[theora] Some questions about Theora IP

Maciej Stachowiak mjs at apple.com
Sat Jul 4 07:09:24 PDT 2009

Hello Theora developers,

I'm doing some cursory research into Theora's IP status in preparation  
for asking Apple to reconsider the possibility of shipping an  
implementation. I have a few questions and I'm hoping knowledgeable   
people can help out.

1) What are the terms of any patent licenses or disclaimers, and do  
they have field of use restrictions or limitations on code for which  
the patents are licensed?

I found the following patent license in the original VP3.2 source: <http://svn.xiph.org/branches/vp32/vp32/VP32_license.txt 
 >. This appears to have a sort of field-of-use restriction on the  
patent license terms; products that don't support the original VP3  
bitstream do not get the patent grant. The patent license is also  
limited to the VP3.2 code itself and derivative works thereof.

I also found this promise of patent non-assertion: <https://svn.xiph.org/trunk/theora/LICENSE 
 >. This doesn't have a field-of-use restriction, but it does appear  
to limit the promise not to assert to the original VP3 code and  
modifications to that code.

While looking for info, I found this thread from 2004 asking if On2's  
patents on Theora were licensed for independent implementations: <http://lists.xiph.org/pipermail/theora/2004-July/000544.html 
 >. There didn't seem to be a clear conclusion at the time.

Does anyone have further info on this? Are there additional agreements  
between On2 and Xiph besides the above documents?

Note: I expect that in practice, On2 isn't looking to stop independent  
implementations, and perhaps didn't intend the limitation on the  
patent disclaimer at all. However, companies have been known to  
suddenly become more litigious (for example if they get axquired), and  
in such a case I would expect a literal reading of the patent license  
and the promise not to assert. It also seems important for the patent  
status to be really clear for purposes of the W3C Patent Policy, if  
Theora were to be required by HTML5 or other specifications. I'm not  
sure a royalty-free patent license that doesn't cover independent  
implementations would be good enough. Perhaps whoever did the  
negotiating with On2 could get them to publish a clearer statement.

2) Is there any public information about what patents On2 holds that  
may read on Theora? (This information isn't essential but I couldn't  
find it stated anywhere.)

3) Has anyone done any research to determine if there might be patents  
that read on Theora held by parties other than On2? I know this is a  
sensitive topic, and anyone who has that kind of info might not be  
able to share it. But if anyone has done even partial research, that  
would be very useful. I'm wondering for example if anyone has done  
some of the more obvious obvious checks, like studying the patents  
that are known to apply to MPEG-2 Video,  MPEG-4 Part 2, or H.264/AVC,  
since Theora has a roughly similar basic design.

4) Has On2 ever said anything about whether parties besides them hold  
patents on VP3?


More information about the theora mailing list