[dispatch] Re: INDF - Subscription Account RSS/Atom feeds: request for other people to take care of this problem

[Michael Toomim <[e-mail]>]

This is an important problem, and we have a number of drafts written 
now, such as Braid, PREP, Mercure, GENA, and SUBSCRIBE:

  * Braid:
    https://datatracker.ietf.org/doc/html/draft-toomim-httpbis-braid-http
    (and see the Multiresponse approach in
    https://datatracker.ietf.org/doc/html/draft-toomim-httpbis-versions#section-2.4.4)
  * PREP:
    https://datatracker.ietf.org/doc/draft-gupta-httpbis-per-resource-events/
  * Mercure: https://datatracker.ietf.org/doc/draft-dunglas-mercure/
  * GENA:
    https://datatracker.ietf.org/doc/html/draft-cohen-gena-p-base-01 and
    https://datatracker.ietf.org/doc/html/draft-cohen-gena-client-00
  * SUBSCRIBE: https://awwright.name/2024/draft-wright-http-subscribe.html

We also have a few surveys written comparing these approaches, written 
by Mnot, Josh Cohen, and Kevin Dunglass:

  * https://www.mnot.net/blog/2022/02/20/websockets
  * https://braid.org/@josh/pubsub-comparison
  * https://dunglas.dev/2023/11/mercure-braid-prep-news-about-subscribing-to-http-resource-updates/

Some of these approaches take advantage of Reverse HTTP / PTTH, by 
having the server send a NOTIFY or PUT/POST/PATCH request to the client.

  * Reverse HTTP / PTTH demonstrated interest in a recent BoF at meeting
    123: https://datatracker.ietf.org/meeting/123/session/ptth/
  * Something like it would be necessary for the approach in GENA, and
    it could give new abilities to Mercure.
  * We are taking notes at
    https://braid.org/protocol/reverse-http#subscriptions-with-pushed-updates

And, of course, there are existing specs for WebSub and SSE, which are 
nice, but aren't good enough.

Before Josh Cohen passed away earlier this year, he was wanting to start 
a discussion in the HTTPWG to organize and compare and contrast the 
design decisions in these different approaches, so that we could make 
sense of the actual choices in front of us, and get some shared mind on 
this problem. I was going to help.

Let me know if you're interested in this work.

Michael

On 8/20/25 2:00 AM, Jonas Korene Novak wrote:
> Hi Rob,
>
> There's a W3C standardized protocol named "WebSub"
> (which can be found here: https://www.w3.org/TR/websub/)
>
> Although not an IETF standard, it seems to fit what you're describing 
> close enough.
>
> If you're just looking for something that works, I think this covers it,
> but if your goal is for it to also be standardized in the IETF and 
> other members
> agree, I think the next step would be for someone to write up a draft.
>
> Regards,
> Jonas
>
>> My request is for the IETF to formulate a standard for feed and
>> podcast clients to get /one/ notification of an update for all
>> subscription URLs. This work has been done, I think. There are "Well
>> Known URIs", "URI templates", and "Webhooks". Think of products like
>> Patreon/Memberful/Substack etc.
>
> _______________________________________________
> dispatch mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]