Re: parsable copyright format 1.0 (jessie release goals)
On Sun, May 12, 2013 at 2:38 PM, Bart Martens wrote:
> I would regret that the new debian/copyright format would become a jessie
> release goal. The cost/benefit ratio is, in my opinion, very low. It costs
> quite some human time to recode the upstream copyright and license information,
> and I have not yet seen the benefit of more easy parsing by software. The
> licenses in the upstream source code are already in text, so they are already
> in a format suitable for parsing by software. In 2013 I don't see the need for
> humans to recode text to be even more easily parseable by software. Our human
> time would be better spent on developing tools to extract the copyright and
> license information from the upstream sources, in my opinion. If we develop a
> really smart tool, then most debian/copyright can be fully generated
> automatically in a format designed to be well readable by humans.
You may be interested in fossology, which was semi-recently removed from Debian:
http://www.fossology.org/
http://packages.qa.debian.org/f/fossology.html
Another one is Ninka:
https://github.com/dmgerman/ninka
It turns out that automatic license identification is a "hard" problem:
https://lwn.net/Articles/547400/
--
bye,
pabs
http://wiki.debian.org/PaulWise
Reply to: