Re: [codec] Mirja Kühlewind's No Objection on draft-ietf-codec-opus-update-08: (with COMMENT)

[Ben Campbell <[e-mail]>]

> On Aug 14, 2017, at 10:56 AM, Mirja Kühlewind <[e-mail]> wrote:
> 
> Mirja Kühlewind has entered the following ballot position for
> draft-ietf-codec-opus-update-08: No Objection
> 
> When responding, please keep the subject line intact and reply to all
> email addresses included in the To and CC lines. (Feel free to cut this
> introductory paragraph, however.)
> 
> 
> Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
> for more information about IESG DISCUSS and COMMENT positions.
> 
> 
> The document, along with other ballot positions, can be found here:
> https://datatracker.ietf.org/doc/draft-ietf-codec-opus-update/
> 
> 
> 
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> While I support the publication as this seems to be the right thing for me to
> do in this situation, I find it really weird that we have to publish an RFC to
> fix bugs in an example implementation in the appendix (that is even encoded).
> While it is a good thing to have code in an RFC as far as this helps
> implementation, maybe rfc6716 went to far with putting a whole reference
> implementation in there. I guess that has also led to a situation where
> everybody is just using the provided code while efforts to reimplement the spec
> could have detected these bugs earlier in the process. Maybe it's an item for
> the IESG to thinking about how to provide a reference implementation with an
> RFC (that maybe can be updated easier) other than just putting it in the
> appendix

The reference implementation in the appendix in RFC 6716 is not just an example implementation. It contains the normative specification of the bit stream decoder—which makes up the majority of the normative bits in that document. 

That has, so far, been a very uncommon approach, and I don’t know that we would do it again. But if we choose to do so, I absolutely agree we should consider whether it can be done in an easier to update fashion. 

Ben.

> 
> One more actionable comment: I think the abstract could say more about the
> updates, especially maybe noticing that this only updates the code in the
> appendix and not the normative language in the body.
> 
>