[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-ohs-talk] GPL as a "chaordic" constitution for augmentation (was re: More bad news...)


At 08:49 PM 4/26/2002 -0400, Paul Fernhout wrote:
> > 5. Freely and fully exchange information relevant to the Purpose
> > and Principles unless it violates confidentiality or
> > materially diminishes competitive position.
>
>The GPL requires disclosure of new patterns derivative of GPL patterns
>from the common pool, promoting a free exchange of information. Those
>wishing to keep things confidential can decide not to put them under a
>GPL license by choosing not to incorporate GPL patterns into their
>approach. Note that the author who chooses to put a work under the GPL
>(and which was not otherwise made as a derived work from a GPL'd
>pattern) retains control over that work such that they can relicense it
>to others under different terms (for any amount of money). Similarly,
>those wishing to maintain a proprietary approach may do so by also
>avoiding GPL patterns, or by using them in such a way that their
>creations are not considered derivative works of GPL patterns, but
>simply coexist with them. (An example of coexistence would be creating a
>proprietary program that generates complex proprietary documents from
>proprietary templates, which are then further edited using a
>non-proprietary GPL'd editor, where the editing results remain
>proprietary.) A work derivative of GPL patterns must remain under the
>GPL, unless either all authors of all works involved agree to a license
>change, or unless some way is found to use part of the new derived work
>with changes in a non-derived fashion. This might be accomplished by
>changing the base subroutines calls it might make from calling a GPL'd
>library (which makes it derivative) to calling a non-GPL'd library
>(which might make it non-derivative).
>
>Note that although I refer above to a common pool of GPL'd patterns,
>that is a bit of a mistake. The GPL does not require you to give
>anything to anybody or to everybody. It only says that if you do give
>someone a derived work, that you give them the source under the same
>conditions as the work you derived it from -- that is, allowing free
>distribution with source. You can always sell the first copy of a
>derived work for a large sum of money, or you can only give derived
>works to those who you think will not redistribute them. However, you
>can't compel those people not to redistribute the derived work. I bring
>this up, because if, say, you personally had a huge library of GPL'd
>works, you could pick and chose which works and to whom you wish to
>distribute them, as can everyone else with those works. Thus is for
>example, you had a copy of potentially dangerous program under the GPL
>(e.g. a nanotech design for a weapon) you don't have to give it to a
>three year old child on request. However, if you did give that design to
>the three year old child, you could not legally prevent them (or their
>guardian?) from further redistributing it under the GPL (unless the
>child was your legal ward or issues of national security or criminal law
>etc. overruled copyright law). My point is that there is some room under
>the GPL for controlling distribution since data can be distributed by
>being given to trusted individuals rather than being posted on a web
>site. In practice though, everything gets posted to web sites...
>However, even then, people may now know about it unless sent a pointer
>from a trusted source -- in the same way people can go to the library to
>read the details about Christopher Columbus and what he and his men did
>to the Arawak Indians but hardly anyone ever does (see for example _A
>People's History of the United States_ or _Lies My Techer Told Me_).    (01)

And, earlier today, Richard Stallman wrote:
"In the free software movement, it is not a central question, because
our position is that non-free software is antisocial and free software
is an ethical imperative. If it is not profitable for a certain large
company to do free software development, that is a practical setback
for us, but nothing deeper than that. We say, "We'll get that freesoftware 
developed some other way." "    (02)

And, to repeat Chaordic Principle of Practice (POP)# 5:
> > 5. Freely and fully exchange information relevant to the Purpose
> > and Principles unless it violates confidentiality or
> > materially diminishes competitive position.    (03)

It seems clear to me that the Chaordic folks _do not embrace_ the free 
software ethical imperative ("...unless it violates confidentiality or
materially diminishes competitive position."). And, if that is so, all that 
verbiage about the ability to relicense, sell otherwise, etc that you use 
to justify satisfaction of POP 5 notwithstanding, I simply don't see how 
you can sell GPL to folks of the Chaordic persuasion.    (04)

Finally, GPL says this about patents: <copied from 
http://www.gnu.org/copyleft/gpl.html>
"Finally, any free program is threatened constantly by software patents. We 
wish to avoid the danger that redistributors of a free program will 
individually obtain patent licenses, in effect making the program 
proprietary. To prevent this, we have made it clear that any patent must be 
licensed for everyone's free use or not licensed at all. "    (05)

Paul, did you talk to the Chaordic folks about patents?    (06)

Now, consider this passage from LGPL: <copied from 
http://www.gnu.org/copyleft/lesser.html>
"7. You may place library facilities that are a work based on the Library 
side-by-side in a single library together with other library facilities not 
covered by this License, and distribute such a combined library, provided 
that the separate distribution of the work based on the Library and of the 
other library facilities is otherwise permitted, and provided that you do 
these two things:
       a) Accompany the combined library with a copy of the same work 
based on the Library, uncombined with any other library facilities. This 
must be distributed under the terms of the Sections above.
       b) Give prominent notice with the combined library of the fact that 
part of it is a work based on the Library, and explaining where to find the 
accompanying uncombined form of the same work. "    (07)

Contrast that passage to this one from GPL: <copied from 
http://www.gnu.org/copyleft/gpl.html>
"These requirements apply to the modified work as a whole. If identifiable 
sections of that work are not derived from the Program, and can be 
reasonably considered independent and separate works in themselves, then 
this License, and its terms, do not apply to those sections when you 
distribute them as separate works. But when you distribute the same 
sections as part of a whole which is a work based on the Program, the 
distribution of the whole must be on the terms of this License, whose 
permissions for other licensees extend to the entire whole, and thus to 
each and every part regardless of who wrote it. "    (08)

In LGPL, you are permitted to distribute combined works with mixed 
licenses, assuming the other license(s) permit that as well (BSD says 
nothing about this), whereas, in GPL, you cannot mix and match projects 
with other licenses.    (09)

One thing I have noticed when I query software authors about their use of 
GPL is that most of them use it simply because everyone else appears to do 
so, and, they were not aware of the fact that LGPL could fly with GPL, but 
was not viral.  Several authors promptly switched to LGPL.  A few simply 
stated words to the effect: "GPL satisfies my needs." without further 
explanation that they fully understood the issues.  This kindof reminds me 
of why and how Microsoft gained its monopoly share of the market.    (010)

It should be clear that I, for one, do not embrace the entirety of the fsf 
ethical imperative. OTOH, all this research is making me rethink my use of 
Apache license; LGPL is not viral (so far as I can tell) but does speak to 
the issue of patents in a way that I think is fair: 'you want to patent it? 
fine! but not with this library'    (011)

LGPL strikes me as much closer to the Chaordic POPs than GPL.    (012)

Jack    (013)