Licensing does matter

[Nothinman]

For those of you that think debian-legal's constant bickering and the ftpmaster's manual review of al new packages, OpenBSD's rewriting core tools/drivers under the BSD license and fighting for free access to docs, firmware, etc is all a waste of time take a look at this. One little licensing conflict is causing a decent amount of work for the Gentoo, *BSD and possibly Darwin folks.

http://farragut.flameeyes.is-a-geek.org...shadow-lies-upon-all-bsd-distributions
http://farragut.flameeyes.is-a-geek.org...-problems-requires-a-development-pause
http://farragut.flameeyes.is-a-geek.org...2007/01/07/the-big-license-mess-part-2

 

[greylica]

Sometimes I think why so many efforts to do things that aren't fully and clearly GPL'ed, if the GPL Kernel is so good. Assuming GPL Kernels is very stable and generally need few tweaks, and not a full recode, other fronts of work is needed to give tools to the end users that simply doesn't know how to operate Linux without the eye candy. It will be helpful to make people switch faster.

 

[drag]

"Sometimes I think why so many efforts to do things that aren't fully and clearly GPL'ed,"

Because GPL license is not appropriate for all circumstances...

 

[Nothinman]

Sometimes I think why so many efforts to do things that aren't fully and clearly GPL'ed, if the GPL Kernel is so good.

The issue here is actually a licensing conflict with the GPL, if the libkvm had been under the no-advert BSD there wouldn't be an issue, but since that BSD license imposed additional restrictions it's incompatible with the GPL.

 

[kamper]

Somehow I have trouble feeling any sympathy for gpl folk who complain about overly restrictive licenses and proclaim something as silly as "A shadow lies upon all BSD distributions". Not that the 4-clause license doesn't suck of course...

 

[Nothinman]

Somehow I have trouble feeling any sympathy for gpl folk who complain about overly restrictive licenses and proclaim something as silly as "A shadow lies upon all BSD distributions". Not that the 4-clause license doesn't suck of course...

The guy who made that claim is the Gentoo developer who discovered it, he's not "gpl folk" any more than any other OSS developer that happens to work on GPL'd software.

 

[n0cmonkey]

I'm not sure why this is an issue for BSD projects. Besides the 4 clause license sucking.

 

[kamper]

Originally posted by: Nothinman

Somehow I have trouble feeling any sympathy for gpl folk who complain about overly restrictive licenses and proclaim something as silly as "A shadow lies upon all BSD distributions". Not that the 4-clause license doesn't suck of course...

The guy who made that claim is the Gentoo developer who discovered it, he's not "gpl folk" any more than any other OSS developer that happens to work on GPL'd software.

That's what I meant, people who release software under the gpl as opposed to "bsd (license) folk" who release software under the bsd license.

 

[kamper]

Originally posted by: n0cmonkey
I'm not sure why this is an issue for BSD projects. Besides the 4 clause license sucking.

Exactly. I'm confused why this guy, just having discovered the problem, feels the need to tell the bsds that they're not doing things properly when clearly they are the people who know the licenses best and have been dealing with them for years. If there were a problem with the advertising clause, it would have come up already and the necessary amendments would have been made. A lot of the people in his list of copyright holders from the 3rd link are still quite active in the various communities and they haven't raised a stink.

 

[Slackware]

This is a problem because it is made into a problem, there is a perfectly good licence to use and release it under.

I guess i just don't see the problem, there have been kickback releases of BSD code released under a BSD licence for as long as i can remember.

 

[n0cmonkey]

Originally posted by: kamper

Originally posted by: n0cmonkey
I'm not sure why this is an issue for BSD projects. Besides the 4 clause license sucking.

Exactly. I'm confused why this guy, just having discovered the problem, feels the need to tell the bsds that they're not doing things properly when clearly they are the people who know the licenses best and have been dealing with them for years. If there were a problem with the advertising clause, it would have come up already and the necessary amendments would have been made. A lot of the people in his list of copyright holders from the 3rd link are still quite active in the various communities and they haven't raised a stink.

It looks like it's being taken care of (I know at least Open prefers 3 clause version or preferably an ISC like license). It's good that he brought it up, but I feel like he blew it out of proportion. These license "issues" slip through the cracks, but it isn't the end of the world.

 

[drag]

Originally posted by: kamper

Originally posted by: n0cmonkey
I'm not sure why this is an issue for BSD projects. Besides the 4 clause license sucking.

Exactly. I'm confused why this guy, just having discovered the problem, feels the need to tell the bsds that they're not doing things properly when clearly they are the people who know the licenses best and have been dealing with them for years. If there were a problem with the advertising clause, it would have come up already and the necessary amendments would have been made. A lot of the people in his list of copyright holders from the 3rd link are still quite active in the various communities and they haven't raised a stink.

Well the advertising clause is well known and has been a problem for many many years.

They people developing using this software probably just realy take the time and look at the exact license this software was made aviable under. This is not unusual. For a long time the majority of GPL violations were people using the software in commercial products without actually taking the time to look at the license. Licensing is boring and confusing and most people don't care.

I don't understand all that is going on (having not taken the time to look over stuff closely), but in situations like this it usually isn't a problem to convince people to change the license, the problem is actually _finding_ the people to change the license. A great deal of them may have done something as simple as changed email addresses 5 years ago and now nobody knows who they are or were to find them, but they still own the copyrights. Some of the original developers may even be dead and transfered the copyrights along with the rest of their property down to a relative. A relative that may not know enough to even know what BSD is and don't use the internet for more then looking for new baking recipes.

 

[kamper]

Originally posted by: drag
Well the advertising clause is well known and has been a problem for many many years.

Sure, but not enough of a problem to stop the bsds. I don't see why it should stop him or why he thinks this is big news. Not that I care if he just wants to cover his ass, that's fair enough and it's not like I'm going to be using gentoo/freebsd any time soon 😛

They people developing using this software probably just realy take the time and look at the exact license this software was made aviable under. This is not unusual. For a long time the majority of GPL violations were people using the software in commercial products without actually taking the time to look at the license. Licensing is boring and confusing and most people don't care.

Yes, but, as I said, he's not talking about commercial vendors, he's talking about the very people who are still releasing bsd licensed code. They know what's going on with the licenses.

I don't understand all that is going on (having not taken the time to look over stuff closely), but in situations like this it usually isn't a problem to convince people to change the license, the problem is actually _finding_ the people to change the license. A great deal of them may have done something as simple as changed email addresses 5 years ago and now nobody knows who they are or were to find them, but they still own the copyrights. Some of the original developers may even be dead and transfered the copyrights along with the rest of their property down to a relative. A relative that may not know enough to even know what BSD is and don't use the internet for more then looking for new baking recipes.

Well, if he does manage to get more of this hounding done that would certainly be useful. There was a story covering this on linux.com which pointed out that RMS was actually the one who raised the stink over the advertising clause in the first place. I guess he is good for something after all 😛 *ducks*

 

[Nothinman]

That's what I meant, people who release software under the gpl as opposed to "bsd (license) folk" who release software under the bsd license.

But that's a terrible way to categorize people. How would any Debian developer fit in there since there's numerous DFSG compatible licenses?

Exactly. I'm confused why this guy, just having discovered the problem, feels the need to tell the bsds that they're not doing things properly when clearly they are the people who know the licenses best and have been dealing with them for years. If there were a problem with the advertising clause, it would have come up already and the necessary amendments would have been made. A lot of the people in his list of copyright holders from the 3rd link are still quite active in the various communities and they haven't raised a stink.

It has come up already, the advertising clause is bad and any software using that license isn't compatible with the GPL. AFAIK most of the BSDs have stopped using it but apparently some slipped through the cracks. It's just that this one specific case has gone unnoticed until now.

Just because no one's raised a stink until now doesn't mean it's not a problem, just like the redistribution of binary kernel modules. Most people just ignore it and hope no one says anything but that doesn't mean the legallity of it isn't questionable.

 

[drag]

Sure, but not enough of a problem to stop the bsds. I don't see why it should stop him or why he thinks this is big news. Not that I care if he just wants to cover his ass, that's fair enough and it's not like I'm going to be using gentoo/freebsd any time soon

BSDs don't generally use the GPL... And it's not so much 'covering his ass', it's more that this is just the sort of BS you deal with when dealing with software licensing. If this issue was caught years ago it would be a non-issue now. But people ignored it or didn't notice it and it has been allowed to fester until it basicly halted development of two fairly major operating systems.

Yes, but, as I said, he's not talking about commercial vendors, he's talking about the very people who are still releasing bsd licensed code. They know what's going on with the licenses.

I don't think there is much of a difference. They probably just never bothered enough to look at it before. Laziness isn't something that depends on weither or not your a commercial vendor or a BSD developer, it's something that inherent in being human.

Also even though the BSD license FreeBSD uses permits this sort of thing, unlike the GPL, it is likely that FreeBSD has been in violation of 4-BSD itself for many years now. It is unlikely that they followed the advertising clause and posted notices in every sort of advertisment material that they produced for FreeBSD.

Well, if he does manage to get more of this hounding done that would certainly be useful. There was a story covering this on linux.com which pointed out that RMS was actually the one who raised the stink over the advertising clause in the first place. I guess he is good for something after all *ducks*

I don't know who made it a big deal at first. RMS has a way of pointing out issues that people don't like to think about or talk about in a rather undiplomatic manner. (it's why people love him so much) (ditto for Debian)

The advertising clause is irritating. The natural end result adopting such a license for a entire OS over years of development would be about 10-20 pages long worth of credits (each person using that license would have their own varient that used a different credit text) that they would end up having a meticulously maintain and make sure was present on every peice of documentation or box art or web page or whatever the hell you want to use it for and would change depending on the configuration. It's one of those things that was clearly a bad licensing decision.

 

[kamper]

Originally posted by: Nothinman

That's what I meant, people who release software under the gpl as opposed to "bsd (license) folk" who release software under the bsd license.

But that's a terrible way to categorize people. How would any Debian developer fit in there since there's numerous DFSG compatible licenses?

I'm sorry if you're offended that I call people who choose to release software under the gpl 'gpl folk'. The fact is, he's coming from a community that works primarily with the gpl even if they use other licenses. I made a generalization. You're nitpicking.

AFAIK most of the BSDs have stopped using it but apparently some slipped through the cracks. It's just that this one specific case has gone unnoticed until now.

If you mean stopped releasing new code with the 4 clauser, yes. But I don't think there's been much of an effort to remove 4 clause code. For example, openbsd's virtual memory system (borrowed from netbsd) still mostly has the advertising clause.

 

[Nothinman]

I'm sorry if you're offended that I call people who choose to release software under the gpl 'gpl folk'. The fact is, he's coming from a community that works primarily with the gpl even if they use other licenses. I made a generalization. You're nitpicking.

I'm not offended, it just makes no sense to me.

 

[Slackware]

The four clause makes very little sense to any of us if that helps any.

Free is apparently stepping away from it and Dragonfly never really found it comfy.

Net never even implemented it and i don't belive Open did either.

So, all is jolly good.

 

[kamper]

Originally posted by: Slackware
Net never even implemented it

Sure they did. Check the link I posted 2 posts above.

 

[n0cmonkey]

Originally posted by: drag
Also even though the BSD license FreeBSD uses permits this sort of thing, unlike the GPL, it is likely that FreeBSD has been in violation of 4-BSD itself for many years now. It is unlikely that they followed the advertising clause and posted notices in every sort of advertisment material that they produced for FreeBSD.

Nope, the clause was removed years ago on a lot of these files. It just didn't get removed from some of the files because there were probably THOUSANDS of these files scattered throughout the tree.

The advertising clause is irritating. The natural end result adopting such a license for a entire OS over years of development would be about 10-20 pages long worth of credits (each person using that license would have their own varient that used a different credit text) that they would end up having a meticulously maintain and make sure was present on every peice of documentation or box art or web page or whatever the hell you want to use it for and would change depending on the configuration. It's one of those things that was clearly a bad licensing decision.

* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.

If you don't mention that software or a feature it provides you don't need to add the name to everything. And what constitues "advertizing materials?"

Link: ?Now everybody get back to work. This is a 7-year-old nonissue??

 

[drag]

Nope, the clause was removed years ago on a lot of these files. It just didn't get removed from some of the files because there were probably THOUSANDS of these files scattered throughout the tree.

It was my understanding that with this paticular issue had to do with the fact that UCB did not actually hold the copyrights to this paticular code and therefore was not able to relicense it, but it's likely that I am wrong with this. If it's all UCB's code then it's a non-issue.

If you don't mention that software or a feature it provides you don't need to add the name to everything. And what constitues "advertizing materials?"

Well that's part of the annoying stuff isn't it? You'd have to assume that it is anything that is used in the attempt to get people to use the software. Posters, shrink wrapped boxes, ads, FreeBSD evagenical websites etc etc.


edit:
This is were I got my impression.. So if he is wrong then so am I.
http://farragut.flameeyes.is-a-geek.org...2007/01/07/the-big-license-mess-part-2

What the problem is, if UCB (University of California/Berkley) allowed distributors to drop the third clause, de facto converting the BSD-4 licensed code to BSD-3? Well, UCB is not the only copyright holder of some BSD-4 licensed software, as it comes out, so we?re not safe to just drop all of the BSD-4 clauses from the source code, which means I still have to clear up if as a mere aggregation we classify as not needing to abide to the clause or otherwise.

 

[n0cmonkey]

Originally posted by: drag
It was my understanding that with this paticular issue had to do with the fact that UCB did not actually hold the copyrights to this paticular code and therefore was not able to relicense it, but it's likely that I am wrong with this. If it's all UCB's code then it's a non-issue.

That seems to be what the various linked stories are saying...

Well that's part of the annoying stuff isn't it? You'd have to assume that it is anything that is used in the attempt to get people to use the software. Posters, shrink wrapped boxes, ads, FreeBSD evagenical websites etc etc.

Nope, you'd have to ask your lawyer what constitutes "advertising materials mentioning features or use of this software" means in your jurisdiction. I don't think anyone will try to convince you that this was a good clause in an otherwise fine license, but I don't think it's as much of a problem as the current stink makes it out to be. Unless you're expecting to be able to incorporate it into a GPLed program, but if you're doing that you're going against the spirit of the license anyhow. 😉

edit:
This is were I got my impression.. So if he is wrong then so am I.
http://farragut.flameeyes.is-a-geek.org...2007/01/07/the-big-license-mess-part-2

What the problem is, if UCB (University of California/Berkley) allowed distributors to drop the third clause, de facto converting the BSD-4 licensed code to BSD-3? Well, UCB is not the only copyright holder of some BSD-4 licensed software, as it comes out, so we?re not safe to just drop all of the BSD-4 clauses from the source code, which means I still have to clear up if as a mere aggregation we classify as not needing to abide to the clause or otherwise.

But I don't see multiple copyright notices in the libkvm files (on Free and Open), so I'm a bit confused. And this has aparently been corrected on both the Open and Free source trees (FreeBSD just did a bunch of commits ~21 hours ago, and a couple by Open around ~20).