[Linux-aus] Fwd: Ogg Vorbis and Theora removed from HTML5

Paul Wayper paul.wayper at anu.edu.au
Mon Dec 17 00:01:19 UTC 2007

Jeff Waugh wrote:
> It's the word "additional" that is the key. Apple and Nokia have taken on a
> certain amount of risk by using the formats they support. Let's call that a
> "known risk".
> Xiph have said there are no *known* patent encumberances for both Vorbis and
> Theora [1]. In particular, some patents related to Theora were freed along
> with its release.
> However, for Apple or Nokia to add Theora to their set of supported codecs,
> they are taking on new unknown risk. (snip)

Sorry, but to me it is disingenuous of Apple, Nokia and Microsoft to say
"well, it might have a patent in there somewhere despite all that
searching, we'd better not touch it being the good corporate citizens
that we are".

Take on the one hand Apple and Microsoft.  They both own technologies
that compete with Ogg.  Likewise, I'm pretty sure Nokia has sued
competitors for using its technologies in the past.  So I don't think
any of them can claim to be above the behaviour they're deploring as an
'unknown risk'.

On the other hand, in the past Apple and Microsoft (and quite possibly
Nokia) have lost patent claims against them, Microsoft's loss to the MP3
patent being the most obvious recent example.  They didn't say "well, it
might still be encumbered, we'd better not touch MP3" then, did they? 
Microsoft in particular are famous for 'borrowing' another technology
and then just settling out of court afterward, knowing that they've got
the sales and important industry buzz off the new product way before the
company they ripped off.  It would be a nontrivial but possible search
to find examples of Apple and Nokia borrowing techniques from
competitors.  AFAIK Nokia are part of a group which is trying to
overturn a CSIRO patent on spread-spectrum radio transmissions which
apply to WiFi, for instance.  So these companies are no stranger to
taking a risk on lifting a technology from somewhere else if it gets

And additionally the 'there's no media in Ogg format' argument is
ignoring history.  Adobe didn't write the PDF spec and give away the
reader because there was a huge base of PDF documents out there. 
Widespread use of MP3, GIF, PNG, JPEG and MPEG came because the tools to
convert them were available.  You can also argue that Microsoft pushed
Windows Media because it was supposedly technically superior to MP3 at a
given bit rate and, while they've certainly got a market share by
convincing the content creators to write in their format, MP3 hasn't
gone away either.  So this is really one of those 'build it and they
will come' situations, and the argument that they won't support it
because there's no media available yet is specious at best.

However, I would say that the revision of the draft HTML5 standard is
better after the revision.  OK, it doesn't specify Ogg.  It does,
however, say that whatever is specified in the standard needs to be open
for development, have no licensing cost, and of sufficient quality. 
None of the codecs that Apple, Microsoft or any other company are
pushing fit those criteria.  The closest from my point of view is Vorbis
and Theora.

Essentially, what the system needs is the ability to have some legally
binding process where the standards group says "we're going to use
Vorbis and Theora, if you have any patent claim on these patents you
must say so now, otherwise we're going to use it and you won't be able
to get a cent out of us."  There should be a way to invalidate submarine
patents on specific products.  But patent reform is a separate issue... ;-)

Have fun,


More information about the linux-aus mailing list