On 12 March 2010 14:15, Nicolas Vérité <[email protected]> wrote: > On Fri, Mar 12, 2010 at 14:02, Matthew Wild <[email protected]> wrote: >> On 12 March 2010 10:37, Nicolas Vérité <[email protected]> >> wrote: >>> On Fri, Mar 5, 2010 at 19:53, XMPP Extensions Editor <[email protected]> >>> wrote: >>>> Version 0.1 of XEP-0279 (Server IP Check) has been released. >>>> >>>> Abstract: This specification defines a simple XMPP extension that enables >>>> a client to discover its external IP address. >>>> >>>> Changelog: Initial published version. (psa) >>>> >>>> Diff: N/A >>>> >>>> URL: http://xmpp.org/extensions/xep-0279.html >>> >>> I am really confused here: Are we reinventing the wheel? Why add this >>> entropy? STUN is there, well thought out, coded and strongly tested, >>> though maybe a bit complex. If we (improve and) accept this XEP, then >>> what CAN/SHOULD/MUST clients and/or servers implement then? STUN >>> and/or SIC? >>
> >> I think the XEP needs to be updated to make this clear. Could the >> people interested in seeing this XEP implemented perhaps suggest some >> alternative text for the introduction? > > The intro might not be the place to work out most in this phase. I > believe we first need to address philosophy issues (iq or stream > feature?), then setup rules for clients and servers developers saying > what to use first, and what else as a fallback. Ah yes, forgot to address this in my previous message. I'm not sure that including it in a stream feature is nice. One minute people are saying no-one is going to want to use this XEP, and then they're suggesting putting it ina stream feature to all clients (most of whom aren't interested). The main reason I don't like it though is that it really isn't a feature. It's not something the stream has to offer, not something the client can negotiate - which is more relevant now that features are more closely tied to negotiation in 3920bis. It was brought to my attention that XEP-0115 also uses stream:features this way, and how to fix that I'm not sure, but I don't think we want to go further in that direction. Matthew (I hope this bikeshed looks pretty by the time everyone's finished with it)
