[Gen-art] Re: draft-ietf-cellar-codec-18 ietf last call Genart review

[Steve Lhomme <[e-mail]>]

Hello Mallory,

Thanks for your review! My comments are inline below.

> On 1 Jun 2026, at 14:58, Mallory Knodel via Datatracker <[e-mail]> wrote:
> 
> Document: draft-ietf-cellar-codec
> Title: Matroska Media Container Codec Specifications
> Reviewer: Mallory Knodel
> Review result: Ready
> 
> I am the assigned Gen-ART reviewer for this draft. The General Area
> Review Team (Gen-ART) reviews all IETF documents being processed
> by the IESG for the IETF Chair.  Please treat these comments just
> like any other last call comments.
> 
> For more information, please see the FAQ at
> 
> <https://wiki.ietf.org/en/group/gen/GenArtFAQ>.
> 
> Document: draft-ietf-cellar-codec-??
> Reviewer: Mallory Knodel
> Review Date: 2026-06-01
> IETF LC End Date: 2026-05-29
> IESG Telechat date: Not scheduled for a telechat
> 
> Summary: This document describes how to map codecs into the Matroska container
> format. The draft is well written and clear in purpose.
> 
> Major issues: None. Thanks for the document. The biggest challenge I faced
> while reviewing this was to do my best to check out all of the different
> reference documents (there are so many! and from many other standards bodies).
> I guess that this is the core function of this work, to bring together many
> different codecs into one standard framework. Caveat that this is a general
> review and not a technical review.
> 
> Minor issues:
> 
> * Decide whether codec is plural and if so delete all "codecs" (love an
> invariable noun).

It seems that the proper way is to use ”codecs” according to wiktionary: https://en.wiktionary.org/wiki/codec#English
We also used it a few times in RFC 9559 as well.
I did a pass on a few cases where a proper plural should be used and make a Pull Request for the next iteration of this document:
https://github.com/ietf-wg-cellar/matroska-specification/pull/1104  

> * There is confusion for a new reader between "Block Addition" and "Block
> Additional", both a kind of necessary mapping. What seems to be the difference
> is that the latter is about mapping additional /data/, not about adding a
> block of data. That's a nuance. It would be helpful to define this in an
> introductory paragraph. This sort of introductory paragraph can also provide a
> guide to the table of contents-- why there are mappings of codecs and blocks,
> both for audio and video, and how to handle the data within them. That
> high-level architectural explanation is missing from the doc. If it exists
> elsewhere, you could link to that and summarize.

I reordered the text and remove the paragraph from section 7 as it would duplicate the text from section 4.
https://github.com/ietf-wg-cellar/matroska-specification/pull/1108

These sections will now be merged towards the end. They were both dealing with the same thing, i.e. the data inside the <BlockAdditions> element that may (BlockAddID=1) or may not (BlockAddID!=1) be directly related to the codec data.

The term "Block Additions" is no longer used.

And this section has 2 pseudo-XML example of how the data are layed out for BlockAddID=1 and BlockAddID!=1. Hopefully it’s enough to clear the confusion.

> Nits/editorial comments:
> 
> * Also on the plural tack: You could probably say "Codec Mapping" rather than
> the plural form in most cases, just for consistency, and for the title of
> section 3 specifically, but not, say, 3.2. or 3.3..

OK. I also added these changes to https://github.com/ietf-wg-cellar/matroska-specification/pull/1104

> * 3.3.1: "doesn't contains”

Fixed via https://github.com/ietf-wg-cellar/matroska-specification/pull/1105