Re: Amendment: GFDL is compatible with DFSG
I will _not_ second this proposal. Moreover, I would like to ask any
Debian Developer who's thinking of doing a second to consider what it
Legalese is not programming. See below.
On Mon, Jan 23, 2006 at 01:45:40AM +0200, Anton Zinoviev wrote:
> (2) The Invariant Sections - Main Objection Against GFDL
> Consequently the secondary sections (and in particular the invariant
> sections) are allowed to include only personal position of the authors
> or the publishers to some subject. It is useless and unethical to
> modify somebody else's personal position; in some cases this is even
> illegal. For such texts Richard Stallman (the founder of the Free
> Software Movement and the GNU project and author of GFDL) says :
> The whole point of those works is that they tell you what somebody
> thinks or what somebody saw or what somebody believes. To modify
> them is to misrepresent the authors; so modifying these works is
> not a socially useful activity. And so verbatim copying is the only
> thing that people really need to be allowed to do.
> This feature of GFDL can be opposed to the following requirement of
> Debian Free Software Guidelines:
> 3. Derived Works
> The license must allow modifications and derived works, and must
> allow them to be distributed under the same terms as the license of
> the original software.
> It is naive to think that in order to fulfil this requirement of DFSG
> the free licenses have to permit arbitrary modifications. There are
> several licenses that Debian has always acknowledged as free that
> impose some limitations on the permitted modifications. For example
> the GNU General Public License contains the following clause:
> If the modified program normally reads commands interactively when
> run, you must cause it, when started running for such interactive
> use in the most ordinary way, to print or display an announcement
> including an appropriate copyright notice and a notice that there
> is no warranty (or else, saying that you provide a warranty) and
> that users may redistribute the program under these conditions, and
> telling the user how to view a copy of this License.
This argument has been brought up a number of times already, but it does
The requirement in the GPL to display warranty or license announcements
is not absolute; it specifically states that you must only print such
announcements if the program 'normally reads commands interactively when
run'. In other words, if you modify the program to do something entirely
different in such a way that these license or warranty messages would
become problematic, you are allowed to remove them.
The primary objection to the invariant sections in the GFDL is precisely
that this is not possible; if you would want to synthesize a manual into
something small, you would still not be allowed to remove the invariant
sections. Worse; since after synthesizing the text the bits that are
about the subject matter could end up being smaller than the cumulative
amount of invariant sections, it might not even be legally possible to
synthesize a manual.
Synthesizing info manuals is something that Debian does regularly (or,
at least, should do; our policy requires that every binary comes with a
manpage, and that info documentation is not sufficient. Extracting the
relevant bits from the info manual would be the logical choice to remedy
> The licenses that contain the so called "advertising clause" give us
> another example:
> All advertising materials mentioning features or use of this
> software must display the following acknowledgement: "This product
> includes software developed by ..."
Again, this analogy does not hold. Advertising clauses only apply to
advertising material, not to the software (or the manual) itself;
conversely, the requirements in the GFDL regarding invariant sections,
acknowledgements and cover texts _do_ apply to the manual itself.
> Consequently when judging whether some license is free or not, one has
> to take into account what kind of restrictions are imposed and how
> these restrictions fit to the Social Contract of Debian:
> 4. Our priorities are our users and free software
This part of the Social Contract does not mean that we should bend the
rules of freedom to accomodate for our users. As such, it cannot be an
argument as to whether the GFDL is free or not.
> (3) Transparent copies
> Another objections against GFDL is that according to GFDL it is not
> enough to just put a transparent copy of a document alongside with the
> opaque version when you are distributing it (which is all that you
> need to do for sources under the GPL, for example). Instead, the GFDL
> insists that you must somehow include a machine-readable Transparent
> copy (i.e., not allow the opaque form to be downloaded without the
> transparent form) or keep the transparent form available for download
> at a publicly accessible location for one year after the last
> distribution of the opaque form.
> The following is what the license says (the capitalisations are not
> from the original license):
> You must either include a machine-readable Transparent copy ALONG
> with each Opaque copy, or state IN OR WITH each Opaque copy a
> computer-network location from which the general network-using
> public has access to download using public-standard network
> protocols a complete Transparent copy of the Document, free of
> added material.
> Consequently the license requires distribution of the transparent form
> ALONG with each opaque copy but not IN OR WITH each opaque copy. It
> is a fact confirmed by Richard Stallman, author of GFDL, and testified
> by the common practice, that as long as you make the source and
> binaries available so that the users can see what's available and take
> what they want, you have done what is required of you. It is up to
> the user whether to download the transparent form.
That would indeed seem to be the intent of that section, but it is not
what is written.
It says that you *must* either include a Transparent copy along with
each opaque copy (thus, if you print a book, you must include a CD-ROM),
or maintain a website (or something similar) for no less than one year
after distributing the opaque copy. As written, it is not enough to
point the recipient to an available copy that they can take if they want
to; you must either include it, or maintain a website.
Not everyone has the ability to do that. If I print out a copy of a
manual that I wish to give to a friend, then I do not want to be forced
to write him a CD-ROM, too; and I'm not sure that I want to maintain a
copy on my webspace, either (in my particular case that shouldn't be a
problem, but I can imagine that not everyone has multiple gigabytes of
diskspace on their webserver)
It should also be noted that the intent of the original authors of a
particular license is of no relevance for a license that can be used by
anyone; if it ever gets to court, then the interpretation of that
license by the court and by the author of the document that falls under
that license is what matters.
> (4) Digital Rights Management
> In fact, the license says only this:
> You may not use technical measures to obstruct or control the
> reading or further copying of the copies you make or distribute
> This clause disallows the distribution or storage of copies on
> DRM-protected media only if a result of that action will be that the
> reading or further copying of the copies is obstructed or controlled.
> It is not supposed to refer the use of encryption or file access
> control on your own copy.
No; however, as written it can be interpreted as such. We all agree that
this is a bug in the license, but agreeing on that does not mean that
there is no problem.
> Consequently the measures of the license against the DRM technologies
> are only a way to ensure that the users are able to exercise the
> rights they should have according to the license. Because of that,
> these measures serve similar purpose to the measures taken in the GNU
> General Public License against the patents:
> If a patent license would not permit royalty-free redistribution of
> the Program by all those who receive copies directly or indirectly
> through you, then the only way you could satisfy both it and this
> License would be to refrain entirely from distribution of the
Note the difference in detail and length between this section and the
above two lines. This section is clear in that it only applies to patent
licenses and non-royalty-free redistribution. Contrary to that, the
above section is not clear in that it applies to DRM; it uses the
(undefined) wording "Technical measures", which could apply to anything.
> We do not think that this requirement of GPL makes GPL covered
> programs non-free even though it can potentially make a GPL-covered
> program undistributable. Its purpose is against misuse of patents.
> Similarly, we do not think that GFDL covered documentation is non-free
> because of the measures taken in the license against misuse of
> DRM-protected media.
That is correct; however, as written, the section that is supposed to be
about DRM (and DRM alone) can apply to anything.
Legalese is an unfortunate accident of nature. It has its own problems,
most of which involve arbitrary interpretations of a text which do not
match up with the intended interpretation of that text while still being
consistent with the literal text as it is written. It has nothing to do
with exact sciences.
Obviously this is unfortunate; however, pretending that such problems do
not exist will not make them go away.
It is a fact that the transparent copy and DRM clause problems are bugs
in the license; I will agree with you that the intent of the license's
authors was not for these clauses to be interpreted in such a way that
we would think they are problematic. However, unfortunately intent is
not enough, as the GFDL should make everyone painfully aware.
.../ -/ ---/ .--./ / .--/ .-/ .../ -/ ../ -./ --./ / -.--/ ---/ ..-/ .-./ / -/
../ --/ ./ / .--/ ../ -/ ..../ / -../ ./ -.-./ ---/ -../ ../ -./ --./ / --/
-.--/ / .../ ../ --./ -./ .-/ -/ ..-/ .-./ ./ .-.-.-/ / --/ ---/ .-./ .../ ./ /
../ .../ / ---/ ..-/ -/ -../ .-/ -/ ./ -../ / -/ ./ -.-./ ..../ -./ ---/ .-../
---/ --./ -.--/ / .-/ -./ -.--/ .--/ .-/ -.--/ .-.-.-/ / ...-.-/