[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: mplayer 1.0pre7



On Mon, Apr 25, 2005 at 04:34:41PM +0200, A Mennucc wrote:
> mplayer 1.0pre7 is ready and packaged at
> http://tonelli.sns.it/pub/mplayer/sarge
> 
> a.
> 
> ps: still no news from ftpmasters... hope they at least will try to read
>   http://people.debian.org/~mjr/mplayer.html

Right, so as an mplayer user and having an interest in its inclusion, I
took a look. Note that I'm a member of the FTP team, and occasionally do
NEW processing, but the large majority of it has been done by Joerg
Jaspert.

First thing I noticed was that it's about 700.000 lines of sourse code.
That's a lot. Of the potential issues, I'll give you my own, personal
opinion, which might or might not be shared by all ftp-masters.

- Copyright: I believe consensus is that this case is settled, thanks
to the great work by the numerous people involved

- Packaging stuff: Of course there are always nitpicks, but IMHO those
are not a consideration for whether or not to accept the package, not
in the last place because of the long time it's already in the queue.
Particular issues that itch the ftp-master that would approve the
package are IMHO best done with filing a bug after accepting, I didn't
see any serious issues in any case.

- Patents: The big issue with mplayer a.t.m. I'm myself not very
following the patent stuff, but as far as I understood, certain
patents hold by the MPEG organisation, esp. those w.r.t. encoding of
MPEG data streams, are actively being enforced, (again afaik) in the
United States in particular. See [1] for more information of what I
believe is relevant here. Unfortunately, links there mostly either
shine in unavailability (404 etc) or utter vagueness and
non-information (I couldn't find any bit of useful patenting
information at [2], for example). The FFII had more useful information
at [3].

All this seems to concentrate on MPEG-related *encoding* though, and
not to decoding. Moreover, Debian contains plenty of MPEG-related
decoding software, and the FTP-master policy at least w.r.t. audio
MPEG decoding has always been to not let supposed patents in this area
stand in the way of distributing this software, on the basis that it
seems to be an unenforceable patent, or at least, it isn't enforced
(and giving in to any patent would mean Debian could not distribute
anything). I see no reason why MPEG videa decoding would be different in
this respect, again, to the best of my knowledge.

So, adding these two tentative[4] conclusions together, it seems
likely that if mplayer were demonstrated with reasonable certainty to be
free of MPEG-encoding code, it would be acceptable for inclusion in
main as far as the FTP-masters are concerned (note: We're not (yet?)
saying it's *required* to strip MPEG encoding stuff, but in my personal
opinion, it seems likely that this is what it'll turn out to be. Don't
take my words on too much value though, maybe stripping this won't be
required after all, but in any case, if it isn't there, we don't need to
think/discuss about it -- reinclusion of the encoding stuff can then
later separately be discussed).

I must mention one big 'but' though: as mentioned above, patent stuff
isn't my expertise, and I could easily have missed a patent (or other)
issue.  MPlayer is definitely a hairy subject, unfortunately, and that's
the reason for the delay in processing it[5], it requires careful
research and reasonable deal of attention to boring patent stuff.

I hope this helps,
--Jeroen

[1] http://www.mpeg.org/MPEG/starting-points.html#ipr
[2] http://www.licensing.philips.com/information/mpeg/
[3] http://swpat.ffii.org/patents/effects/mpeg/index.en.html
[4] Barring mistakes in my reasoning
[5] The 'suboptimal' communication has other reasons that will no doubt
    be talked about in the next FTP-master flame[6], so I'm not inclined
	to comment on that
[6] I've not yet been informed about its ETA

-- 
Jeroen van Wolffelaar
Jeroen@wolffelaar.nl (also for Jabber & MSN; ICQ: 33944357)
http://Jeroen.A-Eskwadraat.nl



Reply to: