On Mon, 14 Mar 2005 17:53:35 +0000 Gervase Markham wrote: [about the "don't remove get_source feature"] > - The requirement to not remove certain particular code is probably > non-free;
I don't think it's forbidding to remove the code: it's merely forbidding
to drop a feature.
You could reimplement it in a better (or even worse) way, but you must
support it.
Anyway I agree it's non-free.
Suppose for example that my derivative work is intended for use on an
embedded system with very limited hardware resources.
I could fail to comply with my constraints, due to this prohibition to
drop a functionality (in the meanwhile, perhaps, I'm distributing my
derivative work separately, through my website, in both source and
binary forms and even through Debian archives, since I'm a DD myself and
have packaged my derivative work for Debian! Thus I'm a very good guy!).
Obviously, this is a thoroughly hypothetical example (I don't write
programs for embedded systems, IANADD, and, above all things, I wouldn't
create derivative works of AGPL'd programs!)
>
> - The general requirement to make code available for download could be
> asserted without the "don't remove code X" clause;
Yes, though I don't think such clauses could be made DFSG-free... :(
>
> - Specifying HTTP is not future-proof, and may not be appropriate for
> some programs or environments;
Definitely agreed.
>
> - What happens if the program interacts with other programs over a
> network? How do you define "interacting with a user"?
Who knows?
I agree that this is very gray and unclear.
--
Today is the tomorrow you worried about yesterday.
......................................................................
Francesco Poli GnuPG Key ID = DD6DFCF4
Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4
pgproudW5yfWV.pgp
Description: PGP signature
