Thanks Elwyn,

Replies inline. Changes can be found here
<https://github.com/ietf-wg-masque/draft-ietf-masque-connect-ethernet/compare/5c39341..ff4d640>
 and in the latest editor's copy
<https://ietf-wg-masque.github.io/draft-ietf-masque-connect-ethernet/draft-ietf-masque-connect-ethernet.html>
.

Thanks again,
-Alejandro

On Tue, Jul 14, 2026 at 9:13 AM Elwyn Davies via Datatracker <
[email protected]> wrote:

> Nits/editorial comments:
>
> s1, para 1: s/can't carry/cannot carry/
>

Done.


> s1, para 3:    states 'This protocol supports all existing versions of
> HTTP by
> using HTTP Datagrams [HTTP-DGRAM]. This statement needs to be qualified
> to identify what the latest existing version of HTTP at the time of
> publication of
> this document.
>

The sentences that follow explicitly mention and reference HTTP/1.x,
HTTP/2.0, and HTTP/3.0; I think this is well covered.
This text was borrowed from
https://www.rfc-editor.org/rfc/rfc9484.html#section-1-4.


> s3, para 1, last sentence: It would be good to explicitly mention what
> .well-known location is being registered (.well-known/masque/ethernet) and
> provide a pointer to s11.2 which specifies the process of registration.
> Additionally this section should request the update of the well-known URIs
> IANA
> registry for masque to reference this document.
>

I've added a reference to §11.2 at the end of §s3,¶1.


> s3, para 4: s/which segment to connect to/the segment to which to connect/
>

Done

s3, para 8 (1st bullet): s/a level 3 template or lower/a template of type
> lwvel
> 3 or lower. See Section 1.2 of [TEMPLATE]./
>

Done

s3, para 9 (2nd bullet): Add a reference to RFC 2396 that defines e
> 'absolute'
> term.
>

RFC 2396 was obsoleted by RFC 3986. Added a reference to §4.3 of RFC 3986.


> s4, Figures 1 - 4: Capsule-Protocol field definition. A note that '?1' is a
> Structured Field Boolean True value as per RFC 8941 might help.
>

I'm not sure how I would fit that in. [HTTP-DGRAM] already covers it pretty
well.
I could add an informative reference to RFC 8941 if you think that would be
helpful, but I'm inclined to leave this as is.


> s5: The last paragraph refers to the possibility of an endpoint receiving a
> context ID that has not been regeistered but does not explain the
> consequences
> of this or whether that is something that would be specified as part of the
> registration process. Thwre is discussion in s6 that may imply this.
>

This text comes directly from
https://www.rfc-editor.org/rfc/rfc9484.html#section-5-3.


> s6, para 4 (one after fields definitions), last sentence: s/include
> include/include/
>

Done.


> s8.1: Does STREAM(44) need some explanation? Also is Quarter Stream ID = 11
> just a reandom value or meaningful?
>

Those identifiers are borrowed from
https://www.rfc-editor.org/rfc/rfc9484.html#name-examples. The Quarter
Stream ID is derived from the Stream ID. 44 / 4 = 11.
https://www.rfc-editor.org/rfc/rfc9297.html#section-2.1-3.2.1 explains
more, and this is an implementation detail of HTTP/3 Datagrams. The
specific Stream ID and Quarter Stream ID used here are not meaningful
except in that they are properly related to each other.


> s9.2, para 2:s/to access to/for access to/
>

Done

s11.2: As noted above as related to s3, para 1, IANA should be requested to
> update the well-known URIs IANA registry for masque to additionally
> reference
> this document.
>

The update is described by the table in §11.2, and already mentions a
reference to the document. IANA has confirmed that the reference will be
included in the update.
_______________________________________________
Gen-art mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to