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

Re: Boilerplate License Revision Proposal



on 28 Jul 2000 22:04-48 Guylhem Aznar wrote (and Stallman quoted):
>     If you guys think it's ok, that would be the best idea :
>      - non free : any license permitting that we redistribute freely the
>        document (ex opl a+b)
>      - ldp : gfdl or opl -a -b only
> 

On Sun, Jul 30, 2000 at 12:47:54AM -0600, Richard Stallman wrote:
> I think that segregating the existing non-free documents would be a
> very good idea.  You could continue to distribute them, but those
> of us for whom they are unacceptable would be able to shun them.
> Meanwhile, it would give new contributors a very clear message
> to make their work free.

I think that the existing policy is not so bad.  It means that any
document we have may be freely copied and distributed in any medium.
It thus rejects opl b and suggests (but doesn't require) avoiding opl
a.  Note that Guylhem Aznar seems to be proposing that we accept
documents that can't be freely distributed in print form, but
segregate them into a separate directory tree, etc.  Richard Stallman
suggests that we only segregate *existing* documents which implies
that from now on we either keep the existing policy or reject new
material which doesn't permit unrestricted modification.

As it is now, anyone may take any (or all) LDP documents and freely
copy and distribute them in any media.  Thus anyone who wants to
distribute (or print) a number of our documents doesn't need to check
the license on each one.  But if one wants to modify a LDP document,
then one needs to check the license to see if the author allows this.
It's not much of a burden since usually a person would only be
interested in modifying one document and doesn't need to check the
licenses on a large number of documents.

A major defect of the existing policy is the case where a document is
not maintained, but prohibits modification without permission of the
author.  We wind up with obsolete documents that sometimes can't be
updated.  I think (in theory) that we need to change our policy of
what we will accept and from now on not accept documents that can't be
modified if the author fails to maintain them to reflect new software
(and new hardware in some cases).

The problem here lies in where to draw the line as to what
restrictions an author may place on modification.  Richard Stallman
would say the the author should have no say in the matter and doesn't
even need to be notified about a modification by someone else who then
puts the modified doc on the Internet.  Our current policy gives the
author full control over any modifications (if they choose to exert
such control).  Both of these extremes are simple to state.  I favor a
policy for LDP somewhere in between these two and it's not simple to
state.

I think (in theory) that a license should at least allow distributing
a modification if the original author failed to adequately maintain
the document.  Who is to judge what "adequately maintain" means?  We
can't afford the cost of adjudicating such disputes in a public court
of law and should have our own internal means of resolving such issues
(although if we were unable to resolve them they might then need to go
to court).  But we now have no such means except the public discussion
group to which people can participate even in cases where there is a
conflict of interest.  It's not a very good tribunal to resolve such
disputes.

Thus I think as a practical matter that we should keep the existing
policy for now (although in "theory" I think otherwise).  What we need
is one free license which we can suggest (but not require): boilerplate,
LDP License, GPL, OPL, or GFDL.  I think that all of these licenses
will need revision before we can use it.  I'm looking now at OPL and
it has an ambiguous requirement for putting the name of the original
publisher on a book cover in print at least as large as the book's
title.  This needs to be changed.

I would argue that regarding benefits to the author, there is not a
great deal to be gained in selling the exclusive right to publish if
such work is first (weeks in advance of the work appearing in
bookstores) put on the Internet at the 200 LDP mirror sites.  The
electronic media has a big advantage here since it appears first and
is free in price.  I think that many people who buy such a work may
not realize that they can get it free.  Another advantage of getting
it via the Internet is that one learns how to get free updates and
some documents are updated monthly (or so).  As time goes by, the
share of the market by electronic media should grow and that of print
media decline.  Thus I don't see the possibility of making a living by
writing such documents and selling them for printing (while
distributing them free electronically in advance of printing).  The
fact that one can get the same info for free tends to reduce the price
of such books and also reduces author royalties.

Thus I don't think Poet's proposal would bring authors much financial
benefit and it would go against our principles of free documentation.
Anyway, authors are now free to grant exclusive printing rights and
distribute the same free electronically.  They just don't put it on
the LDP sites.  However, I think we should link to the sites where
they do distribute their half-free docs.  Do any such sites exist?

			David Lawyer


--  
To UNSUBSCRIBE, email to ldp-discuss-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org