Re: [art] Revising BCP56: On the use of HTTP as a Substrate

[Eliot Lear <[e-mail]>]

Hi Mark,

There is no question that 3205 could use a dust off.  I would suggest
that document really served two purposes.  The first was to provide best
practices on its use, and the second was more of an applicability
statement.  You've gone to some detail as to best practices, but it
seems to me more attention should be paid to the applicability component.

Good engineering principles dictate not only when a mechanism should be
used, but as important, when it should not.  HTTP's use has GREATLY
expanded thanks in large part to REST, but it also could it might narrow
somewhat, thanks to protocols such as CoAP, or through the further
development of QUIC.  A valuable contribution would be to give guidance
as to what substrate is appropriate and when.

Related, we TCP/UDP port reviewers occasionally see requests for new
ports that make use of HTTP as a substrate.  Personally I always welcome
more guidance as to when to approve those ports.  The standing view is
that we approve them in large part to address non-interference with
existing services, but at some level that seems a bit of a waste, and
perhaps we should come out more strongly to say that if HTTP is really
to be used as a substrate, there must be some form of appropriate
application layer dispatch perhaps tied to .well-known.

I won't be attending the HTTPBIS meeting but I hope you'll consider
these comments as you develop the document.

Eliot



On 7/11/17 6:14 AM, Mark Nottingham wrote:
> A number of folks have recently noted that BCP56 addresses the problems of using HTTP for protocols in the early 2000's, but we've moved on considerably since then. 
>
> Given the number of IETF protocols being build upon HTTP these days, it seems timely to reconsider it. I've been working on a candidate for replacing it for a little while; see:
>   https://tools.ietf.org/html/draft-nottingham-bcp56bis
>   https://mnot.github.io/I-D/bcp56bis/
>
> I have a small-ish slot in the HTTP WG session on Wednesday to discuss this. 
>
> Cheers,
>
> --
> Mark Nottingham   https://www.mnot.net/
>
> _______________________________________________
> art mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/art
>