Re: [Standards] XEP-0369 (MIX) - Approach to options for per-user preferences
On 21 Sep 2016, at 16:21, Steve Kille wrote: > The sort of thing I am looking to avoid is where the server provides the > user with a form to fill in. By standardising common fields, we can avoid form requests in the common case without forcing clients to do the request dance (the same approach we took with MAM). Clients don’t have to implement anything other than the required fields, so don’t have any additional complexity, and for servers the only additional enforced complexity is returning a form when requested, which seems trivial. I don’t think we need to avoid forms in this instance, as the complexity is optional. /K ___ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: standards-unsubscr...@xmpp.org ___
Re: [Standards] XEP-0369 (MIX) - Approach to options for per-user preferences
> On 21 Sep 2016, at 14:32, Steve Kille wrote: > > MIX channel participants have the option to configure preferences for JID > visibility.There are some comments on the spec in this area, which I aim > to address soon. > > The current specification is formulated so that an implementation can > provide implementation specific per-user preferences, which it offers to the > user through a XEP-0004. > > I have been considering the sort of preferences that a MIX channel > participant might have. These can generally be addressed by choice of > which channel nodes are subscribed to. > > I have not been able to think of any additional sensible per-user > preferences. Possible ones might be ‘allow vcard requests through proxy JID’, ‘mirror my PEP through the proxy JID’, I suppose. > > I am thinking that the best approach is to only allowing configuration of > the JID visibility preference and not making this configuration extensible. > I am keen to keep things as simple as possible/sensible. I think that where extensibility isn’t going to add any significant implementation complexity (which seems to be the case here), it’s sensible to allow it. /K ___ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: standards-unsubscr...@xmpp.org ___
Re: [Standards] XEP-0369 (MIX) - Approach to options for per-user preferences
> -Original Message- > From: Standards [mailto:standards-boun...@xmpp.org] On Behalf Of Dave > Cridland > Sent: 21 September 2016 15:58 > To: XMPP Standards > Subject: Re: [Standards] XEP-0369 (MIX) - Approach to options for per-user > preferences > > On 21 September 2016 at 14:32, Steve Kille wrote: > > I have not been able to think of any additional sensible per-user > > preferences. > > Users might well want to control whether they receive private messages > through the MIX or not. [Steve Kille] I think that one is worth including in MIX. > > In general, unless there's a compelling reason, where there's already > "space", I'd prefer to keep things generalizable. Allowing arbitrary > preferences doesn't appear to introduce any further protocol round-trips, > and as long as optional preferences are truly optional - that is, there's an > expectation that clients may ignore them - then this seems the more future- > proof option. [Steve Kille] If you define things in MIX, the protocol becomes cleaner (in my view). This mechanism can also allow for proprietary extensions agreed between client and server. The sort of thing I am looking to avoid is where the server provides the user with a form to fill in. Steve ___ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: standards-unsubscr...@xmpp.org ___
Re: [Standards] XEP-0369 (MIX) - Approach to options for per-user preferences
On 21 September 2016 at 14:32, Steve Kille wrote: > I have not been able to think of any additional sensible per-user > preferences. Users might well want to control whether they receive private messages through the MIX or not. In general, unless there's a compelling reason, where there's already "space", I'd prefer to keep things generalizable. Allowing arbitrary preferences doesn't appear to introduce any further protocol round-trips, and as long as optional preferences are truly optional - that is, there's an expectation that clients may ignore them - then this seems the more future-proof option. Dave. ___ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: standards-unsubscr...@xmpp.org ___