Comments on: royalty free codecs at the IETF

By: Franck Martin

Franck Martin — Tue, 04 Aug 2009 01:47:30 +0000

IETF is all about passing the barriers and joining the group. It is a resistance path. It makes great specifications because blood(?!) and sweat has gone into them.

Keep steady, you’ll get there.

By: Monty

Monty — Mon, 03 Aug 2009 18:31:25 +0000

oops; looks ike angle bracketsdon’t get escaped. ‘foo; in “I don’t work for ‘foo’” above is missing the foo because I put it in brackets….

By: Monty

Monty — Mon, 03 Aug 2009 18:29:40 +0000

That… was an exhausting conference.

To answer a few questions, Ericsson had 39 registered employees attending, but that didn’t count independent contractors that work for the company in reality but aren’t official employees. “I don’t work for , I am from a little company” is apparently a whole category of running joke at the IETF.

Also, there is no formal vorting, but there are the ‘hum votes’ as mentioned. They’re not binding, but they do matter. In our the telcos in attendance hummed as a block unanimously against and everyone was fully aware that would happen. The truly damning thing would have been had they hummed unanimously against and there had not been sufficient perceived support for the WG.

As it was, we obviously lost the hum vote, but by a surprisingly small margin; things were solidly against but not overwhelmingly so. We made our case, stood our ground during questioning, and now things go back to the mailing lists to address at least one obvious mistake in preparation— I intend to blog about this myself in a great deal more detail in a day or two.

There was no quick victory as we hoped, and that was an urealistic hope, but we came surprisingly close. In every way our ‘failure’ moved us forward and we may even be close enough that we’ll not need to re-BoF in Hiroshima (IETF76). Hope hope, these trips are exhausting.

By: Mardeg

Mardeg — Thu, 30 Jul 2009 02:35:27 +0000

What would be good is if Xiph actually released Speex 1.2 final or all OSes, since the latest RC1 has been at http://downloads.xiph.org/releases/speex/ since 24-Jul-2008

By: @F1LT3R

@F1LT3R — Thu, 30 Jul 2009 00:47:39 +0000

Such a breath of fresh air to hear this is happening, thanks for the info Chris! Two big names like these with the skill and the scalability to make something happen is a great hope for the future of the Internet.

By: Franck Martin

Franck Martin — Thu, 30 Jul 2009 00:37:35 +0000

It is interesting. If I recall the SIP specification attributes a few parameters to some standard codec (like g723) and allows extension for other codecs (like iLBC). I suspect IANA could start a list of reference numbers and codec to be used in various protocols.
Also IETF has not an obligation of opensource, but requires all people writing RFC to disclose if they know if any technology is under any patent. To avoid someone push a standard and then later come back and claim royalties.
Finally SIP is as much about audio codec as video codecs and there are more people looking at streaming TV. IETF is the right place when you want to bring some technology, increase its robustness and standardize it. And may be at the difference of W3C, anything is a standard if at least 2 independent operable system has been built. You don’t need to wait for the others to agree, and the others can do their own standard too. At the end, it is the user that votes by using.

By: Marc Petit-Huguenin

Marc Petit-Huguenin — Thu, 30 Jul 2009 00:33:17 +0000

I do not see how Ericsson can pack the meeting with 40-100 people, as there is only 39 Ericsson employees registered to attend the IETF meeting (https://www.ietf.org/registration/ietf75/attendance.py?sortkey=3&login=). Also there is no such thing as a vote at the IETF. Closest thing is a hum that will need to be confirmed on the mailing-list.