W3C home > Mailing lists > Public > whatwg@whatwg.org > December 2007

[whatwg] Xiph.Org Statement Regarding the HTML5 Draft and the Ogg Codec Set

From: Christopher <xiphmont@gmail.com>
Date: Thu, 13 Dec 2007 05:51:04 -0500
Message-ID: <806dafc20712130251h7e5e044fx6ba5c0a6bb6af8c6@mail.gmail.com>
On Dec 13, 2007 5:32 AM, Maciej Stachowiak <mjs at apple.com> wrote:

> It's unfortunate that this press release conflates Ogg, Vorbis and
> Theora.

Although it is a point of nerd pride to correct 'Ogg is just the
container, the codecs are....', that truth depends on context.  Ogg is
and always was the name of the whole project.  There is nothing wrong
with referring to the total code (and functionality) set as Ogg.  The
fact that the codecs' internal names became well known was originally
an accident.

(Not meant as smackdown, just clairification)

> They do not have equivalent deployment, testing and review
> status (or for that matter technical quality)

If you consider Theora to be a flavor of VP3 [which it is] they do
actually have similar deployment and testing track records.  VP3
hasn't been open for as long, but it has been fully exposed to the sun
for several years and is actually a bit older than Vorbis.

VP3/Theora is a much simpler, much lighter weight codec than the
current 'state of the art' and yet performs very well.  The current
reference encoder has quite a lot of headroom for improvement, but
that is more a function of technical advance since it was written than
a statement against its quality or reliability (many things that were
state of the art ten years ago seem stupid or naieve now).  h.264 will
continue improving for longer than Theora. But Theora has room to
close with where h.264 is today, and it will do it at a fraction of
the computational complexity.

I'll mention that even the original VP3/Theora survives extended
fuzzing very nicely, which speaks to its reliability track record.

Monty
Received on Thursday, 13 December 2007 02:51:04 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 16:58:59 UTC