Re: Analysis of the Free Art License 1.3 (original) (raw)

On Wed, 19 Jan 2011 21:48:32 -0400 David Prévot wrote:

[...]

Hi,

Hello!

[...]

Le 25/01/2009 07:55, Francesco Poli a Ă©crit :

On Sun, 25 Jan 2009 12:41:18 +0100 Francesco Poli wrote:

May I dig up this almost one year old thread,

I would say almost two years old, but I digress... ;-)

in order to ask or maybe add some enlightenment about this license, used by Kalle's for his design.

Welcome on board! :-)

As an introduction, I would like to point to the “Licenses” page of the FSF [...] that consider, about “Licenses for Other Types of Works”: “We don't take the position that artistic or entertainment works must be free, but if you want to make one free, we recommend the Free Art License.”

This is known and not so relevant as it may seem: the Debian Project is not the FSF and the FSF is not the Debian Project.

Too bad that the FSF doesn't take "the position that artistic or entertainment works must be free"! I take that position, instead. And, fortunately, the Debian Project uses the same guidelines (that is to say, the DFSG) to assess the freeness of any software work [1], not only computer programs. In 2004 it was decided (through general resolution GR-2004-003 [2]) to make this clearer, by amending the Social Contract.

[1] http://www.inventati.org/frx/essays/softfrdm/whatissoftware.html [2] http://www.debian.org/vote/2004/vote_003

Moreover, the fact that the FSF considers a license free (or even recommends it) does not necessarily imply that the Debian Project should come to the same conclusion (and vice-versa!).

This license is about artistic work, is initially written in French, [...] I believe that Francesco's concern about this license to comply with DFSG are mostly due to lack of context and maybe some not best choice of translation.

I admit that I only analyzed the English translation of the license, since I cannot read French legalese (my knowledge of the French language is unfortunately close to zero, apart from the meanings I can guess by analogy with my native language, which is Italian).

Anyway, the fact that the English translation is possibly inaccurate should IMHO be counted among the flaws of the Free Art License...

Free Art License 1.3 [...] 2.2 FREEDOM TO DISTRIBUTE, TO PERFORM IN PUBLIC [...] specify to the recipient the names of the author(s) of the originals, including yours if you have modified the work,

This might forbid anonymous works or anonymous modifications, which is non-free, IMO.

Actually, especially in artistic work, the “name of the author” doesn't need to comply with the civil one (most artists use pseudonym),

I really hope the intended meaning of the license is to not forbid anonymous or pseudonymous works/modifications, but I am still convinced that the actual license text is not clear enough on this point. That's why I said that this clause might forbid anonymous works or modifications.

and I fail to understand the difference with the GPL that ask to mention in the copyright.

Where does the GNU GPL (v2 or v3) ask that?

AFAICT, the GPL mandates the inclusion of appropriate copyright notices, which, however, may well use pseudonyms, or the term "anonymous", instead of civil names.

specify to the recipient where to access the originals (either initial or subsequent).

This condition is a little improved with respect to the corresponding one in the Free Art License version 1.2. However, I am still a little concerned that this could mean that, in order to distribute a work under this license, I am required, as long as I go on distributing, to keep updated information on where recipients can access every previous version.

Please keep in mind that artistic work may have one original, and that, unlike computer data, copy of the original may differ from it.

I am aware that the Free Art License is conceived in such a way to be applicable to material works too (for instance to statues, paintings, and so forth...). However, here we are analyzing the freeness of works licensed under the Free Art License for the purposes of the Debian Project. Hence the primary concern is how the license applies to works that may (technically) be distributed through the Debian mirror network and the Debian website(s): that is to say, software works, in the broadest sense of the term "software" (see again my above-mentioned essay [1]).

Please refer to the “Definitions” part of the license about “Originals”, “Initial work” and “Subsequent works”. Anyway, the French sentence that is “indiquer au destinataire où il pourrait avoir accès aux originaux” that could better be translated in “specify to the recipient where he could access the originals“. In artistic work, original may be broken, lost, or definitely altered by subsequent work for example. It really sounds like if “if possible” is missing (or is implied) in the English translation.

If my concern is indeed caused by an unclear English translation, then I would really love to see an improved English wording in the next version of the Free Art License.

Well, of course, I would like to see other improvements as well: most important of all, GPL-compatibility... But anyway...

What if the original changes, say, URL? Have I to keep track of where it goes?

If possible, but it can be the URL where you did find the original, so you “specify to the recipient where to access the originals” at the time you did your subsequent work or your copy.

If this is the correct interpretation of the license, then, great!, it seems that this clause does not cause any trouble. The problem is: where's an official and legally binding statement that assures this is the correct interpretation? Since the license text is not too clear, we cannot assume that any licensor will interpret the license as we would prefer...

What if the original vanishes? Have I to keep a copy of the original and make it available, in order to be able to distribute a subsequent work?

In artistic domain, original can vanish, it's implied.

Originals can vanish even in technical domains: I think that the original versions of several computer programs have been lost since the beginning of the history of computer science. And many originals still get lost nowadays, no need to go back to ENIAC times!

My point was: since originals can vanish (as you acknowledge), what happens when they actually vanish?

2.3 FREEDOM TO MODIFY [...] distribute the subsequent work under the same license or any compatible license.

This condition appears to be poorly drafted, since it could be interpreted to mandate distribution as a requirement to get the permission to modify. Under this interpretation (which I think was not intended), this license would force modifiers to distribute their modified versions, even when they would rather keep them private. Forced distribution is a non-free restriction, IMO.

The original sentence “diffuser cette oeuvre conséquente avec la même licence ou avec toute licence compatible” can be understand as in : if you “distribute the subsequent work”, you ave to do it ”under the same license or any compatible license.” The way things are written is really with physical object in mind (not computer data).

Well, that's the issue I was pointing out two years ago: unclear wording can lead to a non-free interpretation of the license (an interpretation that includes a forced distribution clause). Since we cannot assume that nobody will interpret the license in this non-free way, we have a problem, even if (as I suspect) the intended meaning of the license was harmless...

[...]

  1. COMPATIBILITY A license is compatible with the Free Art License provided: it gives the right to copy, distribute, and modify copies of the work including for commercial purposes and without any other restrictions than those required by the respect of the other compatibility criteria; it ensures proper attribution of the work to its authors and access to previous versions of the work when possible; it recognizes the Free Art License as compatible (reciprocity); it requires that changes made to the work be subject to the same license or to a license which also meets these compatibility criteria.

I think that, with these compatibility criteria, especially the last two, the GNU GPL (v2 or v3) does not qualify as a "compatible" license, unfortunately. I am not aware of any other license that meets the DFSG and may be considered a "compatible" license.

According to the French Wikipédia page [...], it is compatible withe the Commons Attribution Share-Alike (CC-BY-SA) v3.0

Where does the French Wikipédia page [3] says that CC-by-sa-v3.0 is "compatible" with the Free Art License?

[3] http://fr.wikipedia.org/wiki/Licence_Art_Libre

It currently states that "Elle est l'Ă©quivalent de la licence Creative Commons Share Alike By Attribution", which I think means "It is the equivalent of the Creative Commons Attribution Share Alike license".

Please note that "equivalent" != "compatible".

Anyway, independently of what Wikipedia says, I don't think that CC-by-sa-v3.0 qualifies as a "compatible" license, with respect to the Free Art License: it does not comply with the reciprocity condition and it does not allow modifications to be subject to a license that meets the compatibility criteria defined in the Free Art License.

and a quick search made me find a document proposed with both this licences [...]

The existence of works dual-licensed under any pair of licenses is irrelevant with respect to the compatibility, or lack thereof, between the two licenses. Actually, dual-licensing makes especially sense when the two adopted licenses are incompatible with each other.

We (as in the www-team working on including Kalle's design to the website and many other part of Debian for Squeeze release) would really like to know if we can consider this license DFSG compatible, or if we really need to ask Kalle to change it.

My own personal opinion is still the one I expressed about two years ago [4]:

[...] | In summary, this license seems to be intended to be a Free copyleft | one (but incompatible with GPLv2 and GPLv3). | There are some issues though that seem to make it fail. | | It's not a license that I would recommend, because of its issues and | lack of clarity.

[4] http://lists.debian.org/debian-legal/2009/01/msg00119.html

-- http://www.inventati.org/frx/frx-gpg-key-transition-2010.txt New GnuPG key, see the transition document! ..................................................... Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE

Attachment:<pgp8tEqVmF91V.pgp>
Description: PGP signature

Reply to: