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

Re: could you safely rewrite the DFSG requirement?

On Tue, Feb 12, 2002 at 01:48:13AM -0800, Thomas Bushnell, BSG wrote:
> Sven <luther@dpt-info.u-strasbg.fr> writes:
> > What is written is that free distribution of aggregation with software musrt
> > be permitted. And software originally means a significant amount of machine
> > code which can be executed somewhere. 
> Um, no.  It doesn't mean "significant".  Nowhere in DFSG does the word

Mmm, but not being a native english speaker, i have to rely on the existing
dictionaries for such definitions, and the meaningfull part, i got it from one
of the definitions given by the debian distributed dict dictionary databases.
I cannot find it back though, so either it has disappeared since, i am not
looking at the same polace, or i did misinterpret something.

Notice that foldoc says :

     Some claim that {documentation} (both paper and electronic) is
     also software.  Others go further and define software to be
     programs plus documentation though this does not correspond
     with common usage.

This well shows that this is not the standard admitted definition of it.

> "significant" occur.  The following is a piece of software:

This is by no means an empty one liner, is it ?

do you also consider :

10 REM

as a piece of software ?

> And that's really all it takes.  
> > Even if we go as far as including documentation as software, it
> > still needs to be a meaningfull thing that you aggregate, and not an
> > empoty content one liner.
> Really?  DFSG doesn't say that, and it has never been understood to
> say that.  Still, the following was famous, and meaningful:
> main ()
> {
>   printf ("hello world");
> }
> I assume that counts as software.  If we add that to the manual,
> that's an aggregation, and it's even meaningful.  I want to make sure
> that O'Reilly knows this.

Sure, but it is not an empty content one liner. it is the first program
written by manny programmers, it is the introduction to kernigan & ritchies C
book, and as thus has apart from its meaning a historical as well as mythical

> > Also, even if you consider documentation as software, there is
> > another obscure interpretational leap from there to considering
> > printed documentation as software.
> >
> > I know many in the free softare community want it to be such, but this is by
> > no means a granted interpretation.
> People are of two minds about this.  Some think that documentation
> should be a little looser.  I think that that crowd (which once
> included me) is losing, as far as Debian is concerned.

Yes, but this is not a clear interpetation of the DFSG, and i only recently
discovered, trough the foldoc software entry, that people could consider
printed book as software, which is contrary to what my first thinking on this
subject yielded.

> Debian distributes only free stuff; that's the point of Debian.  If
> you think the DFSG doesn't apply *at all* to documentation (which is
> what you seem to be saying), then what should apply?  It seems to me
> that you are not asking for some interpretation of the DFSG, but some
> new thing entirely.

Mmm, my first thought when i read the oreilly licence was that iut had to go
into non free, then i read the polemic about the aggregation stuff, and looked
at the DFSG, and thought, ok it is far stretched, but it may seem that it can
be interpreted as such. Then came the proposition of the empty content one
liner, and that is something i feel is bad.

> However, what (exactly) is your point?  Why don't you spell out a
> concrete, specific proposal, instead of arguing against the
> all-but-universal understanding of the DFSG?

My point is that the DFSG is not clear on the subject of documentation, and
particularly printed version thereof. It need to either be ammended, as you
say, or extended by a DFDG speaking porecisely about documentation.

The response to it from this list, which in my understanding is the place to
discuss legal things like thge DFSG, but apparently i am wrong, was of the
'you are wrong, please go away" kind.

> > Again, that is what you and other people on debian legal say, but it is by no
> > means clearly written in the DFSG.
> What is clearly written is that it must permit distribution as part of
> an aggregate.  I want to make sure that O'Reilly understands that
> "meaningful" and "significant" and "important" are not any part of
> that term... just aggregate, pure and simple.
> > Well, yes, but the Sun OS is by no means an empty one liner, is it ?
> The license as actually written prohibits distribution as part of Sun
> OS-- which is enough to make it not DSFG free.  In any case, there is
> no question here of "empty" one liners.

huh, i don't follow you here ?

And yes, the licence is non free because of the contamination clause.

> If I wanted to republish an O'Reilly book, I would add the "why free
> software needs free documentation" essay, and consider that an
> aggregate.  By no means is that an irrelevant empty addition.


> I want them to explicitly acknowledge that such an addition--even if
> only one page--is enough to trigger the "aggregate" clause.

but the polemic is around an empty content one liner, not a only one page

> > > The point is that the DFSG requires that free distribution as part of
> > > an aggregation be permitted.  It does not allow *any* restriction of
> > > this, and thus, even a "no content one liner" aggregation is an
> > > aggregation, and free distribution of this must be permitted,
> > > according to the DFSG.
> > 
> > No, see my reply to this at the beginging of the mail.
> > 
> > This is just your interpretation, not what the DFSG says.
> What it says is:
> "The license of a Debian component may not restrict any party from
> selling or givinrg away the software as a component of an aggregate
> software distribution containing programs from several different
  ---1----                         ---2----
> sources."  

Here you take the interpretation that software includes printed versions
of books, and not only their electronic version. This is contrary to common
usage, and needs clarification.

> The single line 
> is such a program; add "several" and we're done.  One page of text,
> and that's it, satisfying the curious restrictive clause in DFSG 1.
> That's what it says.  All this talk of "significant" and "meaningful"
> and "important" are words that you have invented that are not found in
> the DFSG--which requires even trivial, unimportant, and insignificant
> cases.

but not empty one liners.

Mmm, as i can't find again the place were i found the definition of software
containing these, i guess i have to retract this one for now.

That said, the problem remains, we keep relying on something unclear and
subject to interpretations instead of sayign things clearly.

This is called transpearency in politics, and is somthing we may well be
advised to apply here also, to stop any forthcoming polemic.

> > Because you interpret it as what you wish it to say, and since the opinion of
> > the debian-legal folk is mostly the decisive opinion on this matter, it
> > doesn't matter much if it really say so, or say otherwise, does it ?
> Ah, but indeed, we are reading the license.  Your opinion counts, but
> it's only one opinion, and so far *everyone* disagrees with you. 

Err, everyone on debian-legal who read this thread, or more probably only
those who carred enough to reply, what are those compared to 500+ debian
developpers, or to all those which read the DFSG to see what is considered
free software.

> But, of course, feel free to take it to the appropriate forum if you
> want to amend the DFSG to restrict the meaning of "aggregation".  This
> is not that forum.

Either you speak about aggregation, in the intention of meaning it as a true
aggregation, or you don't, and in this case, why not simply state that a book
or other free documentation must be printable and distributable alone, and not
go into obscure interpretations of what an aggregation means ?

And altough the word may permit (or allow?) the aggregation of nothing to the
book, i don't think this is the spirit of what was written.

As i see DFSG 1, we require that it is possible to distribute the software as
part of the debian archive, debian CDs or such. 
If this is not the case, and since you followed this since the begining,
please tell me what was really meant by this aggregation clause.

Ok, i know you don't want me here anymore, but i guess this thread has made
things more clear to me.


Sven Luther

Reply to: