According to the Free Software Foundation, free software includes "the freedom to improve the program, and release your improvements to the public, so that the whole community benefits... Access to the source code is a precondition for this." While I agree that the principles of the FSF are noble, I also feel that there is an unspoken assumption - an assumption that pods of hobby developers across the world can coordinate on the same scale that directed companies with a budget can. Where free software has an important place in computing, so does closed-source commercial software.
Post a Comment
Adam, many companies are successfully using dual-licensing approaches as they find the GPL and various commercial licenses serve two different market segments and two different sets of needs and concerns.
There are definitely times and places where you want code to be freely available. Think of a publically funded medical expert system, for instance. Free software is appropriate for those projects that have benefit to the community and need safety measures against commercial hijacking. Or all the tools you need to build and maintain such a system. It makes sense to have those tools available to the people who funded their creation so no community, society, or country is unable to access better medical care.
And there are times when a commercial license is appropriate. For instance, when selling product into corporations that need/want/expect a certain level of support from the vendor. MySQL for instance, is available in the corporate market via a commercial license.
The dual license provides both markets with what they need. The freely available open source provides trust to the market, something that we find missing for many commercial-only software products. Parts of many governments are switching to Linux not strictly due to cost, but because they can trust the code. This is why Microsoft hastily put together their governmental source viewing program to stem the tide of migration to Linux.
Also, your article is disingenuous when it comes to Linux:
1. Slow mouse tracking is very fixed in the latest development kernels and with the next stable release of Linux, this issue will no longer exist. The next kernel will also have ALSA, providing a stronger foundation for sound.
2. It is clear to most people that the quality of Windows 2003 is not comparable to that of Linux because the resources available. If the Linux community put in billions of dollars of funding into the next version of Linux and took four years, I'm sure it would be very good as well.
The concept of free software is not new. 'Public domain' software has existed for a long time. However, the FSF/GPL take on free software mandates that the software, including changes, be put back in the public domain. I think that's fair and appropriate for many applications which benefit the public.
Commercial software has also been around a long time and it is not going away anytime soon. However, this software exists largely for applications which do not benefit the public at large and have no public accessibility requirements.
There is typically not a single software license that serves the needs of the commercial software developer and the public domain developer. I think in the coming few years, software multi-licensing will become more popular as it enables both public and commercial benefit.
It is clear to most people that the quality of Windows 2003 is not comparable to that of Linux because the resources available. If the Linux community put in billions of dollars of funding into the next version of Linux and took four years, I'm sure it would be very good as well.
That's exactly my point. The resources AREN'T there, because too many Linux users refuse to pay for software. If users would pay for software, companies might be attracted more to the platform, and we'd end up with better software! I don't see how that's disingenuous.
I must say that this was one "free software" themed article that I actually enjoyed reading. Honest, down to earth and void of zealotry – Cheers!
Free software was never a matter a price.
The important distinction is not between free (as in zero price) and commercial software. It's between free (as in freedom) and proprietary software.
Free (as in freedom) software deserves support, monetary and otherwise. Proprietary software does not.
You should pay for your free software. I do.
Not sure I agree with your take on this - as a software company whom is actually making a dual license structure work I see nothing in the article that suggests that this is not an appropriate path. Perhaps I should review it again but I think his read on this is pretty spot on in allot of ways. Note: I said, in allot of ways and not in every way - an important distinction as I don't agree with every point/example in the article.
Perhaps an article should be written on all the possible definitions of free(as in free) as recognized by the FSF.
However, IMHO free(as in freedom) means you have the right to buy(as in spend money) or not to buy(as in not spend money) regardless of the assiociated License.
The GPL, which I'm assuming your referring to, is not about freedom at all. It's about ensuring that you get something back for the work you've done.
First, commercial software gives immediate returns in the form of money. There is no choice in the matter (well, within the law I suppose).
GPL software on the other hand takes the following attitude. I don't particularly care about money, but I care about code. If you make changes to my free code, I want them. An important point to make is that the owner here is not an individual, but a nonexistent entity. None the less, it is all about returns, not freedom.
If you really wanted to be a nice person, you would release code without expecting ANYTHING in return. This would be FREE software. I would have the freedome to do anything I want with it. The same cannot be said of GPL software.
That's exactly my point. The resources AREN'T there, because too many Linux users refuse to pay for software. If users would pay for software, companies might be attracted more to the platform, and we'd end up with better software! I don't see how that's disingenuous.
You need to look at more than just "users pay for software = good software".
1. You've made the assumption that every need can be best met by a commercial for-profit enterprise. This is not usually the case when it comes to software that is for the public benefit. There are many parallels in the real world as well. Many times public benefit companies cannot be private enterprises because the private enterprises tend to put their needs ahead of the public's needs. Government regulation helps in these matters, but do you want every company that produces software for public benefit to be heavily regulated? That would certainly add overhead and cost, taking away from what benefits can be delivered.
2. Microsoft is funded by years of illegal monopoly money, so there is no comparing what Microsoft can do vs. what the Linux community can do. I will say it again, if you gave Linux billions of dollars in funding and four years (for just one major revision) of a zero market pressure environment, I'm sure you could turn out an amazing OS that would make Windows 2003 look shabby by comparison.
3. Just because something is commercial doesn't mean it is going to be perfect. Take the example of QNX which is privately funded small company that makes a well-regarded OS product. You will find many versions of their OS had problems with the mouse pointer too. By the way, this was well before any cheap/free versions of their OS. There are many many examples of commercial products which are very poor quality.
4. When you examine Linux, you find many price points, from free all the way to enterprise server licenses that cost thousands of dollars. Many Linux users do pay for software. Even on the very low end, with the small but growing Walmart market that Lindows and Lycoris have created, there are more people everyday who are paying for Linux.
5. Many people cannot afford a privately-produced OS. Sure many individuals who don't have the discretionary money to pay for a privately produced OS do not pay for their Linux OS. But many of these people come from much poorer countries than the US and Europe. In Indonesia, desktop Windows costs more than the average person makes in a YEAR. It is unreasonable to think this person is going to buy any Microsoft OS. So this person either pirates Windows or copies Linux from someone.
You yourself are a Windows user and a Linux user. I presume you are paying for your Windows software. Do you refuse to pay for your Linux software? Are you completely unwilling to pay for any Linux software? If you were running a commercial enterprise and needed the benefits and support of Red Hat Enterprise Server, would you refuse to pay for it?
When we look at what is happening today, we find two things.
First, more and more people are paying for open source / free software, one way or another. We see great success in dual licensing, showing that those who are deriving commercial benefit from free/open-source software are willing to pay for it.
Second, we find that many commercial products are low quality. Private enterprise does not result in quality. Good software development results in quality and Linux is very good quality. Linux delivers far more quality per dollar invested than any Microsoft OS.
While I agree that the principals of the FSF are noble,
Cool. I think Richard and company are pretty noble folks as well. Besides that, their principles are noble too! 
Andrew B. wrote:
However, IMHO free(as in freedom) means you have the right to buy(as in spend money) or not to buy(as in not spend money) regardless of the assiociated License.
Then I don't think that's in-line with the FSF's meaning of the word. You're thinking of "gratis".
Free (as in freedom) in the FSF sense means that the *software* is free. Regardless of whether or not you pay for it, the source will be freely available if anyone distributes the software.
Not necesserilly in price, but the user should still have freedom to modify the product for his or her needs. just like if I buy a couch or whatever, i can do absolutely wahtever I want to it/with it, same should go for software.
I think the guy has a pretty good point. He didn't say that all software has to be commercial, just that sometimes there are benefits to the commercial approach. I don't think anyone except perhaps Stallman himself can claim otherwise.
It is hard to claim that Free Software does not produce good software. It is, however, equally hard to claim that Free Software produces software that meshes well. The division between KDE/Qt and GNOME/GTK is a perfect example. They both write good software. They do, however, demonstrate the man's point that Free Software doesn't coordinate well with itself.
Sure, the latest windows may feel good to you, but you have to ask yourself: how long did it take Microsoft to get to that point? MS has been around for more than 20 years, and windows has been under development for almost that long, but until very recently, and their resources notwithstanding, windows sucked big time. It took a lot of effort, mistakes, trials, retrials, before microsoft finally started getting it right. So I don't think it is fair to explain whatever advantages windows has by refering to the fact that it is a commercial and proprietary system. The fact is: windows is maturing with time, and so is linux. In fact, I personally think linux is maturing at a much faster pace than windows.
And if you look at webservers, well, IIS has yet to catch up to its opensource counterpart. And things like postfix and bind and way ahead of their closed source versions. Sure, there's a niche for both closed and open source software, but putting a priori limitations on the possibilities of open source for purely ideological reasons is premature.
> how long did it take Microsoft to get to that point?
10 years. All software gets better after ~10 years according to Joel On Software.
> MS has been around for more than 20 years, and windows has been under development for almost that long, but until very recently, and their resources notwithstanding, windows sucked big time.
You have misconceptions here regarding the development of the Windows NT-branch which only started in 1991. And in 1999 they already had a great product out, Windows 2000. The desktop version of it, XP, came 2 years later out. Win2k3 Server is even better.
You can bitch at Win9x and Win3.x if you want, but don't put all Windows in the same bag just because they share the same name. They are using different code and architectures.
Adam,
You should get your facts right. Just because one gets the operating system for free (which is becoming more and more a commodity nowadays IMHO) doesn't mean that he/she won't pay for good quality software. There are other major problems concerning this, like the QT licensing, for instance. QT is as powerful and mature as any toolkit available at Windows, but people refuse to pay Trolltech what is due to develop comercial software which I find a little odd, since you can't get MFC at Visual C for free in Windows neither. But I never hear people complaining about Visual Studio.NET licensing or something like that.
Unfortunately, GTK still is not mature as QT but is getting in there fast. The others toolkits are fighting their place at the sun, like wxWindows and FLTK. And the Xlib, which seems to be the Unix standard, is too hard to develop anything on it and is starting to show it age.
Well, The Kompany has been developing comercial software for KDE using QT for a long time now and I'm sure they are pleased with the multi-platform nature of it. They provide a Windows version of most of their software and in some cases, if you buy the version for Windows, you also get the Linux version and vice-versa. Even Adobe itself has turned a licensee of QT to develop Adobe Photoshop Album and they loved it.
See? There is space to comercial software here...
Personally, I don't know anybody that says "Screw Adobe and their Photoshop stuff. We don't need them. We've already got GIMP and GIMP rulez. Everybody knows that!!!" (Please note that I do love GIMP! :-)).
Some people might point that the opensource nature of Linux tends to lead into several versions of slightly incompatible versions of it, what might scare away comercial developers and therefore the Catedral-style should be widely adopted to make things more feasible for those companies. Well, I don't see them porting their apps to FreeBSD neither, which do provides that kind of environment.
My point is... Adobe, Macromedia, Discreet and all the others can port their software for opensource OSes if they want to. Piracy isn't even a good argument since people already get pirated copies of their software in Windows or Mac OS X anyway. Nobody really cares if closed source is what it takes to they do that (only a small minority of zealots would argue against something like that). It's up to them to see if such move fits their needs market-wise.
But to say that it's Linux/BSD users fault's that those companies don't support those systems as they deserve is really naive, to say the least.
DeadFish Man
> It's not unreasonable to pay for goods and services, and I don't believe that software should be an exception >
I fully agree!!
The cost for an OS or S/W Applications are irrelevant in the corporate world. They use it to improve their business and to stay competitive. In other words they are investment goods, as robots and other machines are in the car machine manufacturing industry.
Take AutoCad as an example. It costs mega bucks, but there is hardly any machine manufacturer who doesn't use it. The same goes for any other specialiced ( and i include OFFICE as well) Application.
Or SAP. It costs hundred thousands of $. Compared to those figures Xp cost you peanuts. (It comes pre-installed anyway).
May I be wrong but, as far as I know, many developers that use to work in FS projects are employees of big companies like IBM, Oracle, SGI, HP, Red Hat, Suse and so on. And many of them are researchers that work in universities over all world.
Of course, that are hobbyists (and many are professionals) working too, but they are a plus (and sometimes, a BIG plus). And, if you believe in statistics, and know how to interpret them, you can believe that more people looking the code can help fix it.
Also, I like to think that programming is one area of science where the theoretical background can be compensated by creativity, what is a lot harder in others areas. Again, with statistics, there is a good chance of a good idea come to light from this crowd and, if something is wrong, be fixed.
One thing that many people overlook, is that even though programmers at Microsoft, IBM, Oracle and in others companies can't cut and paste FS code on theirs projects, they are free to see it, learn and apply it whether they learned something or not.
Anyway, isn't because exist some big company behind one project that it's going to be a success, the history give many cases that proof it.
I believe this article was writting with a certain amount of naivete. I don't mean to say that Adam Sheinberg isn't knowledgable about linux, but simply that I don't think he is plugged into the way that most users think about free software. He seems of the opinion that free software users insist all of their software should be free. This is further supported by his reply to an above comment: ...because too many Linux users refuse to pay for software..
This assumption is ridiculous. Many linux users I know keep a windows partition for games (which they pay for). I believe transgaming does fine with winex (which people pay for). Crossover Office is constantly being improved and sells plenty of copies despite the fact that it costs money. Linux users don't insist on having all free software. It is just that there aren't many companies making their software available to purchase on linux. Look at Adobe. How long could it possibly take to get a working port of Photoshop or Illustrator on Linux? They have a version for OSX so I am sure it wouldn't require an incredible amount of resources. However, they don't even take the time to try and get wine compatibility. How about Macromedia? There is still no linux shockwave plug-in or dreamweaver.
The reason you see so many open source alternatives in the linux world isn't because of any great refusal to pay for software (although there are those who would like to see a free version of everything). It is because companies refuse to make the investment. Personally, I wouldn't mine seeing every company who doesn't have linux versions of their major software by 2005 go out of business. If that were to kill Adobe or Macromedia, then so be it. They have great products, but they refuse to support a community that has been clamoring for these programs for awhile.
Ask any linux user, chances are they will tell you that commercial software being ported to linux is a good thing.
I just read DeadFishMan's comment after posting that last one and he said pretty much the same thing as I said (even calling Adam naive). Sorry for the reduncancy, but it is good to see that others agree with me.
I agree with your point, but you make a few errors:
>they don't even take the time to try and get wine compatibility
Do you realize what a support nightmare it would be? Codeweavers works with it. Regular Wine, however, is (no offense to the developers)a hack designed for users. It is not reliable. It's as silly as saying Nintendo should have designed its games with ROM emulator compatibility in mind.
>There is still no linux shockwave plug-in
There has been one (a proprietary one, not the GPL one) for Linux and Moz/NS compatible browsers since forever. Works fine for me.
>If that were to kill Adobe or Macromedia, then so be it. They have great products, but they refuse to support a community that has been clamoring for these programs for awhile.
Just because someone is clamoring for products does not mean that the company can fiscally and otherwise justify porting something. I'm sure the BeOS community would just looove a Photoshop port, but there are too few of them (us?). Linux is still not on a significant amount of desktops. It's a vicious circle--no famous proprietary apps-->no popularity-->no famous proprietary apps. Linux does need a killer app, one that not only performs better (I could say so for much of my Linux software) but has PR appeal. Something of the GIMP's scale but not a replacement for a proprietary product.
There is a very important point that must be made. Linux has delivered far more quality to the market per software development dollar than any Windows OS.
The fact that so many companies are adopting and using Linux is due to the fact that Linux is a high-quality OS.
Look at the recent big machine benchmarks. There are machines near the top of the list that run Linux. These are machines that are run mission-critical software and you find many of them are running Linux.
A critical factor in the delivery of quality in open source / free software development is that money can be directly targeted on improvements. If my company needs fixes in the USB driver, I can hire someone and have them fix it. There is no such thing available in the commercial software / closed source market. If I need new functionality, I can spend the money and get it built. Again, commercial software / closed source cannot deliver.
The point here is that open source / free software delivers quality software at a much lower cost than commercial software. It also enables the specific evolution of that software in a way that fits the agile enterprises of today.
Microsoft cannot deliver similar benefits to the market. When you go down the Microsoft OS road, you are beholden to Microsoft for all your bug fixes, patches, and improvements. You had better hope they care enough about the welfare of your company. And I hope you are not a competitive ISV as Microsoft is notorious for not fixing the bugs that are preventing their competition from using Windows effectively.
Great, your Windows 2003 setup is working, you like the look and feel, but really, it's not like I care. You just can't customize everything. And why you installed Office on a server?
The thing that I hate the most about Windows and Microsoft products in general is not that is crashes, or the software sucks, etc. but it's the unability to really customize how you want it. That's the beauty of free software.
Of course, newbies doesn't care, but I don't care about them either.
> The GPL, which I'm assuming your referring to, is not
> about freedom at all. It's about ensuring that you get
> something back for the work you've done.
You're wrong. The GPL is about ensuring users' freedoms. It does not require that you publish modifications. It does not ensure that you get something back for the work you've done.
> GPL software on the other hand takes the following
> attitude. I don't particularly care about money, but I
> care about code.
Again, you're wrong. As a free software developer, I care about the freedoms of the users of my software.
> If you make changes to my free code, I
> want them. An important point to make is that the owner
> here is not an individual, but a nonexistent entity. None
> the less, it is all about returns, not freedom.
It's all about the users' freedoms.
> If you really wanted to be a nice person, you would
> release code without expecting ANYTHING in return. This
> would be FREE software.
Then I am a nice person. Thanks. :-) Also, the GPL is a FREE software license.
> I would have the freedome to do anything I want with it.
You should not have the freedom to restrict the freedoms of others.
> The same cannot be said of GPL software.
The GPL protects the freedoms of ALL users.
Actually, you're not entirely right either. The GPL is about protecting the code. The entire GNU philosophy is that code itself should be free. Thus to protect its freedom, the GPL places certain restrictions on its used in non-Free products.
If it was about the rights of the users of the code, it would be more like the BSD license: give credit, and do what you please with it.
It's all subtle differences, but they're still there.
> They are using different code and architectures.
I thought that too until I stumble with a "Incorrect DOS function" ~1.5 years ago when I was copying contents from a CD.
Maybe it has a different architecture, but it has a lot of legacy code too.
I think of Win3.x (and bellow) as a different of Win9x/ME series in terms of archtecture. Win3.x --> Win9x/ME -> WinNT/2K/2K3 are, for many reasons, steps in the Windows foundation.
Win3.x was just a shell over DOS (16bits) . Win9x /ME were mixed 16/32bits. Windows NT/2K/2K3 are indeed 32 (64 for Win2K3 64) bits architecture.
If you ask me I tell you that Windows has 20 years.
But then you could argue that Linux has 30+ years on it's foundations... 
As for "Is Free Software Always a Good Thing?".
No, of course not. Both commercial and free (as in beer, as in chocolate, whatever) are needed.
There are truly amazing software/tecnologies in both sides of the fence.
For example, IMHO:
FS Commercial
Apache
Mozilla
Eclipse,NetBeans Websphere Studio
PostScript PDF
OpenGL and derivatives (this fits both)
Photoshop
Rational Rose
BSD, Linux Solaris, OS X
yacc (bison), lex (flex)
LaTeX
MySQL,PosgreSQL Oracle DB*,DB2
Java (J2SE, J2EE, J2ME)
Quake, Unreal
X11
and much other that I can't remember now.
* Don't ask me about their other products...
Dual license ONLY works if you wrote every single line of the source code. If you start your project from other people's GPL source code, then you CAN'T dual license.
>>>There are definitely times and places where you want code to be freely available. Think of a publically funded medical expert system, for instance. Free software is appropriate for those projects that have benefit to the community and need safety measures against commercial hijacking. Or all the tools you need to build and maintain such a system. It makes sense to have those tools available to the people who funded their creation so no community, society, or country is unable to access better medical care.
It's YOUR tax dollar. Would you be willing to fund a public program that other countries can steal for free ---- and because of that ---- people in other countries (where no public funds were spent) paid lower taxes than you.
Safety has nothing to do with open/close source code. Both are just as buggy (Microsoft ISN'T the whole close source industry). It's actually much costlier if you have to certify both your x-ray machine software application and also the open source OS. It already takes years to get safety certifications when QNX/VxWorks/Lynx certified their OS so you only have to certify your app.
>>>>Microsoft is funded by years of illegal monopoly money, so there is no comparing what Microsoft can do vs. what the Linux community can do. I will say it again, if you gave Linux billions of dollars in funding and four years (for just one major revision) of a zero market pressure environment, I'm sure you could turn out an amazing OS that would make Windows 2003 look shabby by comparison.
Linux is funded by IBM (who had a anti-trust consent decree with the US government). Just because Microsoft is the only one convicted --- doesn't mean that everybody else (who cuts a deal with the government and avoided a conviction) are less guilty.
>>>Take the example of QNX which is privately funded small company that makes a well-regarded OS product. You will find many versions of their OS had problems with the mouse pointer too. By the way, this was well before any cheap/free versions of their OS. There are many many examples of commercial products which are very poor quality.
That's a bad example. For one, many QNX systems doesn't even have a desktop. Secondly, you don't change mouses or video cards once you deploy a qnx-based nuclear power station. This has NOTHING to do with quality, it has EVERYTHING to do with QNX customers (your average nuclear power station owner) not willing to pay extra money for a better mouse driver.
I think the original poster has something of a point. Microsoft has a huge head start in this arena. WinNT was started in October 1996, and coding didn't really get under way until well into 1997. So NT has a six year head start. KDE has hundreds of developers. Microsoft has thousands. The KDE architecture had to be designed from scratch, while the Win32 API was an evolution of the pre-existing Win16 API. In that time, the KDE folks have been able to come up with a product that, on a technological level, is competitive with NT, and even has an edge in many cases. That is an incredible achievement, and speaks volumes about the development model. The KDE folks have at least two years until Longhorn comes out. To put that into perspective, that is how long it took them to do a rewrite for the 1.0 -> 2.0 transition. Who wants to be they can clean up any usability issues in that time frame?
PS> I think the KDE development model in particular has a few important advantages: first, architectural issues are handled by a core few people, not really random developers. Just take a look at the API. It's very clean and integrated, hardly what you'd expect from an anarchic development model. Second, it has a much stronger community than closed source development. Just take a look at KDE programs in general. The strong community encourages developers to do everything "the KDE way." As a result, the level of integration between KDE applications far exceeds that in any other desktop. I don't know if this strength will hold as the KDE developer community gets larger, but it's a nice change from the MS world, where even Office and Internet Explorer use different toolkits.
"It is just that there aren't many companies making their software available to purchase on linux."
-MS holds over 90% of the market, Linux just about 1%. It's just not worthwhile for most companies to invest into the Linux market.
Unless Linux doesn't gain more market shares it won't change; unfortunatly.
It's about freedom and protecting code.
It's about freedom since you give the same rights over the code to your customers/end users.
It's about protecting your code since it won't allow anyone to "steal it" (read claiming that is it exclusively his/hers own).
It's about cooperation since if someone finds another use for your code (s)he has the moral and legal duty to make the changes availlable to you.
It's, definitely, NOT about making money with your code.
>> You can bitch at Win9x and Win3.x if you want,
Actually, I wansn't bitching at all, just pointing out even Microsoft needed time to roll out a good product.
>> but don't put all Windows in the same bag just because >> >> they share the same name. They are using different code and architectures.
Everybody knows that, Eugenia, thank you very much. So what are you saying anyway? That ms developed the other windows from 1982 to Windows Me -18 years - and still didn't make a good product out of it? What does that tell you?
You know what though? No matter how much you spin it, the NT line is still, in a very important sense, a continuation of MS's windows efforts. They leveraged their past windows efforts, and experience, and EVEN code, to build the NT line. So stop pretending like the NT line was a micracle coming completely out of the blue.
And btw, as far as servers go, Win2k was still NOT an exceptional product. It was only good to the extent that it was more stable than previous ms products - in order words, it was good by ms standards. As long as I am not compelled to run sql server or iis, I would, if I was making a technical decision, still choose BSD or linux over win2k.
Why people were/are angry at Redhat is because they changed a lot more than the theme and a few defaults.
check out more here: http://mosfet.org/noredhat.html (Yes, he missed a lot of hamrful changes to KDE, but is hould give you an overview.)
Also, i do agree that OSS neds to become more interoperable and integrated, it just needs more coordanation and team work for a common cause like taking more than 1% of the desktop markey and increasing the server market even more. Instead of havinga billion tiny applications that are intended to do the same thing but have a few differences, jsut make only 2-3 reallly gooooood ones. Competition is good, but waht drives the market the most at improvement is 2-3 companies making the same product, when it gets toof ragmented everything starts going downhill and it becomes a ngihtmare for th user too. (UNIX)
I think when Stallman stated that the freedom to choose any license was not the sort of freedom he advocated was when he lost the last of his cred for me. The simple fact is that the marketplace has been around since the first humans began trading in earnest, and, with good regulation, it works. Don't underestimate the willingness of people to pay money for something that will be valuable to them. Hence people pay for QT licenses etc.. I agree that in many cases software costs were way out of reality, but we are seeing the software market become saner and stabler and the hype die down now.
I love free software, in the license sense, and I think it has already achieved it's goal. There is no stopping the momentum it has already built. To rail against the freedom of individuals or companies in this day, when there are in fact often open source alternatives to most proprietary solutions, to choose any license is silly and wrong.
There is a even a growing symbiotic relationship between commercial companies and open source projects these days.
The bottom line is that not every itch that businesses and consumers desire is going to met by an itch that developers want to scratch. At the same time, the open source development model can't be expected to be expected to fulfill every niche in the market. In other words, giving away your source, in an already small market for some software might be akin to giving away the whole market.
I think open source can be very good at providing a computing infrastructure(ie an os, database, networking technogies, etc) but often stops short of giving a polished "simplified" package to the user.
It is a shame that more closed source consumer software companies(ie. Intuit, Macromedia, Adobe, etc.) have not commited to devloping software on linux.
>check out more here: http://mosfet.org/noredhat.html
Oh, please. "Harmful" to KDE. Right... 
I have KDE on my RH 9 and it works perfectly and QT apps compile perfectly too. I have absolutely no problems with KDE under RH, and I don't believe all this propaganda against RH Mosfet and his friends are brewing.
Ian Pulsford wrote:
I think when Stallman stated that the freedom to choose any license was not the sort of freedom he advocated was when he lost the last of his cred for me.
Do you have a reference for that quote? Doesn't sound like the typical razor sharp writing/speaking about free software I've heard from him. Besides, I'll bet he doesn't support the freedom to rob banks either.
For those who haven't read the GPL,
http://www.gnu.org/licenses/gpl.html
There's more about what "free software" means
http://www.gnu.org/philosophy/free-sw.html
as well as a great faq
http://www.gnu.org/licenses/gpl-faq.html
and lots of other good reading.
Of course, when you're done, don't forget to take the quiz:
http://www.gnu.org/cgi-bin/license-quiz.cgi
:)
Not necesserilly in price, but the user should still have freedom to modify the product for his or her needs.
Why? If I spend 6 months working on a program (that included nobody else's licensed code) and decide to release it under whatever license I see fit, you as an end user don't have the freedom to do shit with my code unless I say you do.
just like if I buy a couch or whatever, i can do absolutely wahtever I want to it/with it, same should go for software.
Of course, software is not the same thing as a physical object such as a couch, which many people will attest to when they want to justify piracy. Just look at it this way - if you could make 1,000 copies of that couch in about a half a minute and then give those copies to all of your friends, do you think it would be legal to do so? Just because it can be done with software doesn't mean that it should be done with software. (And not saying it shouldn't either, but that's not for the end user to decide.)
If someone wants to write an app and hand it out for free along with the source code, that's fine if they choose to do so - a lot of great programs fall under this category. But what these 'free software or death' people need to do is to stop expecting something for free that they did not have a hand in creating. If a piece of software has a license that's too restrictive for your needs, don't use/buy it. Either build it yourself or use something else. And if there's nothing else, well .... that's just too damn bad, isn't it? If you're not willing to do the work involved in building what you want on your own, then why do you expect somebody else who wishes to keep their code under wraps to just hand you the source code and have at it?
Flaws.
"Linux users don't pay for software."
Because right now they're more sophisticated. Unsophisticated users need to buy software -- users that right now use Windows. If they used Linux, they'd be buying software because they can't build it. They'd even have more money for it.
"community backlash can best be described as merciless."
Don't customers ever backlash? Don't employees ever fight management? Isn't Redhat dependent on that very community to make its business model feasible -- low startup costs because they don't employ most of their developers? Who is doing whom a favor here?
The thesis -- free software doesn't pay salaries.
Absent proprietary software, wouldn't customers still pay money for programmers to fulfill their needs? Adam Scheinberg is assuming many things about a completely different software universe that has never yet existed. The financial rules won't be the same, and he needs to think about them.
>>The thesis -- free software doesn't pay salaries.<<
The opposite assumption is -- free software save salaries.
Thus, the new thesis -- non free software waste salaries.
>>If you really wanted to be a nice person, you would release code without expecting ANYTHING in return. This would be FREE software. I would have the freedome to do anything I want with it.<<
OK, now this is what I can do with your code: Make it locked and propietary and charge you for that. Now I make you a slave to make it better. Do you realize that you are a nice slave by now?
I just wanted to get this said once and for all ... comparing Microsoft to Linux in order to compare the quality or usefullness of open source vs. commercial products is not the best and most direct comparison you can make.
The more direct comparisons to be made would either be between Unix and Linux, because they attempt to support the same theory (alhtough linux was originally meant to be a personal desktop OS, and no other unix was - so just like all analogies or comparisons it breaks down at some level).
Another valid comparison would be between Linux and BeOS, because both we're created at aproximately the same time, and both we're meant to be new personal OSes.
Don't get me wrong, there are dozens of usefull connections and comparisons to make, between all manner of licenses, development models, OSes, langauges, etc ... and definately the Windows vs. Linux issue has a place ... but it is constantly talked about primarily for one reason only - popularity. Windows is the most used and popular OS, and Microsoft the largest company. Linux is the most used and popular open source operating system, and Linus Torvalds and Richard Stallman are the most recognized personalities (and I find people who connect them too much even more clueless than people who can only compare linux to windows).
I frequently find myself perplexed by the Linux community. I love the software and I enjoy using the alternative environment for it's benefits and my genuine interest in the technology. However, I never have been able to fully understand the zealous of "open-source." As a college student preparing for a career in programming, I do not see GPL as anything close to a viable career choice. I like free software, and I like community-driven free software, however, I think they are and will remain hobbies. I enjoy programming, that's why I want it as a CAREER not a hobby. About the only useful benefit of opensource over freeware closed-source to me personally is the compiling optimizations of Gentoo
Didn't they debunk the whole "Free software isn't free!" meme? Are free countries less free when they outlaw murder?
Do we need these wordgames to keep us from important things?
Yes, I've read most of Stallman's rhetoric.
Here's what I was referring to:
'However, one so-called freedom that we do not advocate is the "freedom to choose any license you want for software you write". We reject this because it is really a form of power, not a freedom.'
from: http://www.gnu.org/philosophy/freedom-or-power.html
There has been one (a proprietary one, not the GPL one) for Linux and Moz/NS compatible browsers since forever. Works fine for me.
I am afraid you are thinking of Flash. Shockwave is as of yet unsupported. Believe me, I wish it were, my little bro loves the Nick Jr. website but can't use it on my computer due to most of the games being shockwave as opposed to flash.
Do you realize what a support nightmare it would be? Codeweavers works with it. Regular Wine, however, is (no offense to the developers)a hack designed for users.
I am not saying it should be officially supported. But they could throw the wine guys a bone with some hints to making it work and a little effort. Also Codeweavers is based on wine with some proprietary stuff thrown in. It isn't that much of a stretch that certain things that work in codeweavers could be made to work on wine if some companies who owned the ip would give a hand.
The opposite assumption is -- free software save salaries.
Thus, the new thesis -- non free software waste salaries.
How do you figure this?
Let's say I want to write a program that would be the Mac software equivalent to Wine - to be able to run OSX apps under Linux and/or Windows without the need for OSX itself, and have it as close to 100% compatibability as possible. (I'm not sure if anyone would really want to do that, but just go along with it for the sake of discussion ...)
Now, I don't really care about money and would love to make this free software, but I have a problem ... this is not the kind of project you can sit down and write in your spare time and expect to be done in less than 10 years .. especially if you have a wife and kids The most you can hope for is that you would release some code and maybe enough developers would jump on board to help so that you would be done in the next millenium. But considering how long it has taken Wine to get this far, I don't really think I want to go that route.
So, at the very least, I need to do this full time, and probably have a lot of other paid developers helping me, and create a small company. However, to be able to do this, I need money. So I go out and find a couple of investors and hire the developers that I need. Two years later, the project is done (v1.0) and ready for release. Now, I can release this program under the GPL and let the community make bug fixes and/or upgrades for me and be able to do with it what they wish (free software), but I've got a problem ... in developing this software, I've got a lot of debts to pay off. So, how the hell am I supposed to do that by releasing/selling the software under a license that allows people to do whatever they want with the code and give it away for free? And moreover, why do you think that end users should have the right to do this after my company has spent so much time and money developing this thing? Now, I am probably missing something, but I don't see how the hell this is supposed to work when people need/want to make money?
..so I have less to bitch about
Interesting article, and it brings up some good points -- but it also gets a few wrong.
1) We're still missing even passable counterparts for Photoshop
Wrong! GIMP. It is most definitely more than "passable", by anyone's definition. (Whether it's as good or better than Photoshop is another matter entirely)
2) Linux users don't pay for software.
Other postings have refuted this point well enough. I'd be interested in seeing supporting documentation for this statement, btw.
3) And when we implement it, there's no support.
Um... have you ever called MS tech support? Q.E.D.
MS holds over 90% of the market, Linux just about 1%. It's just not worthwhile for most companies to invest into the Linux market.
You ignore many important facts in this comment.
1) Porting from OSX to linux is generally considered a non-ordeal. I don't know how true this is, but I'll have to go by what people say.
2) Porting between linux and all traditional unices is also simple as long as traditional api calls are used.
3) Linux is an emerging market. There is a saying in the world of business: "The bigger the risk, the bigger the payoff." Linux may be a big risk (I don't believe it is, I think it is an obvious winner), but even if it is a big risk it has a chance of a big payoff. I think it is a small risk with a big payoff. The only reason the risk is considered big is because most people don't see how obvious it's success is.
Q.E.D.? Jeez, GoodGrief, could you get any more arrogant?
The GIMP is wonderful. But a Photoshop counterpart it ain't. Ask a real graphic designer, they'll tell you. People who say the GIMP is a professional design tool are generally the ones who only use it to convert those pesky png files into the "more manageable" .bmp format.
As for the paying for software thing - public companies exist to make their stockholders happy and make money. If they thought they could sell, say, Photoshop or AfterEffects or Flash on Linux, they'd do it. They obviously don't.
3) And when we implement it, there's no support.
Um... have you ever called MS tech support?
Yes, have you? They're not God, but they stand by their product and they've sent people out to help us firefight real problems. Anyway, Microsoft isn't the only company that markets their support. Ever called Novell for support? That's worth every penny.
Support, too often, for better or worse, is about blame. Microsoft is willing to stand by their products, even when the shit hits the fan. Open source developers, for better or for worse, usually disclaim themselves from any responsibility in the README file.
Don't cast me as anti-open source, just don't try to sell me with weak, baseless (and incredibly pompous) arguments.
"Porting from OSX to linux is generally considered a non-ordeal... Porting between linux and all traditional unices
is also simple as long as traditional api calls are used."
> I can't judge nor argue it.
"Linux is an emerging market. There is a saying in the world of business: "The bigger the risk, the bigger the payoff." Linux may be a big risk (I don't believe it is, I think it is an obvious winner), but even if it is a big risk it has a chance of a big payoff. I think it is a small risk with a big payoff. The only reason the risk is considered big is because most people don't see how obvious it's success is."
> Let's hope that this is enough of an incentive for developers to invest in Linux. But at the Moment most companies are reluctant to do so.
"Open source developers, for better or for worse, usually disclaim themselves from any responsibility in the README file."
MSFT always disclaims itself from any responsibility in the EULA.
"Microsoft is willing to stand by their products, even when the shit hits the fan."
Redhat employs Alan Cox so he can debug hard problems for their support clients.
"If they thought they could sell, say, Photoshop or AfterEffects or Flash on Linux, they'd do it."
QNX is profitable. Does Adobe sell Photoshop for it? No. Now, if Windows and MacOS were gone, and the only serious desktop were Gnu/Linux, would Adobe sell Photoshop for it? Yes.
btw, the post you're responding to said that Gimp is merely "passable." In no way did he say it was a professional tool.
There are certain reasons why software companies need to make decent amounts of money, and why a GNU world is dangerous. For example, Microsoft keeps hardware manufacturers in check, which probably is a very good thing. And the Microsoft tax is very low.
However, I don't think this article captures these points.
btw, the post you're responding to said that Gimp is merely "passable." In no way did he say it was a professional tool.
He said (indirectly) that Gimp was a passable alternative for Photoshop and since Photoshop is a professional tool, what does that lead you to conclude ?
>Porting from OSX to linux is generally considered a non-ordeal.
Only if you are talking about POSIX command line applications. If you are talking about GUI apps, it is almost impossible to port anything from OSX (just very dificult), as most apps are based on Carbon and Linux has nothing like it, while Cocoa is very poorly supported by Gnustep (plus most people don't have gnustep on their linuxes, therefore devs don't prefer it). And even for the first team, the command line apps, usually is the other way around: people are porting from linux and unix to OSX, not the other way around, in general.
>Porting from OSX to linux is generally considered a non-ordeal.
Eugenia, above claim was from dwilson. Anyway, thanks for your explanation. All I was saying is that S/W developers couldn't justify the costs of porting win aps to linux because there is hardly any demand for it in the corporate world.
Well, Good Grief wrote:
It is most definitely more than "passable", by anyone's definition. (Whether it's as good or better than Photoshop is another matter entirely)
While he wrote it's more than passable (my errer, I've been up too long), it seems he's casting doubt on it being up to Photoshop's standards.
Anyway, it's too late for me to care about trying to delve into the meaning of a someone's post, and I'll drop that point. I misread it too.
I only use Linux and have no interest at all using Microsoft products under any circumstances. This is something that will never change. I definately paid for my Redhat distribution and I would be willing to pay for other Linux services.
It's important to me that the source code is publically available because that means that changes to the platform have to be responsible to the public. I also have control over my applications because the vendor research and development is not locked up, instead I can control the libraries because I have the freedom to migrate or implement new behavior. I think that over time, open source development will become more generic, and provide powerful reuse and popular simplification so that source code accessibility can be leveraged by all Linux users.
Information technology is so easy to abuse. It's very easy for a vendor to slip handcuffs onto you without you realizing it until you are fully locked in. These issues require extensive background information, and they are not understood by the public. Information technology is a glutton for abuse, and open source is the way to educate the public and provide them with control, power, and best of all, freedom.
In the free software world, people don't pay for software. People pay for service, people donate to their favorite projects. Its a revolution and most people are still under shock. But I can assure you in a few years, most non specialist software will be free (as in beer), but you'd be able to pay and get your money's worth of good support for it. Also, custom software would become the order of the day, and that is what people would pay for --- to get software customized for their needs.
The sad fact is that a group of 5 or 6 high school students who probably wear their grandmothers panties over their heads can go into their attics and write software that kicks ass on any Microsoft product. Doesn't that tell you that Microsoft is just a little bit inefficient? I mean I wonder what really goes on there, probably a zoo of people pretending to be working hard.
It's not just in open-source projects that you have to wade through endless beta releases before a decent release. The quality of programs like Windows and Photoshop has been gradually improved over decades. I doubt anyone would say that Win95 was a quality product.
At least with free software you don't have to pay to get the beta releases.
Darius,
your posts about hypothetical projects, and problems of people modifying your code and such make me want to hug you. Good to know there are some logical and sane people out there.
soup,
"In the free software world, people don't pay for software. People pay for service, people donate to their favorite projects. Its a revolution and most people are still under shock. But I can assure you in a few years, most non specialist software will be free (as in beer), but you'd be able to pay and get your money's worth of good support for it. Also, custom software would become the order of the day, and that is what people would pay for --- to get software customized for their needs."
You completely don't get it. Thats not the way things are going to go. Any business model like your will tank. Service as a addon to your revinue from sales is nice, but you can't live off it. If your product is truely good, you will have no service business, since people arn't needed to call tech support when something is good, or rather not nearly enough to make any money. Also, a few random people kicking 5-20 bucks to a company for writing something is not going to support them. Face it, what keaps software companies in business is selling software. When they give something out for free, they are doing it for a reason. Adobe gives away acrobat reader so they can sell people the software to make Pdf's, otherwise PDF's would have never taken off. You can think happy thoughts all you want about your world, but it isn't going to happen. The world runs on money. Companies live on selling goods. Service businesses only work when the whole basis is service. But if a company has to create product, service will not cut it.
Also custom software is the main way of things. I belive it's something like 90% or software written is custom apps for people, not off the shelf stuff. And you can be damn sure software that falls into this catagory is not free. Asking someone to write you and app and telling them you want it for free, but they can charge you whenever you have a question is going to get a door slammed in your face. If they did go for it you can pretty much garrentee one crappy confusing program.
Someone comes out with this unmitigated guff every month or so, and they're always wrong.
First, separate the issues. The development model is entirely irrelevant. Open source doesn't mean you have to *accept* patches or let anyone else play with your baby. A company can release its code as open source yet develop it exactly as they would a closed source product. Imagine if id software open sourced their engine as they wrote (they wouldn't of course, but bear with me). How would that affect its development? Not at all. Someone might take the code and make a fork, but the id product would be the same at the end.
Second, the point is flawed. It's not the open source model that makes desktop Linux a slightly less polished (although still immeasurably better - tried using a command prompt yet?) product than desktop Windows, it's simple immaturity and comparative lack of resources. Try using GNOME apps (NOT GTK) and ONLY GNOME apps for a day, or KDE apps (NOT QT) and only KDE apps for a day. Very close to the same level of uniformity and integration, and they'll both be at or past the level of Windows in two or three major versions' time. The CHOICE you get is just that: you can CHOOSE to use apps that aren't integrated with the DE of your choice, and most people do. What does this mean? Well, it means that to most Linux users, functionality is more important than consistent looks or "integration". But if you want those two, you can have them.
I agree there are benefits brought by a company developing software, namely money and centralised development. However this doesn't mean the software has to close-source, proprietry or not free from cost. Many companies do fine selling support or normally GPL'd products without the GPL.
In the long run nobody benefits from proprietry methods. Also in the long run Linux and other such open and non-commerial operating systems will be mature and overall perfectly adequete operating systems. At that point the impact of companies will be less important.
"Also custom software is the main way of things. I belive it's something like 90% or software written is custom apps for people, not off the shelf stuff. And you can be damn sure software that falls into this catagory is not free. Asking someone to write you and app and telling them you want it for free, but they can charge you whenever you have a question is going to get a door slammed in your face. If they did go for it you can pretty much garrentee one crappy confusing program."
Actually, since the GPL only makes you distribute your changes when you re-distribute, you're free to use whatever GPL code you want in your in-house project. Go nuts! You're not going to be redistributing it, so who cares?
it sounds like it's targeting zealots@osnews.com.
those guys really trolled you in the worst way if you wrote an entire article just to address them.
cause i know that article wasn't written for me or most of the technical people i interact with.
makes for some really lame reading.
basically the fight with the zealots has left the bars and has broken out into the streets.
albeit wrapped up in a nicely worded official article.
why not just rename the article: "ATTENTION LINUX ZEALOTS, BUY YOUR FSCKING SOFTWARE, THAT WILL FIX EVERYTHING CAUSE IT'S ALL YOUR FAULT"
sigh.
Call me picky, but there is no such web browser as 'Firebird' or 'Mozilla Firebird'. There is a 'Mozilla Firebird Project' that aims to produce the 'Mozilla Web Browser', but (and if you look on their page), it is clarrified, they do *not* publicly call their browser 'Firebird'. 'Firebird' is a relational Database based on the opensource Interbase 6.0 code. I honestly with Mozilla had had a little more sense when stepping on other people's IPR, but they didn;t. Could you please alter the article so that we all don't carry on the myth that is Firebird.
Perhaps the problem of Free Software desktop stuff being a lot of inconsistant jumbled crap has absolutely nothing to do with the concept of software being Free (as in speech or beer) and absolutely everything to do with the fact that the Free Software community (and for that matter, Open Source) is largely populated by traditionalist Unix geeks who have long had hostile or indifferent attitudes towards such things as GUI's, usability, and non-geeky people getting things done easily. Why do we keep expecting people who spend all of their time in a terminal and who goose step to the book-on-tape version of Neal Stevenson's "In the beginning there was command line" to produce usable software? Why do we keep expecting a next several years worth of quality after 30 years of nonsense?
How could we make Linux software more consistant? We could have folks just use one toolkit/desktop environment, and they could (horror of horrors) design the UI before they start writing the code (which is what any usability guy worth their salt would tell you to do to have a truly consistant, integrated experience). What would be the response of any Free Software (aka unix geek) to these suggestions? Utter disgust and denial. You'd get responses like "quit trying to enforce your proprietary software views on us" or "one toolkit? it's a matter of choice" or "that's what you want. That's not what I want" or "you obviously don't understand the open source method". Yes, I've been personally told all those things a number of times by a number of different Free Software developers.
Apache and the Linux kernel have succeeded independantly of commercial involvement, so if such polished technical projects that require a massive amount of resources and coordination succeed, we have to start asking the question "Does Free Software not produce usable, highly polished desktop projects that require lots of resources and coordination due to a lack of money, or is it because the Free Software developers really don't consider that area to be very important?" Furthermore, it begs the question as to whether bringing in commercial companies is truly the panacea for the many ills of desktop linux, since the real problem has nothing to do with money and everything to do with lack of developer interest in solving the problem.
In my opinion (which is just my opinion, any may not be valid in all 50 states and is subject to recall at any time), if the you have to bring in commercial companies to save your ass from the apathy of your own developers, you never really had an ass worth saving.
That being said, I'm all for the success of Free Software; once the harmful unix culture has been exorcised from Free Software, things will get better.
"Linux users don't want to pay for anything"
This statement is so wrong... There was _never_ any statistical proof for this, this rumor just came up because it made sense back then but that doesn't make it true. In fact, every informed statement about this seems to imply the opposite, that Linux users pay just as much for worthy software than anyone else (of course there are many Linux users not willing to pay for anything, but so are many Windows users who pirate 99% of their software!).
But that doesn't mean that I will pay for software if I don't think it's worthy. This includes most proprietary software because it usually lacks integration, gets lost for me when I switch architectures (because it lacks portability) or even stops getting developed all together (BeOS comes to mind or this expensive text editor I buyed for Windows which still suffers from very annoying bugs which were never fixed up 'til now).
Please spare me all this "proprietary software is good for you, you should spend money on it, even if it sucks", I just can't hear it anymore. I'll pay for whatever I feel is worth paying for and whatever I can effort. When I don't really need proprietory software enough to pay for it, it's probably just not good enough, won't you agree?
Confusing free software with uncommercial software won't impress me neither, I thought we would be over this by now.
To sum it up, this article is basically just a summary of FUD and redundancies which are constantly spread against free software. Again, no real substance there. Maybe you'll call me an idiot now for not acknowledging your wisdom, but I'm growing tired of it.
There are times that I need to use use Microsoft products. For example, while I dislike it tremendously, Word is the lingua franca of word processing. I know I can send the vast majority of my business contacts a Word document. They likewise expect that I can open/view/manipulate Word docs, as well. Is this bad? I dunno... Is it reality? Definitely.
There are however times when it absolutely a no brainer for me to use Linux and other OSS software. I find it personally offensive that I have to buy licenses for *BOTH* my file/print services (once for the server and then CALs - and those add up quickly!). Samba not only gives me finer grain control, but I don't have to worry about hidden costs or reliability. It just works!(tm)
I find it unfortunate that people will try judge Linux based on what is (or is not) available? I remember when Adobe made a beta of Framemaker available to the Linux community. It was a flop. So what? I do remember people saying things like "Adobe's giving up on Linux." So what? If you can afford Framemaker's cost (big $$$) and need its power, then neither Linux nor Windows is really a good choice. You need dedicated horsepower. And that's been Adobe's thrust with it all along. Believe me, it eats resources like candy.
Not withstanding the MS propaganda, sever commercial entities have swung support over to Linux. SAP has recorded some of their best performance numbers on x86 Linux. (They have a really nice Open Source DB, btw. Check it out at http://www.sapdb.org good stuff even if nobody knows about it.)
I really do think everyone worries/evangilizes/complains about the Linux/Windows GUI too much. I started with a CLI. I will probably always stay with the CLI. My preference. Have a sseen mouse slowdowns with X? Definitely. Have I seen the same slowdowns in Windows? Yes from good Windows 286 (boy that dates me) all the way through XP. What is remarkable is that everyone seems to be able to criticize the other guy's OS while ignoring or downplaing their own OS's weak point(s).
I have and will always for things that work. I have no real ethical paying for Microsoft products when they meet a need for me. I likewise have paid for all my Linux distros and have donated money to support certain OSS projects. I refuse to support crap, either monetarily or in actual use. BTW, as far as Linux and MS support goes I have had successes and failures with both. This is why I will never use Exchange (paying $299 to discover my problem is a known "issue") nor sendmail (I like human readable configuration files, thank you very much).
One final thought, a friend gave me an interesting perspective on the whole MS vs. Everything Else debate. Microsoft would still be an also ran if the original TCP/Ip code wasn't BSD licensed. IE wouldn't exist if Mosaic had not been developed. Likewise, Apache devlopment has been spurned on by IIS. MySQL (or Postgres) would still be minor or research RDBMs if there weren't people looking for cheaper, (and in a lot cases) more reliable alternatives to SQL Server or Access. (One interesting tidbit here, I have come across many customer who were told by so-called MS experts that their websites were SQL Server driven only for me to discover an Access MDB as their backend. Ahh the power of perception).
In short, for me it's the right tool for the right job. Sometimes it's Linux. Sometimes it's Windows. And even sometimes it's something else.
-d.