| |||||||||||||
Interview with Andrew Tanenbaum (LinuxFr.org)Interview with Andrew Tanenbaum (LinuxFr.org)Posted Nov 17, 2011 23:17 UTC (Thu) by chuckles (guest, #41964)Parent article: Interview with Andrew Tanenbaum (LinuxFr.org)
Is it just me or does he sound bitter?
After being very active in the linux community for +10 yrs, I think the reason it's doing so well is _exactly_ because Linus choose the GPL instead of making it BSD. I think the BSD license allowed, almost encouraged, the fragmentation that ended up killing the big UNIXs.*
"I don't think of 5% as that big a success story."
That's on the desktop. Linux pretty much owns the HPC market, has a LARGE chunk of the embedded market, a LARGE chunk of the server market, more than a few stock markets use linux as a trading platform, ...
'No, Linux "succeeded" because ...'
"In fact, one of the reasons we are going into the embedded world is that there is no dominant player there."
"What most ordinary users want is that it ALWAYS works."
I wish him luck, more new people (his students) working on OSS projects is always good.
*Yes, I know HPUX, AIX, etc are still used and developed, but linux is much more widely used.
(Log in to post comments)
Interview with Andrew Tanenbaum (LinuxFr.org) Posted Nov 17, 2011 23:22 UTC (Thu) by chuckles (guest, #41964) [Link]
whoops,
has a LARGE chunk of the embedded market, if you include android, otherwise just a DECENT chunk.
Interview with Andrew Tanenbaum (LinuxFr.org) Posted Nov 18, 2011 0:36 UTC (Fri) by horen (subscriber, #2514) [Link] I'm not a developer or programmer -- just a Unix user/sysadmin since the mid-1980s -- but I believe the "BSD 'coulda been a contenda'" talk is incorrect. During the mid-to-late 1980s, BSD was great because it (SunOS, a modified version of BSD/4.1) ran on Sun's Motorola 680xx-series workstations, and academic institutions (by far the largest user-base) had made a serious and significant investment in that hardware. Sun's move to the SPARC architecture was the beginning-of-the-end, and the move from BSD-based SunOS to SVR4-based Solaris was the death-knell. It wasn't only academic users/developers who got bit; there were lots of small software houses which simply couldn't "toss" their hardware investment. Was SMCC wrong? Dunno... they're history, and I'm still here.
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 1:18 UTC (Fri) by khim (subscriber, #9252) [Link] I prefer contributing to GPL projects because I can give back without having to worry about my work be used against me when/if profit comes into the picture. I'm just one person, but I know I'm not alone. Perhaps there are more people of this mentality than of the "BSD='real' freedom" mentality. Actually I think that AT is right when he says that companies prefer BSD. And he is also right about reason: Companies using it don't have to tell us anything. My experience is that a lot of companies don't like to talk about what technology they are using. What does it mean? Well - you guessed it: small group of core contributes does all the development. Because companies are also doing development - but they don't share. It was discussed long ago.
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 3:47 UTC (Fri) by cry_regarder (subscriber, #50545) [Link]
Companies like it when OTHERS release code with BSD type licences...
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 5:34 UTC (Fri) by BrucePerens (guest, #2510) [Link] Everyone likes a gift. That doesn't mean you should give them one. Having individuals contribute their work to really big companies without any need for the companies to reciprocate in any way sounds a bit unbalanced, doesn't it.
It's not question of contribution - it's question of unification... Posted Nov 21, 2011 14:33 UTC (Mon) by cry_regarder (subscriber, #50545) [Link]
Yes, it certainly does!
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 9:30 UTC (Fri) by job (guest, #670) [Link]
This discussion always comes up and nobody has any hard data. Neither do I.
But we do know that it is much easier for companies to contribute to existing GPL codebases. Yes we have Apple releasing some things under BSD, and some companies in the firewall, routing, and storage business but even taken together doesn't match the long list of every single Linux release.
This reality simply does not reflect that "companies prefer BSD". There is absolutely a lot of software for which it works well, but it is not a license for everone. Many other things, such as community building, may even be more important.
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 13:41 UTC (Fri) by BeS (subscriber, #43108) [Link]
I agree I also don't have hard data to back my opinion up.
But in my experience there are three use cases we have to distinguish:
1. Companies want to use Free Software in their proprietary products. For this company AT is right. For them it's much nice to have non-copyleft Free Software than copyleft Free Software because than they can use the code however they wish and don't have to give anything back.
2. Companies who want to use Free Software in-house. In this case companies don't have to care about the details of any Free Software license because they never has any condition for just using the software.
3. Companies want to use Free Software and contribute to the project. In my experience this companies prefer copyleft Free Software because this guaranties that a competitor can't rip them off. With copyleft they can work in a community which also includes other companies and be sure that they will always be able to use the result of the collaborative work because every party has to give back the same rights.
Now, answering the question what license model you prefer is probably a question about what kind of cooperation you want to have with companies? Personally I think 2. is perfectly OK, 3. is the cooperation I want because this grows Free Software for the benefit of all parties involved. 1. is something I don't care about because even if I can understand the rationals behind this kind of use, for me it makes no difference if they use it this way or not.
It's not question of contribution - it's question of unification... Posted Nov 18, 2011 19:36 UTC (Fri) by wahern (subscriber, #37304) [Link]
#1 does make a difference to you, because Free Software generally tends to be of better quality than proprietary software; fewer bugs now and fewer in the future as many eyeballs submit fixes.
In my experience, the real stumbling block isn't simply using Free Software in proprietary products. Using Linux in an appliance isn't much of an issue if your actual product runs in a JVM.
The problem is using GPLd software in the particular components which companies believe are their competitive advantage and in a way which requires them to give up the source code.
In truth companies vastly overestimate how well secrecy and IP laws work in their favor in the marketplace. But nonetheless they believe they do, and we end up with crappier products because with GPL they'll refuse to use other, better code; or to release the code they do write so it can fixed in a timely manner by others.
Strategic use of the LGPL and BSD licenses can be quite valuable.
Interview with Andrew Tanenbaum (LinuxFr.org) Posted Nov 18, 2011 1:57 UTC (Fri) by jlmassir (guest, #48904) [Link]
"Is it just me or does he sound bitter?"
Of course he sounds bitter. In his fantasy, if he took the decision to promote MINIX at the right time he would be able to develop a complete operating kernel that could match Linux, including device drivers, IN A PROPRIETARY MODEL. Sort of a Linus owning Linux. Imagine how rich he would be?
Interview with Andrew Tanenbaum (LinuxFr.org) Posted Nov 18, 2011 9:34 UTC (Fri) by job (guest, #670) [Link] Inadvertent humor is the best humor. It's almost as funny as when serious people assert that the web would be more successful if it had been patented. Reality is no match for these people.
Interview with Andrew Tanenbaum (LinuxFr.org) Posted Nov 18, 2011 21:57 UTC (Fri) by oak (subscriber, #2786) [Link]
> "What most ordinary users want is that it ALWAYS works."
> > Uh, I'm not sure about that. From what I've seen, most ordinary users want something that's 'good enough'
They want something that ALWAYS works. However, what they're willing to pay for is the 'good enough'.
The "last 10%" to make something that seems to be working, to something that actually works on the real world, for large set of real users (different environment, use-cases, hardware etc), takes 90% of the effort. In proprietary SW it would show in the price & being late.
As it's also in practice impossible to cover all this with tests, what you typically get is "good enough" SW release which is then iterated with later releases, at least in the case of SW that has many millions lines of code worth of functionality and hugely varying uses & use environments.
I think AT's "ALWAYS works" comment was about SW working in a fixed (i.e. academic) test setup.
|
|||||||||||||