Toward true open source (NewsForge)

Posted Jul 16, 2004 19:52 UTC (Fri) by Ross (subscriber, #4065) [Link]

The methods aren't the same either. Since when was the GPL a technical
solution for a contract enforcement problem which has unitended side effects
that prevent fair use of the copyrighted work? Since never. This article is
just FUD.

Posted Jul 16, 2004 19:54 UTC (Fri) by TwoTimeGrime (guest, #11688) [Link]

The author can license the work as he sees fit. If you want to extend a product and don't like the GPL then contact the author and see if he'll license you a copy under a different license just for you. Companies do this all the time. Each customer may have a different license for the same product that says how they may use said product.

If you're looking to sell software you should get a good lawyer. Your lawyer would have pointed out the option of contacting the software author about licensing. If not, get a better lawyer.

(I posted this same comment at the bottom of the story)

Posted Jul 16, 2004 21:12 UTC (Fri) by socket (subscriber, #43) [Link]

The author of this article misses some really basic points, and seems to think that the GPL means something other than what it actually does. It's not the same thing as letting anyone else use it under any terms, and it's not "jealously guarding" the code.

Let's look at the specific issues raised.

"...its legal complexity." Compared to a proprietary EULA? I can read the GPL and understand what it means, so I'm sure most lawyers could, too. And companies may exist that have taken the "No GPL in-house" stance, but I've never heard a specific example given, and without it, it sounds like FUD-mongering.

"The main harmful effect of the GPL is the "gated community" it creates. When you release your source code under the GPL, the only people who can re-use it are those who also license their software under the GPL."

Yes. This is the point. There's no reason for this to be in any section about "Unintentional" effects. Again, nobody *has* to use the GPL or GPLed code. There's a nice, large quantity of good GPLed code out there, but if they don't like the terms, well, whose problem is this again?

"What if my Web browser is proprietary? I'll have to write my own compression routine"..."Ah, but what if I want to release my program under the BSD license, or some other alternative open source license?"

Nothing, absolutely nothing in the GPL forbids you from dual-licensing. In fact, it's been rather common in some of the higher profile projects out there. If you kept track of who holds what copyrights in the code, and can work it out with all the developers, this is completely a non-issue. What you can't do is try to take all your GPLed releases out of the 'net.

The rest of the open source community is standing at the gates of free software, denied access to the masses of GPLed source code within."

Then they can sit down and think about why someone might have GPLed their code in the first place. Perhaps because they want the users of *your* code to be as free as you are to do what you will with the code you received from us. Is that such a bad priority to have? There are always other, conflicting priorities, and some people will rank this variety of freedom below the freedom to make the code proprietary. Both are freedoms, allocated value differently by different licenses. Why, exactly, should someone who writes free software and wants there to be more free software in the world, be happy allowing other people to take the product of their work and make it proprietary for themselves? But if we all took our code proprietary, there would not have been a "open source community" or "masses of GPLed source code" to be at issue in the first place. If you don't like the gift, at least have the decency to not gripe about it to the people who gave it to you.

"You can't stop undesired usage, so leave it open," Tom says, but let's think about how well this goes. Regulations exist to punish undesirable actions, and to deter people from initiating undesirable actions. Why do I lock my front door, or have a firewall on my network, if someone sufficiently determined can get past both anyway? Because it raises the barrier; maybe this is all that laws really do, but a lack of a barrier is always worse than any barrier at all. Not that this horse hasn't been beaten to death in the first place, but the GPL wouldn't be used if people didn't actually want the GPL's terms on their software. If someone is discovered to be violating the terms of a software license, they deserve what they get when it is discovered. Same for free, same for proprietary.

Anyone still whining wants to not play by the rules. The rules are simple: if someone wants to use my GPLed code in their application, they've got three choices:

1. Use the GPLed version and abide by the GPL.
2. Contact me to see if I'd be willing to provide it under different terms. I probably would, if they made it financially worth my while.
3. Go use some other software.

So, why exactly does the GPL not do exactly what it was intended to do? It was never intended to be a blanket "You can do whatever you want with this code." Stallman designed the GPL to have just enough restrictions to guarantee that a given codebase will always have certain specific freedoms associated with it. Removing those restrictions is not the right of anyone but the copyright holder, and never retroactively. If you don't like it, nothing requires you to use it.

The open-source/free-software community exists because it's a group of people concerned with being able to see what we're getting when we get a piece of software. Whether we intend to change the software, look at it, or just use it without ever even glancing at the code, it matters to us whether we can get at the code when and if we want. The GPL is the backbone behind this, because it creates our gate: if you want our code, play by our rules. It's about making sure that the code is good, and works well. What better motivation to write good code than knowing that not only will you get the credit for having contributed, but also that the world of software is better off for your work? We are a community of users and developers that can work together by providing each other help on projects, rather than by building walls between each other. We reinvent the wheel less, because we have the code for all the wheels we've already written, and can figure out how well they're actually going to work for us because nothing's hidden from view. It's the same thing motivating a public scientific process: with more data, and more voices, we'll discover the nature of things faster. With more GPLed code, we'll have better software faster. It isn't by being proprietary that HTTP and HTML made the 'net useful or popular.

And you don't have to use our code if our license bothers you so much.

</rant>

Posted Jul 16, 2004 21:13 UTC (Fri) by hamjudo (guest, #363) [Link]

"The first side effect of the GPL comes from its legal complexity."

Yes, software licensing is hard. It's even more complex when you want to mix commercial and free software. What I've noticed in large companies is that the legal department has already seen the GPL and they mostly understand the rules. That makes it relatively painless to legally use and distribute GPL software along with commercial software.

I've even seen a GPL compliance audit. The lawyers wanted to make sure that the source code actually matched the executables on the distribution media.

It is harder to distribute 3rd party software with other licenses. Sometimes, it is just because the legal folks don't know the other license. Other times, it is because the other license is badly written or has terms that are hard to meet.

It is time to publicly thank the developers of gzip, for their trailblazing work. Gzip let us fit our bloated applications on the limited media of the 1990's. Thus providing a handy training opportunity to many legal departments.

Posted Jul 16, 2004 21:21 UTC (Fri) by jae (guest, #2369) [Link]

Almost typical anti-GPL crap.

Almost, because he states what the GPL does quite lucidly. Which is rare, even for GPL enthusiasts.

But: what's the alternative? BSD? That one allows anyone to take the software I or you wrote and make it closed. And then sell it to use!

To illustrate: say you work on some BSD-licensed SW. You know that SW inside and out (because you`re the creator or you just dug in deep and long enough).

Now, a while later, your employer bought some other big closed-source piece of SW. Which, incidentially, contains that first BSD-licensed SW.

There`s a problem with this new SW. You try to debug it, and find that it's in the part derived from the first SW, the one you know very well.

You can't fix it. Try to explain *that* to your boss.

Posted Jul 16, 2004 21:55 UTC (Fri) by chip (subscriber, #8258) [Link]

"Freedom to modify" vs. "Freedom to hide" is the dichotomy. Again.

Thanks for wasting our time with it. Again.

Posted Jul 16, 2004 23:17 UTC (Fri) by pyellman (guest, #4997) [Link]

WOW. Who is this clown? I mean, it's fine to have an opinion on the merits of the GPL, including negative ones, but he made a real fool out of himself by stressing "The GPL has unintentional harmful effects". Someone needs to tap him on the shoulder and point on that nearly all or all the effects he describes, despite the fact that he uses pejorative language in doing so, are entirely intentional.

Peter Yellman

Posted Jul 16, 2004 23:50 UTC (Fri) by jre (guest, #2807) [Link]

A couple of LWN readers have independently hit the nail on the head:

There's no reason for this to be in any section about "Unintentional" effects.

[N]early all or all the effects he describes ... are entirely intentional.

That's it, exactly. Tom Walker is a bit more articulate than the great boobies at AdTI, but they are philosophical kindred. The common theme is something like "GPL'ed code is not really free, or even open source, because it doesn't permit [insert superficially attractive proprietary use here]."

Hogwash.

They are both wrong, though I suspect Tom Walker is wrong because he misapprehends the purpose and function of the GPL, and AdTI is wrong because they understand it all too well, and don't like it, and lie about it, clumsily.