On Thu, Nov 28, 2013 at 2:40 PM, Dave Crocker <dhc(_at_)dcrocker(_dot_)net>
wrote:
BTW, as distasteful as it might be, is there a reason that making /both/
MTI would not work?
Speaking as a third party to this, so I may have misinterpreted, then yes.
My (possibly simplistic and/or plain wrong) summary follows:
The problem appears to be largely driven by actual IPR issues surrounding
H.264, though it has strong hardware support particularly within the
incumbent VOIP market players.
My impression is that VP8 is largely (though not entirely) thought to be
free from IPR headaches, but lacks the hardware support that is baked into
the market. [I have seen exchanges suggesting that other people suspect VP8
of having IPR issues, but nobody I've seen in the posts I've reviewed has
claimed that position for themselves, so it's not clear to me how IPR-free
it's really perceived]
It's possible to make H.264 an MTI only if you're willing to ignore the
"true" open-source browsers (by which I mean IceWeasel rather than Firefox,
and Chromium rather than Chrome) - Cisco have somewhat mitigated the IPR
issues with their OpenH264 effort, though the precise licensing details
don't align fully with open source, and it's clear they will cause
significant headaches to at least some parties.
It's possible to use VP8 as an MTI only if you're willing to turn a blind
eye to the phenomenal deployed running silicon. (To gain some idea of the
level of investment in the silicon, it's worth thinking about how much
actual cash Cisco are committing to in MPEG-LA fees to cover their
OpenH264).
The problem is simply that both codecs on the table as front-runners have
significant and insoluble downsides for non-intersecting portions of the
market. Both sides are negotiating in good faith, and both sides' concerns
are real and valid.
To make matters more complex, it appears that the encoding side may have
entirely different IPR issues than the decoder for video in general.
The whole situation is undoubtedly a mire of complexity, and I don't for a
minute envy the chairs - FWIW, when XEP-0266 and XEP-0299 were produced, I
was serving on the XMPP Council (read: IESG/WG Chair analogue), and so I've
more than a little sympathy, and no few scars.
Dave.