Re: This is libdv devel list. (WAS: Re: [libdv-dev] Possible bug encoding interlaced material?)

[Roman S. <rv...@su...>]

On Mon, Apr 04, 2005 at 07:41:08AM +0200, Stefan Lucke wrote:
> On Montag, 4. April 2005 03:02, Roman Shaposhnick wrote:
> > <F2On Mon, Apr 04, 2005 at 10:14:44AM +0930, Jonathan Woithe wrote:
> > > Anyway, what do others think?  Is this an issue?  Is there a way to get
> > > around it so interlaced material is encoded correctly to DV by libdv?
> >   
> >   Oh, yeah! Use ffmpeg :-)
> 
> Roman, this list should cover libdv development related things.
> To me it is rather annoying as your default answer to questions like:
> Q: I've a problem using libdv with ..
> A: Use ffmpeg ..
  
  Well, I understand your feelings and attachment to the libdv,
  and I can really see that "Oh yeah!" style can be considered
  inappropriate by some. At the same time, all I can really
  offer to mitigate your concerns is to start making these suggestions
  in private without Cc'ing to the mailing list.

  Please, try to understand me as well -- at this point ffmepg's DV codec
  proved itself to be quite stable and in a number of tests it was
  demonstrated that it offers more quality for the encoding.
  I really don't want people to be stuck with libdv if they have a choice
  and if ffmpeg suits them. They might even benefit from switching
  by gaining access to the whole variety of *other* codecs ffmpeg provides.
  
  However, in the FSF/Linux world DV == libdv. 

  Of course, ffmpeg's DV *FORMAT* handler is inferior compared to libdv's.
  Functionality wise that is. But than again, for some folks (like myself)
  quality of the encoder is much more important.
 
  Hope that explains why I'm making these suggestions all the time.

> If you know the source of some problems related to libdv it would
> be more polite to post a fix .

  I know a number of problems with libdv at this moment, however I really
  have no interest in fixing them. That wouldn't be a problem though,
  it there would be someone willing to act on this information.
  However, at this point my understanding is that libdv doesn't really
  have anybody interested in its further development. Correct me if I'm wrong,
  by you and Dan you strike me more like reluctant maintainers. 

  That said, I would be really interested in working on libdv++ or libdv2
  making it easier for me to maintain both codecs at the same time.
  Most likely this will mean leaving API 100% intact but changing
  some of the implementation details quite a lot. If there's an interest
  in me doing this -- please let me know. 

Thanks,
Roman.