Re: Modification of reference material
On Thu, Apr 15, 1999 at 11:33:13AM +0000, Arnoud 'Galactus' Engelfriet wrote:
>
> The essential difference between software and documentation IMO is that for
> the former, it's more important what it does than what it has internally. If
> I can rewrite your code so it does the function better or safer, there
> shouldn't be a problem. If I rewrite the code and make it more buggy, people
> will notice and stop using my version. But with documentation, it's very hard
> to distinguish between the good and the bad changes.
Well, you can surely make good and bad changes to internal code, too,
without chnaging the functionality too much (for example, youc an write less
portable code etc). However, I can only repeat that I understand your
concern. Read on.
> In our case, the reference (on HTML and CSS) will be read by people to see
> what is and what isn't permitted, and which elements and attributes you
> should use in which case. If someone else adds wrong information, people will
> not be able to see this is wrong because they trust our reference to be
> correct (based on the reputation the WDG has in the Web authoring community).
So the need is that people will NOT get the impression that it is your
original work, but a work based on it (you want attribution for the good
stuff you did, but not for the bad stuff other people added).
[...]
> >+ change the name.
> >+ remove a non-technical introductory chapter, section or paragraph.
> >+ clearly document which changes were made.
>
> I'm not sure how to apply these requirements to our case. If someone modifies
> a file in the reference, then he can document that in a changelog, but few
> people will check that when they read the file. Indicating that in the file
> is possible, but I'm afraid that won't make the reference any more readable.
How about documenting it in a changelog, and marking the file explicitely as
changed. I assume you have an attribution to yourself on every page. Then
you can require that it is changed to something: "This is a derived work
from... WDG (hope I got this acronym right :) is NOT responsible for errors
that are introduced by the changes." or similar.
The idea is that as soon as people start to wonder what this document is,
they will notice that it is a derived work, and not your original. So they
should plester the bad people who did the changes, and not you.
> If I remember correctly, the Mozilla license allows people to modify the
> source code of the free Netscape version, but only the people at Netscape may
> distribute the program as "Netscape." Similarly, the TeX typesetting program
> is public domain, but may only be distributed as TeX if the program passes a
> certain test. Are these licenses considered free? Then perhaps a license like
> Netscape's would be appropriate for this type of documentation.
Yes, those are free in the sense of the DFSG, and we arer distributing them.
We allow certain kinds of restrictions, and we understand the need for them.
For example, you can add the requirement that it passes a HTML validator of
your choice. For all derived works which do not meet your requirements, you
can "punish" with name and version changes, marking changes very easy
spotable etc. If you like, we can work out further details on this list.
Of course you can also say only the very original, unmodified version is
allowed to carry the official name. As long as some form of derivation is
possible, it is very likely dfsg free (although there are some minor
caveeats to avoid. Nothing too serious, though).
Thanks,
Marcus
--
`Rhubarb is no Egyptian god.' Debian http://www.debian.org finger brinkmd@
Marcus Brinkmann GNU http://www.gnu.org master.debian.org
Marcus.Brinkmann@ruhr-uni-bochum.de for public PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/ PGP Key ID 36E7CD09
Reply to: