On Friday 29 May 2009 05:28:51 Jiří Zárevúcky wrote: > 2009/5/29 Dirk Meyer <[email protected]>: > > Peter Saint-Andre wrote: > >> In the XMPP Council meeting yesterday, we discussed the desirability of > >> adding some implementation notes to XEP-0199 (XMPP Ping) before > >> advancing it from Draft to Final. > > > > Stupid question: why use XEP-0199 and not <r/> <a h='0'/> from XEP-0189 > > for this task? > > Very simple answer: Clients not supporting the XEP will still respond > with IQ error (if they are XMPP compliant, which is not the > extension's problem). They wouldn't respond to some unknown "r" > element. Also think of all the client that won't support XEP-189 for > veeeery long time.
You wouldn't send elements from XEP-198 without first ensuring the other side supports it (via stream features) and negotiating that it be used, so there should never be a concern about sending an "r" element to a peer that wouldn't understand it. But yes, the advantage of XEP-199 is that it works with clients that may not actually support XEP-199. :) For clients that aren't XMPP-compliant enough to return errors from unsupported iq types, the server could use a heuristic approach to detect such clients and then not ping them. -Justin
