On Tue, 2008-03-25 at 21:32 +0100, Peter Ueger wrote: > 2008/3/25, Francesco Romani <[EMAIL PROTECTED]>: > > We (I) had no interest in implement every possible format/codec in the > > wild; > > Why would you have to do that? Others have already done that for you, > don't they?
The key point is the two projects does NOT have the same objective. I had no interest in reimplement, say theora encoder: the ffmpeg guys do have (and, needless to say, they are quite brilliant in their job). Of course, tc and ff plays on the same playfield (multimedia), but isn't the same game. The focus of the ffmpeg development are the libraries, not the tool(s) (ffmpeg/ffplay/ffserver/stuff); on the other hand, the focus of transcode is the tool, not the libraries. Are the two projects complementary? I don't think so, or at very least they aren't more complementary than, say, libmpeg2 and vlc. ffmpeg will not gain much if transcode becomes the official frontend (I'm rather convinced that most of ffmpeg devs will see this event as a /regression/, not a gain ;) [1]). I'm convinced transcode will not gain much either; we already use ffmpeg libraries extensively! +++ [1] Of course I wholeheartly agree that a significant part of transcode codebase is poorly designed, poorly implemented or both (yes, this includes parts designed/implemented by myself, of course). On the other hand, I'm also convinced that NIH/NWH isn't a problem per se. -- Francesco Romani // Ikitt [ Out of memory. ~ We wish to hold the whole sky, ~ But we never will. ]
