Re: [dane] DANE Client Authentication draft updated

[Shumon Huque <[e-mail]>]

On Tue, Jan 12, 2016 at 9:10 PM, Kim Alvefur <[e-mail]> wrote:

> On 01/12/2016 11:21 PM, Shumon Huque wrote:
> > On the "_smtp-client" label choice,
>
> That seems like it will cause confusion considering _xmpp-client is used
> in the XMPP world to refer to where the end-users connect their user
> agents to, so basically equivalent to _submission in email.  And then
> there's _xmpp-server for communication between servers.
>

Yeah, I'm aware of that. I think the XMPP community made an
unfortunate choice in those names  - I might have suggested
"_xmpp-c2s" and "_xmpp-s2s" instead.

As John Levine points out, the client service name choices are already
all over the map in the current IANA service registry, so ...

FWIW, I would suggest _smtp-out as a less confusing label.
>

The _smtp-client record was just an example we used in the draft.
We weren't trying to impose a uniform convention for IANA client
application labels here (probably not the right place to do so), but
perhaps we can consider making a recommendation in the IANA
Considerations section of the draft.

The word "client" is a bit ambiguous whether it refers to a user agent
> or a server in the process of connecting to another server. It would
> help to be more explicit about which one of these are being referred to,
> or if it applies to both.
>

In my opinion, it covers any entity acting as a client, so it covers both
the cases you outline. I'm not sure we need to make a distinction in
the draft.

Client-DANE in the later context, enabling mutual authentication between
> SMTP servers could be a nice improvement / complement to SPF & co.
>

Glad to hear it. That's one of the use cases that this draft had in mind.

-- 
Shumon Huque