Re: Bug#239952: kernel-source-2.6.4: qla2xxx contains non-freefirmware
- To: firstname.lastname@example.org
- Subject: Re: Bug#239952: kernel-source-2.6.4: qla2xxx contains non-freefirmware
- From: Nathanael Nerode <email@example.com>
- Date: Sun, 04 Apr 2004 06:48:11 -0400
- Message-id: <[🔎] firstname.lastname@example.org>
- References: <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com>
Henning Makholm wrote:
> Scripsit Walter Landry <firstname.lastname@example.org>
>> Henning Makholm <email@example.com> wrote:
>> > A side question: Does -legal think that it would be OK for Debian to
>> > distribute *checksums* of the non-free magic blob-of-bits, such that
>> > it can be located inside a Windows driver on a disk without needing to
>> > know its exact position?
>> A SHA1 checksum is 160 bits or 20 bytes. This sentence is longer than
>> that. I would say that it isn't copyrightable,
> I agree about that, of course. I was imagining that someone might
> complain that we woundn't be providing "source" for those 20 bytes,
> but it seems I wasn't communicating that clearly enough.
Heh. Let's think about this.
You can replace it with a different, modified checksum based on a different
magic blob; would anyone ever want to do anything else with the *checksum*?
That seems sufficient to be the "preferred form for modification" from my
point of view (as long as the checksum generator is free software). So
yes, I think the checksum constitutes its own source for DFSG purposes.
Make sure your vote will count.