The GPL Impedes Linux More Than It Helps? 386
Anonymous Coward writes "Linux ought to be even more successful than it is. On ZDNet, Paul Murphy ponders the reasons why. For one thing: The GPL impedes Linux more than it helps. Licensing issues, coupled with patent and copyright FUD, have caused developers and VCs to think twice before committing to Linux. Murphy also suspects that desktop Linux is stuck on stupid." From the post: "Basically, legal issues, or the threat of legal issues, caused some key applications developers to back off Linux while the general negativism of Linux marketing caused many of the individuals whose innovations should have been driving Linux adoption to hang fire until MacOS X and Solaris for x86 under the CDDL came along."
Linux and GPL (Score:5, Insightful)
Forgetting development. (Score:5, Insightful)
Re:Who needs Linux when you have OSX? (Score:5, Insightful)
The low profile of FreeBSD when it is used in the enterprise (I'm talking servers, not OSX) is evidence that the GPL does nothing to hinder Linux. With a BSD-style license Linux would have no advantage to developers over BSD and wouldn't be in the position it is now.
Impedance... (Score:5, Insightful)
Note that we could also say the same thing about proprietary, commercial software too: that licensing restrictions and costs impede its adoption. But they also create the circumstances in which that software is created.
The goal of the GPL has never been rapid adoption of software, but rather adoption under particular circumstances.
Anyway, has there ever been a time between 1991 and now when Linux and free software in general have not grown in user base?
from a user's perspective (Score:3, Insightful)
As for Microsoft FUD - that's simply directed against any competitor. GPL is rallying banner for most of the opensource community, so naturally they're targetting it with their immense advertising budgets.
Not surprising (Score:5, Insightful)
You'd maybe see software technologies developed for linux integrated into proprietary commercial closed-source applications, just as they did with the BSD implementation of TCP/IP in MS Windows, or BSD/Darwin into Mac OS X.
It wouldn't bring about the desired effect of keeping software Free, though. What do we want Linux to be?
Re:Forgetting development. (Score:1, Insightful)
1: Make free stuff.
2: ?
3: Profit!
And rightfully so. Therefore, lots of developers use linux but dont make software for it.
And just to clear the abvious, free as in speech also means free as in beer for the vast majority of developers.
Don't underestimate the power of the free side (Score:5, Insightful)
Even if the GPL is slowing corporate adoption, an assertion proved by nothing more than the statement "I think" and a fun little example of the post hoc fallacy, that's no reason to ditch the concept. Sure, more corporations might adopt Linux if it were a closed-source program, but why they'd want a relatively unsophisticated OS by some Scandinavian kid instead of the more robust UNIX is beyond me.
Do you see what I mean? You can't separate the success of Linux from its community and core ideal. They rise and fall together. One of the things I respect about ESR is his realization that good code alone won't win adoption for a GPL'd program. This is about ideas as much as code--and philosophers and salesmen are as much combatants against Microsoft and chattel software* as any F/OSS programmer.
*I asked RMS about that phrase. He didn't think it was all that good, but I still kind of like it. What do you think?
Linux-GPL = BSD (Score:5, Insightful)
Yes, Linus is a talented manager. But he also started without the tremendous codebase that BSD has always had.
Personally, I'm getting a little fed up with the anti-GPL griping. I suspect the gripers of wanting to abuse code they didn't write. People married to the commercial commodity model of software so successfully exploited by Bill Gates. I have yet to hear an objection I find balanced. Most are just "I want more".
Is this in an alternate reality? (Score:2, Insightful)
I believe that Linux has been significantly helped because of the GPL. Anybody that is worried about licensing issues with the GPL can just use a BSD derivative and call it a day.
As for the CDDL I have a feeling it will get little attention since it is not compatible with the GPL. It's like creating their own little island community of developers. Yeah, it's nice that they're opening their source code, but there's not much use in everyone dabbling in it because you are unable to take the work elsewhere. I remember IBM trying the same thing with their own incompatible license and it went absolutely nowhere.
Hidden assumption (Score:3, Insightful)
IBM's endorsement of Linux, the SCO law suit in response, and Red Hat's negative market stance as the Sun killing would be Microsoft of the Linux era combined to destroy the automatic assumption among key innovators in the United States that Linux was "the place to be" -eventually moving many of them to the BSD and Solaris camps where they're now driving the fastest installed base expansions in the history of computing
Murphy talks about an automatic assumption but he's hidden one of his own in this para: that the only key innovators in the US are vendors and venture capitalists. GPLed software lets just about anyone with half a brain and an itch to scratch be an innovator.
Re:"Ought to be"? (Score:2, Insightful)
That's a pretty circular argument, almost like saying "everything that's sucessful can't be improved".
It's entirely possible that licensing isn't one area of Linux that is in dire need of improvement, but don't use the argument "this is what got us here" to back it up.
Software Freedom (Score:1, Insightful)
Re:True to an extent... (Score:5, Insightful)
I don't think takeup would neccesarily be better with a BSD license, either - as evidenced by the fact that BSD takeup lags far behind Linux.
Muddled and Meaningless (Score:5, Insightful)
I was reading opinions like Linux is failing because of the GPL and kept thinking "in what particular way? Give me an example where the GPL is failing Linux - a hard real example such as 'technology professional X reviewed Linux and found this failing in the GPL so decided to go with another choice'". Or the opinion that Linux should try to be something other than a WinDOAs look alike - such as what precisely? I mean it's really easy to point out flaws, but just a tad more of an undertaking to provide real answers and solutions.
Reading all this felt a bit like someone saying they think my shoes are ugly without any real information on how they could be better or why particularly they're ugly. I mean he has a right to his opinion of things but ultimately, if he was hoping to actually keep my attention, I would think he would try to at least give me something concise, with real value and of some interest to me. Ultimately I was left with the impression that he can insult Linux, and the point in that exercise is what? Was it just me who was left feeling that way?
How about a stable ABI? (Score:2, Insightful)
Notice how on most OS's you can own a CD of an application and just install it? Because there are STANDARDS! That's what hurts third-party support.
Flexibility is good, but if they would make a usable standard and stick with it, we might not have to worry about recompiling so often.
Re:Subject (Score:3, Insightful)
It does help (Score:5, Insightful)
Right now, MS could support BSD and kill the market from under Apple. That is what happened in Unix, after it was closed. The big players slowly killed off the little guys by adding closed source that was unavailable to them.
Besides, keep in mind that only Windows is a moneymaker (and that is due to the monopoly in Office). No other OS makes a direct profit. Not even Apple, or any of the linux distros.
Re:True to an extent... (Score:5, Insightful)
Re:True to an extent... (Score:4, Insightful)
Re:Impedance... (Score:5, Insightful)
Exactly. Very, very well put. (Bonus: air is also needed by the engines.) It's like he's saying "Ferraris are great, but they won't be popular until they're less than $10,000." You can't have it both ways. What makes a Ferrari great can't be done for less than $10,000. Yes, there are places where Linux being non-GPL would help, but Linux would not be where it is today if it weren't GPL in the first place. Everything has its pluses and minuses.
And desktop Linux is not stuck on stupid, the author is stuck on stupid:
"...Linux growth didn't slow because of competition - something else must have caused it and we need to understand what that was before we can work up a plan to do something about it."
Um, maybe Linux just got to the point where everyone who wants it, has it? There are such things as saturation and natural limits. Just because Linus jokes about world domination does not mean that Linux is a failure if it isn't the only system in use on every computer everywhere. Would he consider it a success if humans killed off every other species on the planet?
Re:Forgetting development. (Score:4, Insightful)
Most software developers simply don't beleive in the businessmodel: (1: Make free stuff., 2: ?, 3: Profit! ) And rightfully so. Therefore, lots of developers use linux but dont make software for it.
Any developer who thinks that is the proper business model or any business model at all is too stupid for me to want to use their software anyway. The GPL is a great license for software and brings many advantages to the user of the software, not the developer. Your post implies the GPL is about getting more for doing less; this is not true at all. The GPL is a feature of software. By your argument no product should have any features beyond what is needed for it to function. Why would a car maker include air conditioning? It costs more money to include. The answer: because customers want it. If I build my own car from scratch, I'll include air conditioning, just as Linux users who created their own OS included the GPL.
The GPL is a license that is designed to benefit the end user of software. It was written by end users who also happen to be developers. As an end user of software the GPL means I can use a product and modify a product and redistribute it however I like. I can hire anyone to work on it I like. No one else can take all the effort I have put into it, add something, and make a profit off of it without giving me back something in exchange for all my work. It enforces fair collaboration on projects. All this is great for me, as a user.
That said can people make money by creating GPL software? Hell yes. Can construction workers and engineers make money constructing a bridge that is not a toll bridge? Yes. Can artists make money creating a commissioned mural in a building? Yes. Bridge builders, however, don't use the business model of, I think I'll build a bridge here and then try to get the county to pay me for having built it. Artists don't go into buildings, paint murals, and then try to negotiate a payment for it with the building owner. Software creation using the GPL can be plenty profitable if you find someone or some group that wants to hire a work to be done. The problem is that shortsighted and slow people cannot understand using any business model except the one used by current commercial software developers, even if it is one that is much more beneficial to them, personally.
As an addendum, you can make money for additional commission work adding features and customizing software, and in some cases with advertising revenue and by supplying expert support and/or consultation. Any businessman who cannot grasp the advantages of using GPL software tools to solve their businesses needs should be fired immediately. It is not always the right or best solution, but it certainly has some compelling advantages in terms of immediate cost, competitive supply, sharing expenses, industry interoperability, shared research costs, free advertising, and in-house talent development. Any developer who does not want to create GPL software, that is fine, create licenses that favor you as much as possible, just don't expect your customers to be willing to put up with it in the long run when they have better alternatives.
Yeah, GPL hurts Linux (Score:5, Insightful)
Translation: We can't take the code developed by thousands of programmers over 15 years, make it proprietary, and contribute nothing back.
Response: Yep. that's the whole fucking point!
Re:Linux and GPL (Score:2, Insightful)
I would really like to know more about those legal searches you did for your company's sources. What did you have to check? How did you make sure you were allowed to use the tools the way you used them? How did you interpret the licences? etc.
My company thinks about porting its proprietary application to Linux, but definitely doesn't want to GPL the application. Could we use GCC4 to compile it? I saw it includes some libc files that are GPL with a tiny comment that you can still use this file without making the program GPL, so using those files are ok, but does GCC4 include other files that doesn't have that very important exception? That kind of thing can scare the managers as hell...
Re:Forgetting development. (Score:1, Insightful)
Personally, I don't develop for linux because I can't be sure of the target system. It's not like the Open Source community of developers have never "depreciated" anything before, or broke it, or made something that used to work *not* work because they felt it was better to do it a different way. Of course if you question that wisdom you're essentially told to fuck off. It's their sandbox and they'll be the first one to tell you that if you don't like something fix it yourself. Additionally once you've done that job, and fixed it yourself they want you to come crawling back to them with your tail between your legs wimpering, and uploading the patches at the same time.
For me, the GPL isn't hindering development, it's the other developers.
Re:True to an extent... (Score:3, Insightful)
Re:Linux-GPL = BSD (Score:3, Insightful)
I don't know about that. BSD and Linux have two different philosophies as far as design goes. BSD is a system, Linux is a kernel. You need a separate userland (not provided by Linus himself) in order to do anything with the Linux kernel.
I agree. He didn't have the codebase that BSD has had access to since the 1970s. However, he had access to GNU. GNU had everything needed to build an operating system except for a kernel. Linus had just a kernel, but no userland. This was a perfect match, since the GNU userland and development tools are quite portable and of great quality.
I think you'll have a Linux kernel without the GPL, but you wouldn't have the current Linux system without those utilities from GNU, which are licensed under the GPL. Indeed, GNU has made a huge contribution to the computing world with its tools and license, and even BSD users like myself benefit from GNU; most of my software (KDE, OpenOffice, GIMP, GAIM, etc.) are GPL-licensed, and GCC is the main compiler on all BSD systems.
Re:"Ought to be"? (Score:3, Insightful)
I think that they discovered that this kind of article provoke outrage in Linux community and they publish them for the money from the ads they serve to the outraged (and curious) public.
Who cares? (Score:3, Insightful)
What's so great about "quick"?
All of these businessfolk, always wanting things to grow quickly. I'm much more concerned with Linux adoption growing the right way, than as quickly as possible.
I know what some of you are saying, "With that attitude the Linux world will lose a lot of business." Yep. Get over it. ANd don't be so greedy, kiddo.
Re:Linux and GPL (Score:5, Insightful)
To what end? To allow businesses to re-licence their modifications under proprietary terms? The BSDs already exist to that end. To be sure, they've had some prominent business adoptions. The only problem is that afterwards they're not really BSD anymore, and it's hard to see how OSX for example actually brings any benefit BDS.
What else could appeal to business? Dropping the source code requirement for modifications? That more or less morphs back into the same case as a BSD licence. In addition, it's hard to see how this ameliorates any suspicion over a gift culture.
I really can't see what changes you'd make.
It should also be pointed out that for many of those who write GPL software, corporate adoption is not a high priority.
Re:Forgetting development. (Score:4, Insightful)
You have a good point, but in most respects developing for Linux isn't any worse than developing for Windows. I work for a software development company, and we have had fits over the years with the multiple versions of Windows, service packs, IE updates, Outlook versions, etc.. that our customers want to use with our software. Compatibility is not a problem exclusive to Linux.
It's not like the Open Source community of developers have never "depreciated" anything before, or broke it, or made something that used to work *not* work because they felt it was better to do it a different way. Of course if you question that wisdom you're essentially told to fuck off.
Again, not much different than developing with a Microsoft product. MFC has plenty of problems and the bug you worked around in this version will probably be completely different (not fixed) in the new version requiring a complete work around, and if you think Microsoft cares about your problems you are sadly deluded.
There is no perfect environment, and there are challenges regardless of where you are developing. If you can find a Linux product that there is enough market share for to generate revenue you should take advantage of the opportunity. Problem is finding people that will pay for Linux apps.
Re:Linux and GPL (Score:3, Insightful)
Care to explain? How is it a hindrance to India in particular?
Re:Linux and GPL (Score:3, Insightful)
Well... you suggested different licences for different countries, rather than a per-nation modification of the GPL. However I will confess that I see little functional distinction. Whether we change the GPL to operate like the BSD licence in, say, India, or whether we Licence it under the BSD licence in India, the results are the same so far as I can see.
Also I don't see why the BSD license cannot be used for distributing Linux in countries like India where GPL is more of a hinderance.
I've yet to be convinced of the benefits of a change in or to the licence. We already have BSD under a BSD licence. For all the good points of BSD, Linux remains more sucessful. That's hardly a compelling reason to change.
What's the problem with the GPL in India, anyway? It's not mentioned in TFA. May I assume from your handle that you have first hand experience?
Re:Forgetting development. (Score:3, Insightful)
OTOH, you are completely correct that open source newsgroup posters can be hostile and arrogant. Personally I rarely post, and If I do I have thoroughly searched previous postings, other newsgroups and the web in general for the answer to my question. If my question is well thought out and phrased politely I usually get a positive response. The difference is an Open Source developer doesn't have an image to protect. If your product is good people will use it, especially if it's free. Microsoft is in a different position, they have a corporate reputation to think about, so they are nice when you ask a question.
In the end the result is usually the same. The problem doesn't get fixed. The Open Source developer tells you to fix it yourself and Microsoft promises it will be out in the next relase that you have to pay $500 for. Only real difference is how easily your feelings get hurt when someone you don't know is mean to you in a newsgroup, and of course how much you want to pay for the next release of a product.
For cripe's sake!! (Score:5, Insightful)
Now back to your scheduled flamewar.
Huh? I must REALLY be missing something. (Score:5, Insightful)
Bruce
Re:It does help (Score:3, Insightful)
Who is this guy and why should you even care?!? (Score:3, Insightful)
But seriously, there's nothing here. He's jumped on the same old anti-GPL train that has been going around for a while. Let's not give this guy anymore airtime...there's no value in his suppositions.