Re: [codec] Authors and disclosures, regarding Last Call: <draft-ietf-codec-opus-12.txt>

Jean-Marc Valin <> Wed, 02 May 2012 00:16 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 3F16521E8086 for <>; Tue, 1 May 2012 17:16:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id L-uuw+utuUFE for <>; Tue, 1 May 2012 17:16:39 -0700 (PDT)
Received: from ( []) by (Postfix) with ESMTP id 5061021E8020 for <>; Tue, 1 May 2012 17:16:38 -0700 (PDT)
Received: from [] ( []) (Authenticated sender: by (Postfix) with ESMTP id 3A6344AEDE1; Tue, 1 May 2012 17:16:38 -0700 (PDT)
Message-ID: <>
Date: Tue, 01 May 2012 20:16:36 -0400
From: Jean-Marc Valin <>
User-Agent: Mozilla/5.0 (X11; Linux i686 on x86_64; rv:12.0) Gecko/20120428 Thunderbird/12.0.1
MIME-Version: 1.0
To: Joe Baker <>
References: <>
In-Reply-To: <>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [codec] Authors and disclosures, regarding Last Call: <draft-ietf-codec-opus-12.txt>
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Codec WG <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 02 May 2012 00:16:43 -0000

On 05/01/2012 03:33 PM, Joe Baker wrote:
> Can the authors of the draft please indicate when the major components
> of the codec were first offered to the public, in any form, so we can
> better gauge the risk of substantial components being covered by
> undisclosed applications?

Back when we had the first codec BoF (Jul 2009, Stockholm), both SILK
and CELT were already being deployed, and though they were further
modified after that point, the "core technology" hasn't changed that
much. As far as I can tell, SILK was "first offered to the public",
early 2009
while CELT was first released late 2007. Since January 2011, the CELT
layer has been completely frozen and the SILK layer only got minor fixes.

> I have audited the entire WG discussion archives, meeting minutes and
> source code repositories and was not able to find any technical
> contributions by anyone from Huawei.  I assume that the IETF policy
> does not permit working group participants to follow the working group
> and secretly patent the working group's inventions as their own?
> It would be helpful if anyone here is aware that their specific technical
> contributions will be covered by currently undisclosed applications which
> may not have royalty free licensing if they would point them out to the
> group so that we can decide if we should remove those parts.

As far as I know, the contributions of Huawei have been limited to
providing feedback, comments, review, and requirements rather than code,
algorithms, or design.

> I did notice one potentially important discrepancy.  There are many
> more authors of the draft which are not are listed on it as authors. Are
> these people treated equally to the listed authors in terms of the IETF
> IPR disclosure requirements?  I believe some of them have never posted to
> this list although I may have failed to match some of the names.

As far as I know, the IETF only allows about 5 authors to be listed at
the top of the draft, even though many more people may have contributed
to the document. You are correct that some people listed in the git logs
have never posted. There are two reasons for this. Some people have
merely sent us small bug fixes, draft typos, comment fixes and the like.
In the case of Mark Borgerding and Erik de Castro, they are authors of
pre-existing BSD code that we decided to include in Opus. This code
implements an FFT and fast float-integer conversion functions, so
nothing I'd be too worried about when it comes to IPR. As for Soren Skak
Jensen, he was a Skype employee, so his contributions are already covered.

On top of the names you found, there's actually one more significant
contribution. This was made by Raymond Chen on the mailing list. Because
I completely re-implemented that contribution in C, his name does not
show up in the git logs. However, Broadcom has properly disclosed IPR
and licensed it under the same license as the one Xiph.Org used.

Hope that answers your questions.