the OS and web server software are free, but how much money is IBM going to make helping you to set all that up, quite a nice profit. It these sorts of services where companies that support open source software will succeed. Companies, of various sizes, will always need and be willing to pay for experts that can walk in and help them out.
"Even Netscape, bankrolled by one of the world's largest companies, AOL, can't keep up, via open source, with expensive protocols like XSL and so on."
Errr... irrelevent, I would contend that AOL doesn't really care that much about Netscape, execpt to own the technology (hedge against MS) and to own the programmers that go with that tech (another hedge.)
This will take a while though - the first thing to happen will be the death of consumer open source. I posted an article on this to Kuro5hin, and although the poll died, the majority of people agreed with my conclusion that open source isn't viable for consumer software.
Here we somewhat agree, I don't believe open source will every take off for end-user applcations (The Gimp excetped of course).
But I do think it will continue with the writing of drivers and improvement of applications that provide services (like Apache, mySQL, in other words applications used to make end user applications, or provide services, via the web to end users.)
With the rest of your points (lack of funding, lack of innovation, etc.) I think your off base. Similar to Java applets will run everywhere, I believe the open source movement has moved passed the users will update/change etc. the code on thier machines. I think r+d will continue on those applications that provide services (again web servers, db's, etc.) and that the commercial potential will be in tying it toghether as a solution for a customer. The end user won't know what's going on because they'll access these applications via a broweser, or some other interface. Even MS believes this is the future, look at .NET, another attempt at distributed computing. .NET as a unified solution, or will companies like IBM be able to market a collection of open source applications that can do similar things.
The future will be fun... pcThis is where the battle will be fought next. Will MS be able to market
Funny, I think "free and it runs pretty well on an old PII-450" is pretty good value compared with the costs of four Athlon-class servers and four Win2000 server licenses, a MSexChange license, and another $500 per user for Orifice, $200 per user for the OS, when all I need to do is give my developers to send email to each other.
But let's take a closer look at his points:
Think about the pointiest-haired boss you ever worked for.
Now imagine him as CTO of your favorite bank or brokerage, and running into a board meeting, hollering "Oh my God! SUNW and ORCL are down more than MSFT from their 1999 dot-com-hype highs! Throw out that obsolete Sun E10K server running Oracle and get me a farm of Quad-Xeons, we need .NET, M$Exchange and M$Access!"
OK, maybe there are some PHBs dumb enough to base technology decisions on today's stock quotes, but not many. Evolution in action, and all that.
"So I'm a Linux user. But I don't think Microsoft cares. The reason is simple: both of my copies of Windows (one 95 and one 98) are licensed, as they came with my computers (both Dell). Microsoft is getting paid even if I don't use their software. Most of you probably know the name this has been given: the Microsoft tax."
Oh, I think they care. Did you get a complimentary copy of Office with those machines? Are you signed up with MSN? Are you creating content in their proprietary file formats or are you using your Linux box for that? Do you surf the net with IE or Netscape? Do you use Real Audio or whatever that Microsoft format is?
The operating system license is just the tip of the iceberg for Microsoft. They wouldn't be after their competitors so viciously if all they wanted was operating system money.
Microsoft wants it all. Right now, they're spending stupid amounts of cash on engineering and QA and just giving stuff away but, as they say, free is a pretty shitty long-term business model.
Of course, Microsoft and many others still don't seem to have grasped the free part. Nobody is trying to make money selling Linux. Linux is an operating system kernel and is freely available to anyone who wants it. People are making money selling Linux products, services, hardware, appliances, etc. These people aren't using "free" as the business model, they're using "cheaper, faster, better".
c.
Log in or piss off.
"...recent security problems". Compare NT vs Linux intrusions here.
If you want to make a convincing argument against something, first make sure that the reasons can't be turned against you...
Richy C.
Actually, they're using NT servers. They just forgot to remove the linux identifiers from the stolen code. I'm surprised it doesn't say "Stacker".
Over to you, Bob . . .
Without a standards-based, crossplatform language, any hope of "taking over the Net" is so much vaporware. Such an effort is even more pitiful when you have to contract out to the competition for basic services.
The irony of the situation is that IBM, the company Bill nearly killed off in the 80's, is at the vanguard of the host of companies set to sweep the 800-pound gorilla off its feet of clay. Big Blue has spent the last 20 years turning itself into a 600-pound Rocky Balboa, a lean, mean, fighting machine.... and, at the end of it all, Tux is their mascot.
But as I said, it's not JUST IBM, not JUST Linux even.... it's BSD, and Apple, and the Alpha platform, and Sun.... and the fact that Windows STILL doesn't run on ANY 64-bit platform and at this rate may never....
Open Source has been around a lot longer than Linux. (I remember downloading "less" in 1986... and the comp.sources.* archives were pretty huge even then.) It's not going to die anytime soon. Furthermore, IBM is not stupid, not anymore. It wouldn't have put such a huge investment into this if they thought it was a short-lived technology. You don't see IBM stock losing 80% of its value, do you? In fact, IBM is outperforming the S&P, the NASDAQ, and the Dow. The Street prides itself in being able to predict future performance very accurately. (Please also note that MSFT does none of these things...) So I'm not just blowing smoke here... IBM will have its revenge - living well while a greying Bill stands off an I-405 exit ramp holding a sign, "I will code for food"...
I'm shocked that people that high up in Microsoft and other "consultancy" firms completely miss what Linux is about. It's not ABOUT money. It never WAS about money, and frankly, redhat, VA Linux, and everything in between can go tits-up tomorrow and it won't make a lick of difference to me. I'm sure there'd be a lot of unhappy investors - but let me say it again, Linux is not about money.
Linus Torvalds did not write linux because he wanted to be rich - although a nice side effect - he wrote it because he wanted to do something; he wanted an operating system that just sucked a little less than all of the other ones out there. That's the beauty of the GPL. That's why I give code away - It did what I wanted, and if someone thinks that it sucks less, then all the power to them!
I use linux because it does what I want, and so do a lot of other people. Linux won't lose because a bunch of ill concieved business models go up in smoke - all that GPL'd code will be there, waiting for the next Linus Torvalds to hack on it and make it suck less. Those drivers weren't written by people who wanted money; they were written by people that just wanted their hardware to work. There's no rocket science in there - just a pile of time.
Unless microsoft is proposing that they ban free development - free as in speech - then there's a segment of the market that they'll never, ever get - and that's the real linux mainstream, the core of people that use it because it sucks less and makes their lives easier. Does anything else really matter? If you're happy with MS, fine. Enjoy. I'm not.
..don't panic
So if Microsoft is so against this horrid O/S because of security problems.. why are they using some Linux DNS Services?
Look Here For The Info
Let's see which name servers Microsoft is using right now: microsoft.com. 1d20h7m8s IN NS DNS4.CP.MSFT.NET.
microsoft.com. 1d20h7m8s IN NS DNS5.CP.MSFT.NET.
microsoft.com. 1d20h7m8s IN NS DNS7.CP.MSFT.NET.
microsoft.com. 1d20h7m8s IN NS DNS6.CP.MSFT.NET.
microsoft.com. 1d20h7m8s IN NS z1.msft.akadns.COM.
microsoft.com. 1d20h7m8s IN NS z2.msft.akadns.COM.
microsoft.com. 1d20h7m8s IN NS z6.msft.akadns.COM.
microsoft.com. 1d20h7m8s IN NS z7.msft.akadns.COM.
Let's do a queso on the last four.
$ sudo queso z1.msft.akadns.COM:53
216.32.118.104:53 * Linux 2.1.xx/2.2.xx
$ sudo queso z2.msft.akadns.COM:53
32.96.80.17:53 * Linux 2.1.xx/2.2.xx
$ sudo queso z6.msft.akadns.COM:53
207.229.152.20:53 * Linux 2.1.xx/2.2.xx
$ sudo queso z7.msft.akadns.COM:53
213.161.66.158:53 * Linux 2.1.xx/2.2.xx
It's Linux, all right.
Oh, and that S390 will run Linux. So if you want a true enterprise OS, it seems VM/CMS, MVS or Linux are possible choices.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
That is hinted at is that in the free software world it is often much harder to get "big" features implemented in a timely fashion. (In the article they are specifically talking about enterprise features but those are just one particular kind.) I mean, take a look at how quickly commercial operating systems like BeOS and Windows NT supported journalling file systems. Then take a look at how widespread it is among free operating systems. How many clustering solutions are there for linux? Now compare that to the number of mp3 playing front ends. The easy stuff gets done over and over again while the hard stuff gets done once. If it gets done at all.
With free OSes there is often little in the way of financial backing for more ambitious undertakings. Look at who extraordinary the recent support of the perl hacker is. I mean, it makes front page news that some guy gets to spend 100% of his time working on improving the product. When was the last time you saw Microsoft trumpeting the fact that they had hired a person to work full time on Visual Basic?
Of course, it's not IMPOSSIBLE to get good funding to implement more difficult features in free software. IBM and SGI are both doing so, more or less. However, the article does mention that many/most linux based companies are suffering from financial difficulties, which in turn will make it harder for people to get the kind of funding they need to do more ambitious work.
So let's see where we're at:
- "The Linux kernel lacks key enterprise elements
.."
- "You wouldn't want to install Linux on a laptop
.."
- "Free does not sustain a business
.."
- "Linux growth is leveling off
.."
I think the CEO of LinuxCare said it the best: the significant thing here is the degree to which Linux is registering on Microsoft's public radar. We must be doing something right, folksReeeeeeaaalllly. What "key enterprise elements" are those? With the latest Linux developments, we've got everything from a journalling filesystem to enhanced multi-processor support. Sure, it's tough to make the claim that Linux is going to be superior to Solaris or other "big-iron" Unices for "big-iron" applications, but IMHO it's tough to make that claim about Windows, as well.
This, of course, conveniently ignores the fact that the vast majority of Linux installations (just like the vast majority of Windows installations) do not require these "key enterprise elements" that Ballmer is bleating about. And what are these elements, anyway? Mindlessly throwing out buzzwords might make "PC Magazine" swoon, but people who are interested in specifics are going to yawn and be on their merry way.
Is that so? Funny; I just installed Mandrake 7.1 on a Dell laptop last week. The installation went flawlessly. I was up and running and connected to the Internet, reading Slashdot, within two minutes of finishing the installation. As a matter of fact, the PCMCIA modem that I'm using with the laptop was not recognized by Windows. Linux didn't have any problems with it. What was this nonsense about lack of drivers again?
Maybe, maybe not. It depends on the business. But the real issue here is the one that they missed; if every Linux-based business goes belly-up, that does nothing to hamper the continued development and release of the Linux system itself. Sure, companies such as Red Hat have got people working on value-added software such as RPM, but if Red Hat were to vanish from the face of the earth, it would not prevent the Linux kernel from evolving and undergoing continual development.
I think we can chalk this up to simple ignorance; people just don't get that there is no single, controlling corporation behind Linux. They look at Microsoft and see them as the source of the software that runs their computer(s). They don't understand the Linux development model (or if they do understand it, they don't like it because it is so far removed from their expectations.)
Show me the numbers, baby. At my workplace, we've got Linux replacing Windows NT on many of our development workstations. We've got Linux servers coming in the door to handle many specialized data applications. We're putting together Beowulf clusters to do distributed data processing. We're getting rid of clunky Oracle Forms-based user interfaces and replacing them with ones developed using Troll Tech's Qt toolkit. In short, we've seen a Linux explosion over the past year or so, and I know that the same is true of several other places.
We're going down, in a spiral to the ground
Howdy. I work for MontaVista Software. We support embedded systems development on Linux. Incidentally, many of our full-time employees are among the core developers of Linux/PPC. Many of our clients use PPC-based systems. Thus, we pay these folks to work on the issues our clients experience -- but the source goes out to everyone.
I package and port applications. Should I make a significant portability fix, it goes back to the author as a matter of course, to be included in the next version.
This isn't an unusual thing; not only do we operate like this, but Lineo (our primary competitor) does the same. I expect that non-embedded support companies also have very large numbers of individuals doing linux full-time.
Why some other companies put out press releases when they hire folks to work on open source full-time is something I don't understand. We do it as a matter of course.
Doug Miller, Microsoft's group product manager for competitive strategies,says, " the new Linux kernel lacks some of the key elements required for enterprise use".
Well, there it is right there. Now we know why there were so many problems with the Enterprise: Starfleet was running Windows.
Microsoft didn't subcontract to Akamai to manage their DNS servers.
They subcontracted cached content delivery to Akamai, basically as a means to reduce the effects of DoS attacks by distributing their content across multiple Akamai servers across the globe, thus preventing an attack against one machine from taking everything offline.
So now when you contact the microsoft web site to grab something, instead of going to Seattle it may be routed to a Akamai server in Chicago which has the content cached.
Obviously in order to do this, Akamai has to be able to manipulate DNS entries for Microsoft's web servers, thus you now have Akamai DNS servers listed as authorative for Microsoft.com.
This was all discussed in numerous news articles this week, which you apparently missed.
I'm probably a typical Windows > Linux convertee. Up until about 18 months ago I used Windows and Windows alone. A few months before I switched to Linux I started a private FTP server (for legal files, mind you) on my computer that was in my dorm. Well I got tired of rebooting every day, so I made the Big Leap. I started off dual booting, then moved to a Linux-only system about a year later. I recently got a laptop that dual boots, but only because I have to use certain Windows applications for school.
So I'm a Linux user. But I don't think Microsoft cares. The reason is simple: both of my copies of Windows (one 95 and one 98) are licensed, as they came with my computers (both Dell). Microsoft is getting paid even if I don't use their software. Most of you probably know the name this has been given: the Microsoft tax.
So I really don't think Microsoft gives a damn about the desktop market, for the most part; they've got it locked up. Server market is a different story. The article makes some good points. I don't think there's much of a market for "Linux companies," perhaps with the exception of the well-knowns like Red Hat. But does Microsoft really have to fear Red Hat? I don't think so.
-- "Complacency is a far more dangerous attitude than outrage." -Naomi Littlebear
"So in some senses (that) puts the Linux phenomenon and the Unix phenomenon at the top of the list"
Unix phenomenon? You mean the fact that most servers run Unix and not WIndows? This is very telling. An uneducated businessman will read this and think, "Gee, Windows must be under attack by this thing called Unix." which by impication means Unix is the minority.
If you use the language of a champion, you will project the fact you are the winner by default and people will all beleive you are the winner. That's the best advertising anyone can get. "Use Window2k, it's the future". But the problem is, there are a lot of people who don't know that Microsofts future is Unix's yesterday...
Burn Hollywood Burn
What Microsoft is afraid of here is a de-centralization of technology. You see it's a very old war. One that I have seen fought at least twice on a grand scale. It's a war that is only now is starting to expose possible Information models. This profound impact on a info culture could be better understood by once again looking at how it was when Microsoft was fighting the good fight against centralized Technologys. Yep thats right. There was a time when people toughted MS as the savior of the age of computers.s
So what do i mean by centralization of technology? I'm talking about the way one group can control the use and propagation of of technology. Back in the day when hooking a computer to a TV was a neat idea. IBM ruled the jungle of technology. It did this because at that time no one shared computer design. IBM won the hardware war and as a spin off also dominated the bussiness software side too. One thing that IBM did was to aquire companys that had solutions that IBM could put to advantage. Yes even Microsoft was looked at as a possible purchase solution. What Microsoft used was the fact that everyone ( I say that very loosly because only a few people were real excited about the pc revolution back in the early eighties ) wanted there own computer. It was not good enough to run down to the campus (school/work) inorder to get computer time. Accountants could not work at home with the latest marvel spreadsheet. But we have this theme of wanting a personlized and handy form of computing. It was this notion of not personal use but personalized user of the computer that took IBM by surprise.
Microsoft provided what seemed like the greatest solution. A OS that ran on the cheapest computer hardware. ( sound familar? ) Apple showed the market existed and only controled it till someone did it cheaper. Also lets look at the competition. To the public at that time which computer to get was confusing. ( I had a Vic20 ). When IBM backed a cheap off the shelf based computer the bussiness world made the plat-form stick. ( how many times have you heard RISC is better and wondered if it was better then why don't we use it). There is a push for the home computer not because everyone wanted the same thing but because everyone wanted the convence of personal computing with out the access restrictions of central computing on MainFrames.
What changed? Well we have had may years to enjoy the advantages of cheap hardware that has de-centralized our need for large computer with less then friendly access restrictions. Now we have this great platform to automate our more mundane calculations. But it's the thing we would not give up that gave MS it's power. That thing was interoperability ( sorry but I'm not going to correct spelling on something this long when i should be working ). with other computers. We all wanted to run the same programs. Share files and print with the same fonts. This translates into centralizing the software as a trade off for computers in the home. And for the most part its a great trade for everyone.
Microsoft is now the target of de-centralization of Technology. Well we all have cheap computers. And they all come with MS because everyone wants to shop at the same place and eveyone wants to send there thought to others that will be able to understand them. But wait! What about the people that are looking for a different way? The Heretic of centralization in all of us looks around and thinks that maybe I want it to do this. And I want it to do it this way. When your a developer this seems to come up more then when you bought your computer for e-mail. but as more people start wondering what they want to do with there computers they are finding out that it can't be done like that. The reason is allway because the central controling forces just don't have the resources to make everyone happy. "Do you want one thing done right? or Manythings done half assed?" The people that want one thing done right are not happy with a Operating system that wants to do everything you can think of and not let you see the gears.
Does the fact that Microsoft is the largest mean that Microsoft is the best? I don't think so. I think i means they were able to take advantage of the fact that they did not have to worry about the Heretic in the early days. And why would they? At that time I had to go to the local college or too my mothers place of work to get computer time befour the Vic 20.
Do i think linux will ever "Go Down!" ? No I don't. and here is why. Opensource packages like linux distributions come with source code. that everyone knows. But what makes source code valuable? It opens up the software for de-centralized evolution. This is the same de-centralized evolution that gave us leaps in Video card technologies. The more people find out that they can do exactly what they want instead of what the instructions will let them do the more de-centralization wins.
So we get to the linux changing the world part. With out the cost of a centralized Operating system companies now have a choice to invest licence fees into people instead. People that can get technology to do that one thing really well. If you have ever had a Microsoft consultant come out to look at a deployment the first thing you will mostlikly notice is that person doing all the things you've been putzing with for the last few weeks. This is because the world of Canned software does not have much in the way of configuration choices. This is great if your people are not so brite. It also helps bring the cost of people down. One reason why I saw IBM's OS/2 lose it's base was because of a lack of people to write code and admin. Actually there were people there. But Microsoft knowlege was cheaper then IBM knowlege.
What you get with the Microsoft solution is the same solution that your competor has. You will be more at the mercy of MS for inovation then your own IS staff. You will have to spend more money on software that could go to keep people and hire new people.
The winning motion of linux is the ablity to foster real inovation from the ground up. Take what you need from the CVS repositorys. Beat it into submission till all the data is processed right and you have inovation. Real inovation. And Technology is fed with de-centralized revolution not centralized predicted evolution.
I don't think linux will ever die. And I know it will not get less "market share" in future years. I'm sure that de-centralized revolution will change linux well beyond what we know as linux. If linux did go away then I'm sure that instead of MS saving market share it would be because a new and more malable Operating system grabed the attention of the inovators.
When I grew up and went to school I was taught that a good programmmer writes portable code that was expandable. To me portable meant any operating system and expandable meant being able to be used for purposes that I did not forsee. Linux lets me see things that windows hides from possiblity.
Last one in jail is a fascist.