Re: WGLC for Datagram Extension

[David Schinazi <[e-mail]>]

Thanks Spencer! I've approved the PR - we can merge it at the end of WGLC
assuming no one objects.

David

On Thu, Sep 23, 2021 at 6:08 PM Spencer Dawkins at IETF <
[e-mail]> wrote:

> Hi, David,
>
> On Thu, Sep 23, 2021 at 6:53 PM David Schinazi <[e-mail]>
> wrote:
>
>> Hi Spencer,
>>
>> Yes, your interpretation is correct. The language about whether it's
>> interchangeably one frame type or two frame types mirrors RFC 9000's
>> definition of the STREAM frame which is also one or eight frames depending
>> on how you look at it <
>> https://datatracker.ietf.org/doc/html/rfc9000#section-19.8>.
>>
>
> Thank you for the background here (I was watching the base QUIC drafts,
> but spent a lot of time looking at diffs, rather than re-reading carefully,
> so your pointer was super helpful).
>
> I did see a couple of places where the STREAM frame type was using plurals
> (for example,
> https://www.rfc-editor.org/rfc/rfc9000.html#name-stream-frames, plural,
> versus
> https://www.ietf.org/archive/id/draft-ietf-quic-datagram-04.html#name-datagram-frame-type,
> singular), so I included that in the PR. I swear, if I wasn't looking for
> the second datagram frame type in the second called "datagram frame type",
> I wouldn't have noticed anything 🙂)
>
> And it really was a couple of small changes. Thanks for considering my
> comments.
>
> Best,
>
> SPencer
>
>
>> Regarding your editorial comments, could I ask you to send them our way
>> as a PR to <https://github.com/quicwg/datagram> ?
>>
>> Thanks,
>> David
>>
>> On Thu, Sep 23, 2021 at 4:33 PM Spencer Dawkins at IETF <
>> [e-mail]> wrote:
>>
>>> Hi, Lucas,
>>>
>>> I'm top-posting to say that I'm not talking about multiplexing
>>> datagrames now (except to thank you and Christian for your replies), but
>>> I'm still looking at the datagram draft, and noticed something odd. Could
>>> you put your working group chair hat back on for a moment?
>>>
>>> From the Introduction:
>>>
>>> *This document defines two new DATAGRAM QUIC frame types, which carry
>>> application data without requiring retransmissions.*
>>>
>>>
>>> And most of the document about "datagram frame types" (plural), but
>>> pretty much all of
>>> https://www.ietf.org/archive/id/draft-ietf-quic-datagram-04.html#name-datagram-frame-type
>>> talks about "the datagram frame type" (singular) - even the section header
>>> uses the singular phrasing, along with the caption to Figure 1.
>>>
>>> I understand what the document is saying, to be
>>>
>>>    - there are two datagram frame types, 0x30 and 0x31
>>>    - the only difference between these two datagram frame types is that
>>>    datagram frames with frame type 0x30 do not include a length field, while
>>>    datagram frames with frame type 0x31 do contain a contain a length field,
>>>    and
>>>    - otherwise, whatever this document says about "datagrams" is true
>>>    for both datagram frame types.
>>>
>>> Is that about right?
>>>
>>> I can dig that out from the document, but if that was clearer in Section
>>> 4 - "Datagram Frame Types", and a couple of clarifying sentences, that
>>> might be a bit easier for the reader (and, of course, for the various
>>> review teams that will be looking at this document for the first time
>>> during IETF Last Call).
>>>
>>> As long as I'm still typing, I might mention one other thing - I
>>> searched for the string "reliabl" in the draft, and there are 24
>>> occurrences, which is fine, although perhaps repetitive, until you get down
>>> to
>>> https://www.ietf.org/archive/id/draft-ietf-quic-datagram-04.html#name-behavior-and-usage,
>>> which says
>>>
>>> When an application sends an unreliable datagram over a QUIC
>>> connection, QUIC will generate a new DATAGRAM frame and send it in the
>>> first available packet. This frame SHOULD be sent as soon as possible, and
>>> MAY be coalesced with other frames.
>>>
>>>
>>> Could "unreliable" be dropped from the first sentence? It would belong
>>> there if reliable datagrams were an option, but (after 20 occurrences
>>> explaining that all datagrams are unreliable), maybe this is just inviting
>>> confusion.
>>>
>>> Thanks for considering my comments, of course. Do The Right Thing!
>>>
>>> Best,
>>>
>>> Spencer
>>>
>>> On Thu, Sep 23, 2021 at 1:12 PM Lucas Pardue <[e-mail]>
>>> wrote:
>>>
>>>> Hi Spencer,
>>>>
>>>> On Thu, Sep 23, 2021 at 6:52 PM Spencer Dawkins at IETF <
>>>> [e-mail]> wrote:
>>>>
>>>>> Hi, Lucas and Matt,
>>>>>
>>>>> On Thu, Sep 16, 2021 at 3:35 PM Lucas Pardue <
>>>>> [e-mail]> wrote:
>>>>>
>>>>>> Hello QUIC WG,
>>>>>>
>>>>>> This email starts the Working Group Last Call for "An Unreliable
>>>>>> Datagram Extension to QUIC", located at:
>>>>>>   https://www.ietf.org/archive/id/draft-ietf-quic-datagram-04.html
>>>>>>
>>>>>> Please take time to review it carefully and raise any remaining
>>>>>> issues you see either on the GitHub repo issues list[1] or on this mailing
>>>>>> list.
>>>>>>
>>>>>
>>>>> I've re-read the discussion in
>>>>> https://github.com/quicwg/datagram/issues/6, and now better
>>>>> understand the points of view that resulted in the lack of datagram
>>>>> multiplexing in this specification (I was following that discussion when it
>>>>> was happening, but hindsight during re-reading helped a lot!)
>>>>>
>>>>> I especially appreciate the addition of the text in
>>>>> https://www.ietf.org/archive/id/draft-ietf-quic-datagram-04.html#name-multiplexing-datagrams
>>>>> .
>>>>>
>>>>> I accept the wisdom of getting more experience with various people
>>>>> using application-defined multiplexing in various ways before adding any
>>>>> flavor of multiplexing at the transport layer, and agree that holding this
>>>>> specification up while that experience is gathered, would not be The Right
>>>>> Thing To Do.
>>>>>
>>>>> I'm aware of at least two "RTP over QUIC" proposals that assume that
>>>>> they'll need to multiplex datagrams, so I won't be surprised if we return
>>>>> to this question in the future, but for now, the chairs are Doing The Right
>>>>> Thing, and I support requesting publication of -04 (modulo any changes that
>>>>> pop up in WGLC, of course).
>>>>>
>>>>
>>>> Thanks for your comments.
>>>>
>>>> Chair hat off, speaking as an individual. To add to the examples, the
>>>> MASQUE working group is working on a draft to define HTTP's use of QUIC
>>>> DATAGRAM frames and it includes multiplexing [1]. The working group's 00
>>>> draft started with a single variable-length integer multiplexing identifier
>>>> called the Flow ID. After some fruitful WG implementation and discussion,
>>>> in draft 01 it switched to an identifier tuple (Quarter Stream ID, Contenxt
>>>> ID). I'd say that we're still figuring out exactly how the things the
>>>> __use__ datagrams, want to use datagrams. I think it's too early to
>>>> understand the right common transport solution at this time, and that
>>>> leaving it undefined does not prevent people from using datagram frames for
>>>> their own precise needs and purpose.
>>>>
>>>> Cheers,
>>>> Lucas
>>>>
>>>> [1] - https://datatracker.ietf.org/doc/draft-ietf-masque-h3-datagram
>>>>
>>>