[Standards] XEP-0369 (MIX): Early stages of a clients connection
steve.kille at isode.com
Thu Jun 15 12:22:00 UTC 2017
I believe that the Option #4 change has resolved this issue, and has enable me to delete the offending text
> -----Original Message-----
> From: Standards [mailto:standards-bounces at xmpp.org] On Behalf Of Kevin
> Sent: 14 June 2017 08:46
> To: XMPP Standards
> Subject: Re: [Standards] XEP-0369 (MIX): Early stages of a clients connection
> On 14 Jun 2017, at 08:36, Steve Kille <steve.kille at isode.com> wrote:
> >> One thing I could imagine is for the server to include that
> >> information in the roster version. E.g. if it usually versions the
> >> roster with incrementing numbers, it could add "+mix" to the
> >> stringified number if MIX entries are included. This would allow it
> >> to recognise that the client has or has not seen a roster version with MIX
> and can create the diff accordingly.
> > That feels very ugly.
> It’s up to the server how they encode roster versions, so a server choosing to
> do this seems entirely sensible to me (I suggested it offlist yesterday). It
> doesn’t need standardising, though, it’s a server implementation issue
> (although if we want to put in a sentence suggesting it’s possible, that
> wouldn’t be daft).
> >> It took me a while to come up with this idea, so maybe it makes sense
> >> to add that to the wording. Also I have never written a server-side
> >> roster implementation, so I’m not sure if this is even a realistic way of
> doing things.
> > [Steve Kille]
> > I am thinking that Kev's idea of adding "send MIX format" option to roster
> request is the best way to go.
> I think Jonas’s argument that if this is an ‘enable MIX’ rather than just ‘add
> MIX to the roster’ request, its own iq is right.
> > Server will need to maintain things so that it can respond to both MIX and
> non-MIX requests.
> > If a client changes its setting, it needs to clear its cache and fetch the full
> This is wrong, clients should not need to purge their roster cache, servers can
> always cope, and if they cope with a full push rather than a diff, that’s fine.
> Standards mailing list
> Info: https://mail.jabber.org/mailman/listinfo/standards
> Unsubscribe: Standards-unsubscribe at xmpp.org
More information about the Standards