Re: [xmpp] PROTO review of draft-ietf-xmpp-posh

[Peter Saint-Andre - &yet <[e-mail]>]

On 2/23/15 3:31 PM, Peter Saint-Andre - &yet wrote:
> On 2/23/15 3:10 PM, Ben Campbell wrote:
>> Hi Peter, further comments below. I removed sections that do not seem to
>> need further discussion.
>>
>> On 23 Feb 2015, at 15:58, Peter Saint-Andre - &yet wrote:
>>
>> [...]
>>
>>>>
>>>> — Material Comment: IANA Considerations
>>>>
>>>> These seems a bit unusual, since we are registering a “fragment” that
>>>> other protocols will use to register actual URIs.  This does not
>>>> seem to
>>>> have been contemplated by RFC5785. This also the side effect of
>>>> establishing rules for certain entries in the well-known URI registry
>>>> over and above those from RFC5785.
>>>>
>>>> Does it make sense to actually register the prefix itself, since it’s
>>>> not really a URI? It would seem reasonable to leave the actual
>>>> registration to protocols that need to register posh URIs.
>>>
>>> You make a very good point. I think you're right that it makes more
>>> sense for the POSH spec to provide instructions to protocols that need
>>> to register POSH URIs - as, for instance, draft-ietf-xmpp-dna does -
>>> but not to register a URI prefix (instead, just say "please use the
>>> prefix").
>>
>> So if I understand the implications of this path, it would mean we give
>> guidance that any protocol that uses posh should register a “posh.*”
>> URL.
>
> Yes but that's really just for a friendly consistency.
>
> Matt and I chatted about it over IM and we both recalled that we were
> trying to prevent crazy registrations in .well-known (this was around
> the time that Mark Nottingham was starting to work on the evocatively
> named draft-nottingham-uri-get-off-my-lawn, now RFC 7320).
>
>> (I note that the well-known URI registry uses the specification
>> required policy. Is that adequate for POSH use?).
>
> I think so. Over time we've tried to become more liberal about IANA
> registrations policies.
>
>> But this doc would not
>> register anything.
>
> Correct.
>
>> Would that guidance still go in the IANA
>> considerations section? I suspect the answer is probably not.
>
> It would go in another section since we wouldn't be registering
> anything, nor would we be providing actionable guidance to IANA.
>
>> Also, does anything implode if some protocol chooses not to use the POSH
>> prefix?
>
> No imminent implosions to worry about.
>
>> Do we actually care?
>
> Not much, although as I said it provides a friendly consistency.
>
>> I assume we don’t expect developers to try
>> to guess well-known URIs.
>
> That's true.
>
>>>> I see Mark Nottingham is the expert for the well-known URI registry. By
>>>> any chance has anyone run this by him?
>>>
>>> I have a vague recollection of having talked with him about it once,
>>> but I can find no evidence of that in my email folders. If we agree
>>> that what you suggest is the best approach, then I think it makes
>>> sense to update the document before reaching out to him (and in fact
>>> that might not be necessary since this document wouldn't be doing
>>> anything unusual).
>>
>> I agree that if we don’t register anything, and don’t try to change the
>> well-known URI registration policy for posh purposes, it might not be
>> necessary to involve Mark. (But for the record, if he were okay with the
>> way things are currently documented, I would back off on this point.)
>
> Well, I think the way things are currently documented is misguided, so
> we might as well correct it.

Here is proposed text:

###

8.  Guidelines for Protocols that Use POSH

    Protocols that use POSH will need to register well-known URIs wth the
    IANA in accordance with [RFC5785] (the IANA registration policy
    [RFC5226] for well-known URIs is Specification Required).

    For the sake of consistency, it would be best if the URIs registered
    by such protocols match the URI template [RFC6570] path "/.well-known
    /posh.{servicedesc}.json"; that is, begin with "posh." and end with
    ".json" (indicating a media type of application/json [RFC7159]).

    For POSH-using protocols that rely on DNS SRV records [RFC2782], it
    would be best if the "{servicedesc}" part of the well-known URI is
    "{service}.{proto}", where the "{service}" is the DNS SRV "Service"
    prepended by the underscore character "_" and the "{proto}" is the
    DNS SRV "Proto" also prepended by the underscore character "_".  As
    an example, the well-known URI for XMPP server-to-server connections
    would be "posh._xmpp-server._tcp.json" since XMPP [RFC6120] registers
    a service name of "xmpp-server" and uses TCP as the underlying
    transport protocol.

    For other POSH-using protocols, the "{servicedesc}" part of the well-
    known URI can be any unique string or identifier for the protocol,
    which might be a service name registered with the IANA in accordance
    with [RFC6335] or which might be an unregistered name.  As an
    example, the well-known URI for a hypothetical "SPICE" application
    could be "posh.spice.json".

###

Peter

-- 
Peter Saint-Andre
https://andyet.com/