Man, did anybody actually read the article? What he was saying is absolutely right. The value in F/OSS is more the community of developers than the software itself. In other words the collaboration itself is the value. He hit the nail right on the head. Don't sell software; empower companies to work together to achieve their common goals. Everybody chips in to buy the pizza and everybody gets a slice (except in this case the slice is actually a whole pizza because there is no scarcity involved).
That you would buy over and over again with each new upgrade. Software is not a fixed cost, it is always a recurring cost no matter how you look at it. An additional, often overlooked, cost of proprietary software is having to mold your workflow to match their model (using F/OSS and some dev time you can guide the project in the direction you need). You can start a project with a well organized website stating project goals, and let people build from there. State that developers are needed, offer rewards for certain features. I might suggest seeding the project with a small useful core (pay a dev for that). And just try to grow it from there. Make it a pet project.
You also don't get to use the GPL'd code because of that policy. The people who made the GPL'd code are not the only ones losing because of a bad company policy.
Well, that and you don't have the freedom to fork it. Essentially, leaving you to whatever their whims may be. F/OSS works because its far more of a survival of the fittest type situation. With F/OSS if the software no longer works for you, you are free to fork it.
With the install base it has, I really don't see OpenGL going anywhere. In the mobile field its pretty much THE standard. As far as features go everything is supported through extensions correct? Sometimes even before DirectX gets support? As great as it is to declare a proprietary single-platform piece of claptrap the defacto 3D standard, it would be doing a bunch of other great platforms a severe injustice. While I agree OpenGL is not great, it is also the best cross-platform solution there is. And despite what some people want to believe, Windows is not the world. BTW, why not just use something like Open Scene Graph or Irrlicht if OGL is too low level.
I see no proof that Windows threads are any faster than Pthreads. Process forking obviously goes to Unix since it does light weight copy-on-write forks.
DX10 is a comparable API to OpenGL C# is harder to work with than Python (Java is great and all, but there are many other languages. And before you say it, yes I've heard of IronPython, ever heard of Mono?) You are probably right on that last one. They all keep how shit works on the game consoles so close to their chest that it makes it pretty hard for me to know.
Even if they develop a highly parallel programming language, that doesn't mean that everybody who can program C# will be able to use it. Hell, if you want a functional, cross-platform, highly scalable, fault tolerant, naturally paralellizable+clusterable programming language that is available now for free just use Erlang (like everybody else with a clue). Its already written. And yes you can run it on your HPC clusters if you want, but why would you want to?
Yes, because people fighting for software which can be used however one sees fit is totally a problem. Making software work "for you" not "despite you" is totally a problem.
The problem definitely isn't with a piece of useful proprietary software becoming so large and all engulfing that no competition can usefully coexist. Thereby, creating a stagnant monocultured environment with a money swollen corporation killing all forms of competition using underhanded tactics because it has the power to do so. No, that totally isn't characteristic of any kind of forced servitude.
You know slaves had an opportunity to runaway also. But the environment they were in made doing so suicidal.
Stallman has excellent ethical direction. Unfortunately, there is a large proportion of "software houses" that don't (ignoring the fact that open source does not dictate how you sell your dev time). But the ethical and financial benefits of delegating the direction of a software product to the entire human community is huge.
I speak from my semi-limited experience having delegated some of my (paid) time to a couple of F/OSS projects. Specifically, I benefit from getting a nice pre-existing codebase for nothing (inkscape/gphoto2). The community benefits because they got my changes for free. Additionally, I authored a widget for wxPython, posting it to their wiki asking for some help to reach the stated goals. It was useful to enough people that eventually my little widget was flexibly extended to the point of encompassing all my goals and then many more. A culture of cooperative development had pushed my meager contribution into something many people wanted and could use. This was the ultimate testament, to me, of the efficacy of free software.
Slashdot Mirror
Porn.
I'm already gainfully emloyeed thanks. Why did you use vowels for the words 'you' and 'a'
Man, did anybody actually read the article? What he was saying is absolutely right. The value in F/OSS is more the community of developers than the software itself. In other words the collaboration itself is the value. He hit the nail right on the head. Don't sell software; empower companies to work together to achieve their common goals. Everybody chips in to buy the pizza and everybody gets a slice (except in this case the slice is actually a whole pizza because there is no scarcity involved).
Wouldn't it be even more wild if the message was: "You die now!!"
I vote for butt rape filesystem.
That you would buy over and over again with each new upgrade. Software is not a fixed cost, it is always a recurring cost no matter how you look at it. An additional, often overlooked, cost of proprietary software is having to mold your workflow to match their model (using F/OSS and some dev time you can guide the project in the direction you need). You can start a project with a well organized website stating project goals, and let people build from there. State that developers are needed, offer rewards for certain features. I might suggest seeding the project with a small useful core (pay a dev for that). And just try to grow it from there. Make it a pet project.
You also don't get to use the GPL'd code because of that policy. The people who made the GPL'd code are not the only ones losing because of a bad company policy.
Well, that and you don't have the freedom to fork it. Essentially, leaving you to whatever their whims may be. F/OSS works because its far more of a survival of the fittest type situation. With F/OSS if the software no longer works for you, you are free to fork it.
With the install base it has, I really don't see OpenGL going anywhere. In the mobile field its pretty much THE standard. As far as features go everything is supported through extensions correct? Sometimes even before DirectX gets support? As great as it is to declare a proprietary single-platform piece of claptrap the defacto 3D standard, it would be doing a bunch of other great platforms a severe injustice. While I agree OpenGL is not great, it is also the best cross-platform solution there is. And despite what some people want to believe, Windows is not the world. BTW, why not just use something like Open Scene Graph or Irrlicht if OGL is too low level.
What app did you develop?
The fact that I can't use Sun's software how I want is a problem with GPL how?
D3D is better than OpenGL how? OpenGL supports all features that Direct 3D does.
I see no proof that Windows threads are any faster than Pthreads. Process forking obviously goes to Unix since it does light weight copy-on-write forks.
The toolsets are already far superior on Linux. And your mom is fat. ;-)
DX10 is a comparable API to OpenGL
C# is harder to work with than Python (Java is great and all, but there are many other languages. And before you say it, yes I've heard of IronPython, ever heard of Mono?)
You are probably right on that last one. They all keep how shit works on the game consoles so close to their chest that it makes it pretty hard for me to know.
Even if they develop a highly parallel programming language, that doesn't mean that everybody who can program C# will be able to use it. Hell, if you want a functional, cross-platform, highly scalable, fault tolerant, naturally paralellizable+clusterable programming language that is available now for free just use Erlang (like everybody else with a clue). Its already written. And yes you can run it on your HPC clusters if you want, but why would you want to?
Honestly, why would anyone want to roll-out something like this on Windows. A lot of extra expense for little practical value.
Yeah, but while you guys were talking about that, I trademarked Subcasting (TM).
Yes, because people fighting for software which can be used however one sees fit is totally a problem. Making software work "for you" not "despite you" is totally a problem.
The problem definitely isn't with a piece of useful proprietary software becoming so large and all engulfing that no competition can usefully coexist. Thereby, creating a stagnant monocultured environment with a money swollen corporation killing all forms of competition using underhanded tactics because it has the power to do so. No, that totally isn't characteristic of any kind of forced servitude.
You know slaves had an opportunity to runaway also. But the environment they were in made doing so suicidal.
Stallman has excellent ethical direction. Unfortunately, there is a large proportion of "software houses" that don't (ignoring the fact that open source does not dictate how you sell your dev time). But the ethical and financial benefits of delegating the direction of a software product to the entire human community is huge.
I speak from my semi-limited experience having delegated some of my (paid) time to a couple of F/OSS projects. Specifically, I benefit from getting a nice pre-existing codebase for nothing (inkscape/gphoto2). The community benefits because they got my changes for free. Additionally, I authored a widget for wxPython, posting it to their wiki asking for some help to reach the stated goals. It was useful to enough people that eventually my little widget was flexibly extended to the point of encompassing all my goals and then many more. A culture of cooperative development had pushed my meager contribution into something many people wanted and could use. This was the ultimate testament, to me, of the efficacy of free software.
H.264 is a more advanced variant of MPEG-4.
You understand that you are advocating less freedom for no reason right?
Well, since a lot of closed source drivers have redistribution issues, I would say that makes it more difficult to use with clusters of computers.
You can use the SPEs from within Linux. You can even program for them using free software tools.
As long as they contribute back, then its not hurting anybody.
Doesn't Sphinx do this? http://cmusphinx.sourceforge.net/html/cmusphinx.php