Well there are a lot of misconceptions in the software world, and probably the biggest one of all is that "software is/should-be free." Truth is, software is probably the most expensive thing on the planet, in terms of the manpower required to write it and keep it running. It is so expensive, in fact, that the conventional business-model begins to buckle under the strain ... how many pure-software companies of any size do you actually know that have existed for more than twenty years? (IBM doesn't count... they sell hardware too.)

In my count, M$ seems to be the only one, although of course I'm missing some. Their business tactics might be cutthroat but they have survived, and they've written a lot of good stuff. As far as the conventional business-model goes, they've done about as well as could be hoped. But is history going to count them as "wise" or merely "fortunate?" Will they prove to be a product of the genesis of the microcomputer revolution, or will they also prosper in its maturity? IMHO, the jury is still out on that, because the industry is still adolescent.

At the same time, the demand for software far outstrips the supply, and the cooperative model of software-development (open-source, un-sold by itself) seems to be filling a large gap ... although quality remains a problem and sites like sourceforge.net are littered with more abandonware and barely-finished-ware than sturdy commercial-grade products. Still it is a start. The model seems to be the Gillette Principle: "give 'em the handle (the software itself), and sell 'em the blades (service thereupon)." That might well work, if OSS can really handle the development (to the expected level of quality across the board), and "service" costs can be predictably priced and controlled. Promises made, must be kept. Profitably!

Companies fully understand that nothing's free, as they themselves well know. (They sell their stuff. If they didn't, they'd just be a bunch of guys hanging out in an office building waiting for ... what? ... their welfare checks to arrive?) The OSS movement, rather than gloating over its accomplishments, needs to be thinking very hard about how to improve upon the quality-problem and how to keep "service" from sneaking its way back in to "funding someone to finish writing software-X."

Computer software is expensive, but it is also mission-critical, to borrow a by-now hackneyed old phrase. What it does, and the fact that it can be fully relied upon to do it, and the risk of failure, is far more expensive than the software itself. OSS is an important linchpin in moving this process forward, but it is no panacea and we should not be gloating. There's much more work to be done; much more improvement yet to be made. And we all have to eat.

I just got back from LinuxWorld in San Francisco last week..
This Linux thing is catching on with Large Vendors like HP, IBM.. to
name the top 2 at the conference..

Also had the opportunity to attend many sessions...several of which were
"case studies" of how companies were migrating to Linux..

In most cases, businesses and state govenments were migrating
areas to OSS apps like OpenOffice and firefox and then moving the
OS to Linux as the last step..
California and Mass. were the 2 that were at the session, telling
how everything was progressing..

The general trend in moving to OSS/Linux was to put it where
it made sense....
In most cases, this was the back end, servers (first)
then desktops that didn't use any "special" apps that only
ran on Windows...

Bottom Line.. Mixed environment.. Although I don't totally
agree with the above posters about M$ being a great OS..

It is however; USE the right tool for the right job.. and if that
tool is an app that only runs on M$.. then use it..

Personally, I really think more and more "non tech" people
are trying Linux NOT because of cost... but because of
the vulnerabilities...


Better said.. the endless stream of constant vulnerabilites..

Everyone these days has broadband
and most folks don't know the first thing about keeping
a machine up to date.. UNLESS it is automated..
which is something M$ is getting better at..
(i make a lot of my extra $ working on Windows boxes )


I have mixed platforms at home.. XP and Linux for desktops,
Linux for my mailserver and OpenBSD for my firewall...


The right tool for the right job...

my .02

The original Windows-NT design was done by a team that had worked on DEC Vax/VMS. It does have quite a different approach to things, and it was well-done. But there is no way that it can match the rate-of-development of Linux with its cooperative model: witness how "the next release of Windows" has just been delayed another two years. By the time M$ manages to release it, it'll probably be irrelevant. And I can't see it being profitable, either. "Windows 95" is not likely to be repeated.

There's no question in my mind that "Linux is here to stay. Period." And not just the Linux operating-system but this entire modus operandi. It may well be the only way to accomplish the rate-of-improvement that software needs. And as they say, "a rising tide lifts all boats." Attention has been focused on the OS first, and on mainstream office-applications second, "so that we can advance beyond these and move on." No one is going to stop it, not with patents nor with anything else, and it would be sheer insanity to try.

But the process still needs improvement. Contrast the development progress of AbiWord, which is still mostly still-born (imho), with OpenOffice which has the support of a (conventional, mainly-sells-hardware, etc) company. We can't just move forward using "pro bono work" as our only business-model. And it would be both simplistic and inaccurate to credit "free, OSS" with some of its most crucial successes given that so many of these successes came about either through subsidies offered by "conventional" companies or as a means-to-an-end to facilitate the sale of other things, namely hardware.

When we opened up the Internet, all those years ago, we de-centralized everything, and that includes software development. We also enabled the work of a few people ... Linus Torvalds of course, but how about also the inventor of Napster? ... to make an impact upon many. Then we turned the whole thing around and enabled the many to make improvements upon the work of the one! That last step is "what has never happened before" ... until very recently, it never could.

Continuing these random ramblings just a tiny bit more ... it occurs to me that we have de-centralized distribution, and we have de-centralized development, but we have not yet quite hit-upon the most effective new method for getting paid. We still rely upon one "centralized" thing, which we call "a company." We still rely upon these companies being able to reach some kind of critical-mass, which means "big." Which also means having to commute long distances to them every day burning ever-more-expensive gas. Somehow, the Internet must provide us with a better solution for this, which yet achieves the same results...

I agree that GNU/Linux and OSS are here to stay. OSS is the future of software development, IMHO.

I don't agree, however, with the opinion that Microsoft is a good company simple because they have "survived." Nor do I agree that Windows, be it 3.1, 95/98, or the NT/2K/XP line are good products.

The idea that Windows "just works" is foolish and without merit. This is nothing more than mass misconception. Windows has held a monopoly on the desktop for so long, that the flaws in its very design are just accepted as part of using a computer by the general public.

Windows relies too much on applications to release the resources they have allocated. Now, I am a programmer and I do make sure my programs release their own resources, but not every application is written with the same good practices. And in the event an application crashes, the OS should be responsible for releasing those resources. Windows does a poor job of this and doesn't give the administator enough control over processes and resources to make up for it.

Another major problem is the inadequate separation of user-level and kernel-level code. Any application can install DLL's or drivers, introducing uncertified code into the system. This code can run at kernel-level, completely unchecked and unprotected. An application can also modify the system registry (a good idea with bad implementation) without any verification or protection. These changes can cause other applications to malfunction, or even cause the whole system to crash. And Windows is the ONLY OS on the market that can and does break frequently when a user installs a new application.

When Windows does crash (and this happens frequently for most users), the error messages themselves are very lacking. They rarely give useful information, even for support technicians. This leads to the first rule of Windows tech support: reboot.

Now, no system is 100% perfect. Even Linux has problems. But Linux and other OS's offer some sort of "maintenence mode" for an administrator or technician to perform repairs to the OS filesystem. Windows does not offer anything similar. There is "Safe Mode", but all this does is swap the normal system configuration with a more conservative (safe) one. It doesn't give you any additional control over the system to perform repairs. The 2K/XP line does offer the Repair Console (only from the installation CD), but this isn't much better. It merely gives you access to the file system of the damaged OS. It doesn't help you to find the corrupted files. Most of the time a complete reinstall of the OS is required.

Yes *another* Windows flaw is the lack of code-sharing. Only DLL code is shared among applications. If I run 2 instances of an application under Linux, code is shared amost the two instances, greatly reducing the strain on system resources. Only part of the second instance is loaded into memory. If I run two instances of an application under Windows, two FULL instances of the application are loaded into memory.

And then there is the DLL system itself. There is no version control here. Windows can't distinguish between two DLL's with the same name, even if they contain COMPLETELY different code. And since installing an application can overwrite DLL's that other applications and even the system itself uses, a new application can break others, or even Windows. Granted, nowadays, a lot of apps will check to see if a DLL is newer than the one replacing it and prompt you for direction, but this should be handled by Windows. In Linux, I can have multiple versions of the same library installed if I want to without a problem.

Now, these flaws can be fixed. But will they? No. Why? Because it would require a complete rewrite of the OS from the bottom up. And that would break MS's monopoly.

Microsoft is not innovative, they don't care one bit about their customers, and they do not help the computer industry in any way. Every product they have put out has been bought or outright stolen. The only originality is in the packaging and marketing.

And as far as the US supporting the OSS movement? Not likely. Just look at the US government's history of trying to control the world from behind the scenes. We sell weapons to our enemies, instigate coups, and kill our own citizens to strengthen the power of those at the top. Look at the current "president". He is a war criminal. He was ILLEGALLY elected (do some research on the 2000 Florida recounts). And, I'm sorry, but his hands are not clean when it comes to the events of 9/11. If he wasn't in power, I doubt the incedent would have ever occured.

Want an example of US support for open source? Here you go: http://www.freesoftwaremagazine.com/..._blog_2_short/

OSS and the future of technology is in our hands. Don't buy Microsoft.

people are lazy and they expect something you plug in type on and forget about it that is the main reason why microsoft took off like it did ppl dont wanna hear " you have to type yum install update" they get scared.notice linux over the years is getting to be point and click with option of terminal. if people would learn that point and click doesnt "make you know anythiung" they might start going the linux way. by typing "yum install update" it gives u a sence of like oh god i just made the computer do something good" instead of "the computer did that"

Yes, ease of use is the big selling point for desktop users. My above post means nothing to them, and rightly so. And I have nothing against GUI's. But the simple interface that greets the End User should be built on top of a firm foundation. This is lacking in Windows. They slapped together a GUI (actually, they stole a lot of it from IBM and Apple, then when Apple threaten to sue, MS told them if they didn't license the MacOS code, they would pull Word and Excel support from the Mac) to sit on top of DOS, which wasn't a great OS to begin with. There have been design flaws from the moment Mr. Gates bought QDOS (a hacked, stolen copy of CP/M) from the Seattle Computer Company. These problems have never been addressed. They just keep trying to hide their sloppy code with a pretty interface.

And I don't think the average user should be stuck with an OS that is any less stable and reliable than a mission critical web server or mainframe or whatever.

Let's say you went to a car dealership and they had two cars to choose from. The first has no seatbelts, no brakes, no windshield, and an engine that blows up every 10,000 miles. Now, you can drive it with one hand (don't even have to use your foot to hit the gas pedal), and the interior is GORGEOUS, but....if you try to turn it to the right or left too sharply, the whole car falls apart. And don't try to install a 3rd-party stereo or put in non-certified oil or gas or wiper fluid...that will also make the car fall apart. It's a safety feature, you see, to keep evil companies from selling you substandard products.
Now, the second car is very reliable, has excellent breaks, a shatter-proof windshield, runs on corn oil, gets 150MPG, and offers a wide array of interior choices and stereo options. Oh...and aside from the cost of a service and support contract, the car is free.
Now, most people are going to want the second car. But the salesman...um...we'll call him...oh...I dunno...Bill, tells you that you can't have that car! No, no...that car is only for high-class executives carrying important documents and going to important business meetings. It's what they call a "mission critical" car. No, no, you want the other model. It's for the average driver that just goes to work and the store and such.
Now that is just ridiculous, right? Why should such a scenario be acceptable in the software world?

i agree 110% and as much as i cant stand windows i only have 3 linux distros now no more xp but when linux became popular werent they maketing for (carrer world instead of personal use and now if you really look at it the comapnies went to microsoft for same reasson now they can hire ppl that can click a mouse for 8 dollars and hr instead of a "linux user" for lets say 25 an hr. now ive had my share of linux problems but other than sound cardss they were my own fault. we all know linux is trying to be more user friendly but i think people here the terminal and run like hell little do they know thats an advantage not a ffearful thing. im working on getting a few ppl off microsoft due too the money end of it i cant see spending 300 dollars for office 2005 or whatever office is out now and at 150 for an os when you can download one for free. and ever since a friend had a problem with her pc problem stemming from win ME me and her called ms and next thing i remember is i got them to buy her a new pppc cause they screwed it up. so free phone tech service isnt worth a damneither.

You just gotta take a hard look at Macintosh OS/X! It is built on the BSD Unix kernel, with a smattering of Mach, and it blows the doors off of Windows, not only down in the basement where the engines are but also upstairs in the passenger compartment.

And Microsoft has good reason to be very nervous about Apple's stated intention to use Intel hardware, because that undoubtedly means that OS/X is going to be cross-compiled to the x86 processor family... where it can compete head-on-head with Windows. And Linux, for that matter!

In saying that last, I don't mean that "Apple will become the new Evil Empire," but simply that severe competition now exists on Microsoft's formerly-exclusive (or so they thought...) "turf," and frankly Microsoft does not have an effective response for it.

What the end-user wants is a machine that is reasonably secure, rock-solid, and easy both to use and to manage. If you want to see first-hand how this can and has been done in the Unix world, look at OS/X. If you want to see what Linux should be doing in those markets, also look at OS/X. The trick will be how to address the needs of both types of users, and do it profitably, without compromising the needs of any party.

I would like to also cautiously suggest that the Linux community must be very careful not to "compete" with itself in situations where there is no money involved. It's okay to explore for a better solution to, say, the packaging and maintenance problem, and to have two or three different development groups pursuing each alternative with vigor ... but when it comes down to the customer's question of "What shall I buy?" the Linux community needs to present a cooperative, well organized, unified front. ("A house divided against itself, etc.")

The Linux community must also embrace the notion that the customer will indeed expect to pay, and will have definite ideas of what s/he is paying for. And that is, "service." We need to de-geek this Linux thing. We need to de-geek the perception of "this Linux thing." It will take a workable revenue-model to pay for the level of support and service that customers demand, and the Linux community must not fight, on quasi-religious grounds, the very notions of that.

if i remember right OS/X was made by microsoft to let macs run a "windows like" envornment weather they used unix or dos im not sure but before OS/x macs didnt have a big source of operating systems othe rthan linux. dont get me wrong i dont hate MS but i dont like the prices they charge for the shit thats in the boxes.

I agree with your points on MacOS X, especially Tiger. I've never used a Mac (I'm poor), but just reading about their OS makes me drool. hehe
Tiger is a prime example of what an OS should be. Apple's decision to build a sold GUI on top of a solid *NIX foundation shows that somebody over there looked at what we were doing with Linux and thought "Hey...they might be onto something..." Tiger is what Linux could be. Its major drawback, I think, is its reliance on the PPC platform. The engineers at Apple don't really have to worry about compatability with various hardware platforms. So there is still some question, in my mind, of whether or not the OS would be as stable and intuitive on the multitude of different hardware combinations present in the Intel/AMD PC market.
As far as Apple stealing marketshare from MS on the x86 platform...I don't really see that happening. I think it would be a VERY risky move on Apple's part, and could ultimately bring about its doom in the desktop computing world altogether (they still have their music empire to draw revenue from).
I actually believe that MS is a greater threat to Apple now than ever before. In the past, MS has used its Mac ports of Office to keep Apple in line. They used this strategy to slap down a law suit from Apple when they outright stole MacOS GUI code for their fledgling Windows GUI. A simple threat to pull Word and Excel from the Mac got MS a license to the Mac GUI code. I actually think that MS is getting ready to move into the PPC world. The new Xbox 360 will be powered by a tri-core PPC derivative. The initial game demos for the console were run on Macs. Now, at first glance, its seems that MS will be putting all Xbox game development onto the Mac/PPC platform. But I highly doubt, giving their history, that this is the case. The Xbox 360 will use the same 3D API (DirectX 10/WGF) that the new Windows Vista OS will use. I doubt they will port WGF to the Mac. I have a feeling Windows Vista will be ported to the PPC platform. This could be very bad for Apple. And it would be a sensible move on MS's part. With OpenOffice.org becoming so popular, MS's control of Apple through the Office ports is slipping. What better time to pounce?
There is more info on the Xbox 360 stuff on my blog.

i was at book store today seeing a friend and i was looking at the OS/X tiger book and on the book it says that it doesnt crash and from what i know of macs that can be very true but im with you i dont have 3000 us dollars to spend on a small mac system than 300 for the OS/X. but it doesnt look that bad just alot of money

Well, you also have to think about this: How easy is it to switch? And if Billy thinks he can just throw money and get what he wants, he is right, until he has none left....

A point no one is hitting is that M$ didn't start out writting Operating Systems. They baught (or stole code from UNIX when necessary) ideas & products from companies too small (or poor) to properly market thier product. They would then copywrite these ideas, mass produce & move on. All of thier great achievments are stolen or purchased. Bill Gates didn't write DOS in his basement. That's urban legend. He may have wrote command.com, but M$ baught DOS. Then they gave it an interface similar to Mac & things havn't changed much since.

http://www.skrause.org/computers/dos_history.shtml

http://nwc.desktoppipeline.com/showA...leID=173602723

http://www.darwinsys.com/history/mslies.html

"2004-04-15 Microsoft using the State to kill free software?
Microsoft has started a campaign to license all implementations of "FAT", the filesystem type used in MS-DOS. This filesystem format is used all over the place, including most media interchange by flash memory cards, USB-based "memory disks", and more. It is also used for floppy interchange on all operating systems that support floppy drives, including the open-source operating systems like Linux and the BSD family (including OpenBSD). Since Microsoft have refused to offer a no-fee license to open source groups, it could easily be inferred that they are trying to use the Patent Office to put an end to the challenge they face from open source development, and in the process put the lid they have always wanted on all software innovation other than their own. One of the leading groups fighting this effort is the Public Patent Foundation; please check out this information on their intervention filed with the U.S. Patent Office to re-examine Microsoft's claim based on prior art. See also my history pages for where Microsoft originally got MS-DOS from."

-- http://www.darwinsys.com/

bill gates couldnt have wrote DOS DOS has been around long before windows ever hit same with UNIX long before linux. as for bill writing his own stuff. i hate M$ with a passion but if i had to guess he did what alot of people do take a source code or buy one and change it to make it their own. As for even writing his programs he DOESN'T WRITE CRAP he has people write it for him atleast the guy that wrote linux (name slipps my head right now) he still has his hands in linux. GATES became popular due to "his" mirochip. thats where he got the name for microsoft.

I believe vista/longhorn will be Microsoft's demise as Microsoft see's more pressure from American company's like the riaa/mpaa as far as mp3 and dvd etc.. Many people will be looking at alternative ways to listen to there music and or watch there movies. Bill Gates and company are walking around on glass over these issues and know that many are not going to like there new OS. They have to keep there customers along with riaa/mpaa happy. On the other side linux is not based in the US. Remember a few months ago the push to get linus to add drm to the linux kernel. These company's aren't making as much money as they use to and are up Microsoft's butt to help. Watch alot of people complaining about this in the future.And these company's will all be attacking linux and those company's supporting linux like IBM,Novell etc..Whatever happened to the IBM SCO(Microsoft) thing by the way? Microsoft is losing control and they know it or they wouldn't be attacking gnu-linux or trying to employ people who helped in developing gnu-linux. They're not the dominate OS anymore. And my guess is give them one more year and Bill Gates and company will be applying for jobs at Novell.