--nextPart1774734.ILGjHAnFbq
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

This is how I could imagine Rosetta being useful for KDE (and other softwar=
e)

=2D Rosetta would have a list of maintainers of files. For example it would=
 know=20
that Joe Translator maintains the chinese translation of kdelibs.
=2D Rosetta would use KDE svn-repo to get its .po-files.
=2D A Rosetta user wants to translate kdelibs via Rosetta. R. gives him the=
=20
files which are not yet 100% finished. The translator translates using R. a=
nd=20
when done, click on "Finished" or whatever.
=2D Rosetta sends the diff (or the whole file) to Joe Translator who commit=
s the=20
file to KDE's svn (or not, if the quality is not good enough).
=2D Rosetta syncs its database ever 6hours or something like that so that t=
here=20
are no clashes.
=2D When Rosetta send the mail out it "locks" the file until Joe Translator=
=20
commits or rejects. This rejection could be done by an email or by a backli=
nk=20
in the mail to Joe Translator ("click here to reject the translation" or=20
something)

The point of all this is that all translation is inside KDE and there is on=
e=20
guy inside KDE responsible for the quality-check. This is pretty much like=
=20
right now: Somebody translates with KBabel and sends the maintainer the fil=
e.=20
Only the maintainer can commit.

What do you think?

Carsten

--nextPart1774734.ILGjHAnFbq
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEeXqC29GaGyAowFcRAgCkAKCXmibTuYBfg7kWjEp3HJRL3jBIxACeNPyN
5m4eIue5qyjVmz1aRjHP0Qc=
=uOCa
-----END PGP SIGNATURE-----

--nextPart1774734.ILGjHAnFbq--