liblicense [creative commons work]

Kevin Ottens ervin at kde.org
Thu Aug 30 17:08:43 BST 2007


Le jeudi 30 août 2007, Rafael Fernández López a écrit :
> As well, with the file manager you will be able to set the license to
> more than one file at a time.

Good point.

> Of course specific applications can 
> benefit from this too.

Yeah, I might have been unclear, but IMO it'll shine for those apps.

> > I'm pondering about this, at first it looked like a tempting idea, I
> > thought "ok, one more C lib to wrap... no big deal". But the more I look
> > at how it's done the more I wonder if it's in our best interest.
> >
> > My current thinking on this is that we shouldn't use this lib, but share
> > the RDF schema and data. Because what liblicense basically do (AFAIU) is
> > 1) read/write tags from file to set their license, and 2) give
> > information on a particular license based on the RDF information
> > provided.
>
> Hmm, probably you're right... but the easier thing right now to avoid
> more work is to add liblicense on kdesupport if it is not packaged on
> distributions when 4.1 arrives. If distributions package it, then
> we're done, we just depend on it. We only have to add the part related
> to KDE (kcm's...).

Ah ok, well if you're going to provide subpar integration (basically only the 
bits for the file manager) sure that's enough. If you want to integrate it 
better if more than "we're done, we just depend on it". ;-)

> > For 1) we already have the infrastructure in place thanks to Strigi
> > analyzers and KFileWritePlugin. So we'd basically duplicate a specialized
> > subset of our stack only for license while our framework is already able
> > to do it.
> >
> > About 2) we already have Nepomuk to access RDF information in a nice way.
> > Since they already have all the RDF schemas and RDF files needed for
> > this, I think it shouldn't be hard to feed Nepomuk this kind of
> > information.
> >
> > Both combined, that'd mean we'd have the whole data indexed for free,
> > while using liblicense would require extra work (by at least patching our
> > strigi analyzers).
>
> Yeah, but that's more work for us. I thought of that and for that 
> reason I want to move all this for 4.1.

Well, if you want to do it properly it'll require to patch our analyzers 
anyway. Actually I think that in the long run it's less work for us than 
using liblicense thanks to the benefits of using the RDF based facilities we 
already have (indexing for free, and neat api included). Otherwise, you patch 
the analyzers, then you have to make sure the info is available correctly 
through Nepomuk, and finally you have to make the wrapper API of the relevant 
liblicense parts for the applications to use. Looks like more work to me.

> We also can see in the 
> meantime if distributions want to package liblicense or not.

Well, I hope the rdf files will be shipped separately because that's really 
the part we need. Worst case if it's not shipped by distributions we'll be 
able to ship a private copy in kdelibs I guess.

Regards.
-- 
Kévin 'ervin' Ottens, http://ervin.ipsquad.net
"Ni le maître sans disciple, Ni le disciple sans maître,
Ne font reculer l'ignorance."
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20070830/e15edb58/attachment.sig>


More information about the kde-core-devel mailing list