[whatwg] Google's use of FFmpeg in Chromium and Chrome

Daniel Berlin dannyb at google.com
Sat Jun 6 14:20:51 PDT 2009


On Sat, Jun 6, 2009 at 5:00 PM, King InuYasha<ngompa13 at gmail.com> wrote:
> On Sat, Jun 6, 2009 at 3:47 PM, Daniel Berlin <dannyb at google.com> wrote:
>>
>> On Sat, Jun 6, 2009 at 4:35 PM, Håkon Wium Lie<howcome at opera.com> wrote:
>> > Also sprach Daniel Berlin:
>> >
>> >  > >>> "For example, if a patent license would not permit royalty-free
>> >  > >>> redistribution of the Library by all those who receive copies
>> > directly
>> >  > >>> or indirectly through you, then the only way you could satisfy
>> > both it
>> >  > >>> and this License would be to refrain entirely from distribution
>> > of the
>> >  > >>> Library."
>> >
>> >  > Note that the actual *clause* (not the example) in question says
>> >  > "If you cannot distribute so as to satisfy simultaneously your
>> >  > obligations under this License and any other pertinent obligations,
>> >  > then as a consequence you may not distribute the Library at all. "
>> >  > It then gives the patent example as an example of when you could not
>> >  > fulfill your obligations under the license.  The restrictive license
>> >  > in the example falls afoul of this condition (part of #10): "You may
>> >  > not impose any further restrictions on the recipients' exercise of
>> > the
>> >  > rights granted herein."  Nothing in any licenses we have with other
>> >  > parties imposes any *further restrictions* on the recipients who get
>> >  > ffmpeg from us.  You get *exactly* the same set of rights and
>> >  > obligations we got from ffmpeg.
>> >  > As such, we can simultaneously satisfy our obligations under this
>> >  > license (which again does not require us to pass along patent rights
>> >  > we may have obtained elsewhere, it only requires we grant you the
>> >  > rights you find in terms 0-16 and place no further restrictions on
>> >  > you) and any patent licenses we may have, and do not run afoul of
>> > this
>> >  > clause.
>> >
>> > I get parsing errors in my brain when reading this. While I understand
>> > that you do not impose any new restrictions (as per #10), I still
>> > don't understand how you can claim that #11 (the first two quotes
>> > above) has no relevance in your case. To me, it seems that the example
>> > in #11 (the first quote) matches this case exactly -- assuming that
>> > Google has a patent license that does not permit royalty-free
>> > redistribution.
>> As i've said in other messages, this example doesn't match this case
>> at all, since the patent license was not given to us by the same
>> people who gave us the library, *and* our patent license doesn't even
>> say anything about the library used to do encoding/decoding. I.E. Our
>> patent license has 0 to say about our distribution of ffmpeg, only
>> something to say about our distribution of Chrome, which is only
>> covered by section 6 of the LGPL 2.1 (which allows distribution under
>> whatever terms we choose so long as we meet certain requirements,
>> which we do).
>>
>> > I also understand that the LGPL doesn't explicitly "require [anyone]
>> > to pass along patent rights we may have obtained elsewhere". However,
>> > it seems quite clear that the intention of #11 is to say that you
>> > cannot redistribute the code unless you do exactly that.
>> > What am I missing?
>> >
>> That our patent license does not restrict/grant/say anything about
>> ffmpeg, only Google Chrome, and Google Chrome itself doesn't fall
>> under the LGPL 2.1 except through section 6.
>>
>> --Dan
>
>
> So are you saying you DO have a patent license for ffmpeg and Chrome? Or
> don't you?
Why are you conflating both of them together?
Our patent license covers user use of Chrome so they can play H.264/AVC/etc.
It says nothing about ffmpeg.  Zilch. Zero.
If you were to replace ffmpeg in Chrome with another decoder, our
patent license should still cover your use (I would have to double
check this, but this is my recollection).

--Dan



More information about the whatwg mailing list