[vlc-devel] [RFC] Bounties for VLC
remi at remlab.net
Tue Aug 14 09:35:20 CEST 2012
Le vendredi 10 août 2012 18:38:59 Jean-Baptiste Kempf, vous avez écrit :
> I have computed a list of bounties from important and/or annoying
> issues/shortcoming of VLC and libVLC, that you can find on our wiki:
There are a lot of problems:
This bounty is never going to be done. It requires a Mac. And I do not know
any Mac developer that cares about Linux and Windows. This needs to be split,
one for Mac, one for Windows and one for UI.
#380, #5178 and #6443 are all client issues. This does not belong in a
paragraph about RTP *packetization*. Furthermore, I do not think we want to
sponsor work on live555.
#6505 is an encoder problem. PCMU and PCMA packetizations work just fine.
This does not belong under RTP.
Raw video is underspecified, since a lot of codecs are raw video.
This has to depend on sout. Sorry, that is a fact, unless you want to fork the
Should I stop working on audio sync? If I continue working, I might either
disrupt the work of a bounty hunter, or help him for free. I want neither.
* DVB scanning
This requires Windows. I have done 50% of the work by reunifying Linux and
Windows. Now I am going to give up on my incomplete DVB scanner since it would
mean I do 95% of the work, and someone else cashes in.
That bounty is counter-productive as it stands.
We already have DTS over HDMI. We have had DTS over S/PDIF for longer than
HDMI has existed.
I think the bounty should be per codec.
As things stand, you cannot implement any of those bounties except MPGA
anyway, because there is no lavc audio conversion filter.
* Lua preferences
This is not feasible. Caching and localization issues are intractable. If a
genius finds a _correct_ solution, s/he'd deserve more than 1200€.
* Lua downloader
This is useless until Lua supports running untrusted which it absolutely does
not ("Told You So" when extensions were merged). And securing Lua is unsuited
for a bounty, since there is no way to verify that the hunter understood the
implication and did the work seriously...
It's a HUGE amount of work. Just about every codec, video output and video
filter needs to be reviewed to pass correct information. Then the video output
core needs to be extended. OpenGL could be split out to a dependent bounty,
but regardless I wouldn't do all of that for 800€.
C/C++ software engineer looking for a job
More information about the vlc-devel