When Stallman is Attacked

writes "Linux Tech Daily has an editorial slamming a recent Forbes.com attack piece on Richard Stallman and GPLv3. Loved or hated, do you agree with the author that the piece is FUD and completely unprofessional? Love him or hate him, is this unfair treatment of rms? Does he leave himself open to these kinds of attacks with his behavior?" The problem with the editorial of course is that many of the points made in the original Forbes piece are completely valid and true. So basically you get to choose between the linux zealot, and a writer who is obviously fairly hostile towards Stallman's ideas.

Of course Daniel Lyons is spreading FUD

[br00tus]

The Forbes piece is written by Daniel Lyons. Lyons bashes Stallman, GPL, Linux, free software, open source etc. every chance he gets. He has been writing FUD for years. Just do a Google search for Daniel Lyons [google.com] and you can read people's thoughts on this. He came to the article with an axe to grind.

Re:That's GNU/Linux to you!

[Anonymous Coward]

and the zealot is the admin of a site called 'Linux Tech Daily'

What typo are you talking about again?...

Re:You don't have to choose...

[Raffaello]

"baffoon" - is that a buffoonish baboon?

Re:I frequently disagree with Richard Stallman

[Omnifarious]

They do fork over every single GNU program included in their distribution already. You seem to be very confused about how commercial companies in the free software space operate. A request of that nature by Richard Stallman or anyone else would be to point at the .torrent file for the source CDs. It's all there.

That's how Mandrake/Mandriva got their start. They grabbed all of RedHat's source CDs, and re-branded it as their own after making some changes they considered usability improvements.

I think you're so mired in thinking one way about how software can be sold that you can't see the reality that's right in front of your eyes.

Re:True of false?

[Anonymous Coward]

A glance at http://news.gmane.org/gmane.emacs.devel [gmane.org] indicates that the gentleman stays fully engaged in emacs development, though one could contend that he does more managing than hacking, I suppose.

Or one could realize, that the far superior Xemacs forked long ago because of Stallman's "attitude problem."

GPLv3 is incompatible with GPLv2

[greginnj]

Anything released by me in the future will carry a modified GPLv2 that does not permit the use of any future version of the GPL simply because this is a deliberate railroading of the purpose of the GPL.

You don't need to do this, just specify "GPLv2 ONLY" when you describe your license. Read on:

If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. (GPLv2) [gnu.org]

The 'any later version' stipulation is not part of GPLv2, it's part of the text that many people use to announce which license they're using. So you don't need to modify it (furthermore, you can't, if you're releasing a mod rather that something you coded up from scratch).

More interesting to me is article 6 of the GPLv2:

Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.

I can see a number of lawyers making hay of that, saying that GPLv3 introduces further restrictions. Note that when someone releases under "GPLv2 or any future version", the choice belongs to the recipient, not the licensor. The licensor can't grant rights under a license that doesn't yet exist when he makes the grant.

So, what do we get? All existing GPLv2 software is, and will forever continue to remain, licenseable as GPLv2. Even a new version of an existing program released under GPLv3 will still have its prior version and source available under GPLv2. And the first lawyer for IBM/RedHat/Novell who cares enough will use article 6 of GPLv2 to declare the extra restrictions of GPLv3 invalid for the new version of the existing program.

It's a shame that Stallman has gone on this crusade, but my money says GPLv2 is here to stay. Not many people will be releasing GPLv3 code, and the GPLv2 mainstream will fork them into irrelevancy.

Re:True of false?

[gclef]

Honestly, the whole DRM fight is really a mistake, if you ask me, for several reasons:

• Calling out specific parts of the US code just *begs* for someone to go change that part of the law. Once they do that, we'll need a GPL v4, and fast.
• Calling out specific parts of the US code make it much harder to internationalize the license. General principles are always better, since they translate better.
• It's very narrowly focused on DRM. If someone comes up with another scheme to accomplish the same thing, without cryptography (don't ask how, I'm projecting the future here), then we will need GPL v4 (or 5 or 6, whatever we're up to at that point) to cover that loophole, as well.

If Stallman/et al really want to handle the DRM stuff, they shouldn't put in the convoluted bit about encryption keys, etc. They should simply include a new line along the lines of:

If you distribute this work, you may not restrict how the recipients use it, nor may you restrict how derivates of this work are used or executed by the recipients.

The second clause would prevent TiVO from distributing both a DRM'd box *and* GPL'd software. They could distribute the box still, but they couldn't distribute the code with it, since the box is restricting how derivatives are executed.

Re:You don't have to choose...

[Anonymous Coward]

In short, you've done what most mediocre programmers do.

VIPER

[br00tus]

Whether you're serious or not, there is a VI within EMACS called VIPER (Viper Is a Package for Emacs Rebels). Scary.

Re:True of false?

[CableModemSniper]

No then he'd just demand it be called BSD/Linux.

Re:True of false?

[OrangeTide]

exactly. I don't like GPL, but there is stuff out there that is way worse. Companies somehow got it in their head that open source was the same as Public Domain, and start crying when the FSF gives them a call to explain the difference.

The software licenses that Apple and Microsoft give you are more like renting than buying. You didn't really purchase Windows Vista, you paid for the right to install it on one computer. That's the sneaky thing about copyright, it allows the holder of the copyright to determine the conditions in which you may use and copy the material. It's all legal and okay, but sometimes it seems a bit unfair. Especially when we end up paying $100+ for software that we can only use under very specific circumstances. (linux is great in this way, less bullshit than the competition)

sure, no strings attached to users of the software. that's nice. I'm a developer, so I tend to focus on how a license applies to modification and redistribution. When I license software it tends to have expensive royalties, paying an extra $3 on a device that might only cost $150 on store shelves is a pretty big deal. It's no surprise that some of the more aggressive companies are willing to accept the terms of GPLv2.

Many are pretty terrified of GPLv3. Some networking/router companies (who will remain nameless) are concerned v3 will require that users have the ability to upgrade software built-in to the device/appliance. This is a problem for this company (and others like it) because to be more efficient they sell the same hardware but install different software on it and "license" software through hardware. There are little authentication/DRM chips on the devices to ensure that only Software A is run on the device, and not the more expensive Software B.

These companies will have to adapt their business model to be compatible with GPLv3, or not use GPLv3 software in their products. Likely the later, especially since Linus does not seem to want to push Linux into using GPLv3. It could be a problem if glibc goes to GPLv3 though, it is hard to find a Java JVM for Linux that runs on anything but glibc. you end up having to twist the vendor's arm to get updates.

Re:Developers don't deserve freedom?!

[mrsbrisby]

"Once you buy the software, you agree to abide by the terms set by the publisher (seller). If you don't want to do this, please don't buy it."

Nope.

US Courts ruled way back in 1979 that this was _not_ the case.

I have a certain number of rights to use whatever I buy. If you want to take those rights away, you must use a contract. See 17 USC 117 [cornell.edu] for some very specific rights. Heck read ALL of Code 17: they're your rights. You should at least know what they are.

In Vault v. Quaid, 847 F.2d 255 (5th Cir. 1988), Federal courts ruled that so-called "click-wrap" licenses are NOT enforcable, and NOT legal.

Galoob v. Nintendo, 780 F. Supp 1283 (N.D. Cal. 1991), affirmed, 22 U.S.P.Q.2d 1587 (9th Cir. 1992) and also Foresight v. Pfortmiller, 719 F. Supp 1006 (D. Kan. 1989), say that you are looking at it from the wrong perspective.

(This does not necessarily reflect my opinion, but I like to look at issues from as many angles as possible)

I can appreciate this, but you're still wrong.

Fact is, no matter what Microsoft says, you can reverse engineer, and make changes to their software. It is 100% legal to take a Windows "demo disc" or a XP Home, and apply a patch to the registry that turns it into XP Pro. It's also legal to redistribute that patch.

No matter how many times the SPA says it, doesn't make it law. I'd appreciate it if you only say what you know to be true, instead of what you have heard to be true.

Re:True of false?

[Alphager]

There is a solution for all those problems you mentioned: encryption. Encryption is nowadays inexpensive (the software is free, and the cpu-cycles used by it don't matter nowadays).