From: Patent Tactics, George Brock-Nannestad
Oh, dear, now he comments on himself !!
> With an audio file it is no great thing to automatically flag changes of
> quality according to pre-defined criteria, so possibly the actual listening
> may be limited to 30 seconds before and after a change, in order to
> the next item. The more there is a need for this kind of processing, the
> easier it will be available.
----- Well, I just thought I would add that the automatic flagging of
changing sound would probably require a linear, non-data-reduced file,
because such a system would rely on a change in spectral content.
I have put forward the heretic view in the
> International Association for Sound and Audiovisual Archives - IASA that a
> trace of the sound is better than no sound at all. Which would mean that data-
> reduced versions are better than no versions at all. And, honestly, they do
> not take up much space in that form.
----- well, that is precisely what makes MP3 files tick. But for a record
company to claim that the MP3 file is a rip-off of the original sound is
technically erroneous - it is only a fraction of the sound.