Re: [rtcweb] H.264 IPR status (Re: VP8 litigation in Germany?)

[Stephan Wenger <[e-mail]>]

Hi,

From: Harald Alvestrand <[e-mail]<mailto:[e-mail]>>
Date: Monday, 11 March, 2013 23:20
To: "[e-mail]<mailto:[e-mail]>" <[e-mail]<mailto:[e-mail]>>
Subject: [rtcweb] H.264 IPR status (Re: VP8 litigation in Germany?)

On 03/12/2013 02:27 AM, [e-mail]<mailto:[e-mail]> wrote:

[…]


I know of 3 sources of IPR information related to H.264:

- The ISO database of disclosures ( http://www.iso.org/iso/standards_development/patents)
- The MPEG-LA list of IPR available for licensing via their license pool
- The list that Cliff Reader did related to H.264 Baseline work in MPEG

The first 2 are public on the Web; I don't know the status of the third one (MPEG doesn't make its input documents public by default).

I know of a fourth one: the ITU patents database.  H.264 is an ITU recommendation, so ITU patent declaration can be found in the ITU database.  Of course, there is a large overlap with the ISO/IEC database.  The IT database, though, is more easily accessible.  See here, and search for H.264 on the right sidebar.  http://www.itu.int/ipr/IPRSearch.aspx?iprtype=PS  Much prettier than the ISO xls spreadsheet with all its reportedly existing typos.  In that xls one, look for ISO/IEC 14496-10.


I think it's fair to ask the proponents of H.264 to present the IPR situation for their candidate codec; saying that it is "well known" is not the same as having an input contribution to the IETF stating what the situation is.

Describing H.264's IPR situation has been done before—more than once--but is easy to repeat.  Parties contributing to H.264 in JVT, and its parent bodies MPEG in VCEG, have to license any H.264 essential patent in their portfolio under reasonable and non-discriminatory terms, on a world-wide bases, often subject to reciprocity.  That includes at least all the patents and/or portfolios listed in the patent databases of ISO/IEC and ITU.  (There may be additional patents or portfolios where the rightholders contributed to JVT, MPEG, or VCEG, but did not make a declaration.  Whether or not those patents are enforceable is questionable.  Undoubtly, RAND terms apply to those as well.)  Further, a large percentage, but not all, of these patents are licensable through a pooling arrangement from MPEG-LA  (see http://www.mpegla.com/main/programs/AVC/Pages/Intro.aspx), at a cost of no more than $0.20 per codec, plus (in some cases) applicable content fees.  Known parties whose patent claims are not licensable through MPEG-LA include Nokia and many semiconductor companies.  Outside of the pooling arrangement, what exactly RAND royalties are is subject to negotiations between licensor and licensee.  There is currently ongoing litigation in a number of legislation which attempts, among other things, to further clarify the meaning of RAND in terms of maximum royalties.   Finally, there may be parties that own essential IPR and have not contributed to the development of H.264, nor are ISO/IEC or ITU members, and those patent claims may not be subject to RAND terms.

Hope I didn't forget anything.

Anyone wants me to characterize the VP8 sitation similarly?  :-)

Stephan