[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: The invariant sections are not forbidden by DFSG



Anton Zinoviev <anton@lml.bas.bg> wrote:

>> >  Would that be inconvenient to Frank? -- Yes.  Does this
>> > inconvenience obstruct the software freedoms somehow? -- Certainly
>> > not, the users get the file Frank wants to give them.
>>
>> No, many won't download the file if they know they have to download 10
>> MB in order to get 900kbyte of content. 
>
> The invariant sections in the GNU manuals are not that large but I
> suppose you are not talking about some existing document?

May I ask you to please read the mails you answer to?  If you do, you'll
know. 

> If so I will agree with you -- it is possible to create the invariant
> sections that large that it becomes serious burden to distribute them.
> If this is realy the case, then this document would be non-free.

Since AUCTeX's manual is *not* under GFDL, I cannot take realistic
numbers from it, so let's take Emacs'.  In my hypothetical
Internationalization documentation, I'd have a separate section for
configuring the keyboard to be able to input the characters relevant for
one's language, so Emacs' "Input Methods" sections are irrelevant.  All
I need is

* Specify Coding::          Various ways to choose which conversion to use.

In the pdf version of the manual, this is a little less than 3 pages.
The invariant sections are:

- Distribution  (2 pages)
- The GNU Manifesto (9 and a bit pages  
- GPL  (7 and a bit pages)
- one page each for front and back cover texts

That makes more than 20 pages of invariant sections, or less than 13% of
interesting material.  Do you agree that the GNU Emacs Manual is non-free?

>  The
> invariant sections with offensive material give us a similar example
> -- documents that contain such invariant section would also be
> non-free.

The GNU manifesto might well be considered offensive by the authorities
of some not-so-hypothetical country.

>> Moreover, I doubt that it would be allowed to structure the text
>> like this:
>>
>> 1. Intro, including explanation of the structure
>> 2. Content
>> 2.1 to 2.12 the individual documents' internationalization docs
>> 3. Legalese
>> 3.1 to 3.12 the individual documents' invariant sections
>
> Yes, this is allowed.  Acording to GFDL "Section numbers or the
> equivalent are not considered part of the section titles" and
> "multiple identical Invariant Sections may be replaced with a single
> copy".

Of course I assume that the documents are all different, from different
authors, with different political or unpolitical opinions to express in
invariant sections.  I don't think that RMS would appreciate if the part
from the Emacs manual would not only come immediately after the one from
the Foo manual, but somewhere 40 pages down his Manifesto would follow
immediately after Michael Foo's "Why GNU is bad Manifesto" with only a
small note saying "End of invariant sections from the Foo manual" and
"Beginning of invariant sections from the Emacs manual".

>> This would make the manual basically unusable.
>
> This would be required only if you are creating "aggregation with
> independent works".  You will have to create such an aggregation only
> if some of your sources are covered under incompatible with GFDL
> license.

That will probably the case.  Moreover, I have the feeling that the GFDL
is incompatible with itself in the sense that I can't have more than one
front cover text.

> But even in that case you may combine your GFDL sources and
> as a result all invariant sections will be grouped in one place.

Maybe you're right and it's not necessary to mix invariant and
interesting sections, but still a ratio of 87% of rubbish is a bit high.
I think this would make it not just inconvenient, but instead non-free.
For example, even copying costs forbid to to distribute 11 sheets of
paper to a group of students if I want to hand them out a 2-pages
condensed version of the above nearly-3-pages section on coding
selection in Emacs.

Regards, Frank
-- 
Frank Küster
Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich
Debian Developer (teTeX)



Reply to: