[Standards-JIG] Re: JEP-0134 (Protocol Design Guidelines): Does it warn against "Atom over XMPP"?

Peter Saint-Andre stpeter at jabber.org
Mon Oct 25 22:57:25 UTC 2004


In article <20041024015939.3283F99D8 at mailrelay.t-mobile.com>,
 "Bob Wyman" <bob at wyman.us> wrote:

> JEP-134 says at section 2.5 "Know your strengths":
> > the average size of an XMPP stanza is probably on the order of
> > 200 bytes, and in general the largest effective size for stanzas
> > is perhaps ten times that large; although stanzas can be much
> > larger than that, normally it is not wise to consistently generate
> > larger stanzas
> 	JEP-0134 seems to state that "Atom over XMPP" would constitute an
> "unwise" application of XMPP since "Atom over XMPP" would often result in
> the generation of stanzas larger than "the largest effective size for
> stanzas" (i.e. 2000(10x200) bytes.)
> 	Given that the author of JEP-0134, Peter Saint-Andre, is also an
> author of draft-saintandre-atompub-notify-01, I find the JEP-0134 text to be
> somewhat confusing... Perhaps it would be useful to say something in the
> JEP-0134 text about those "abnormal" conditions under which it might be
> considered to be not unwise to generate large stanzas such as those which
> will be typical of "Atom over XMPP"...
> 	My expectation is that "Atom over XMPP" and similar uses of JEP-0060
> will soon grow to become at least the second most popular use of XMPP --
> with IM remaining the most common use for some time. If XMPP is the wrong
> protocol for use by these applications, it would be good to make that clear
> as soon as possible. 
> 
> 		bob wyman

How large would you expect Atom entries to be, in general? (Not full 
feeds, but entries.)

/psa




More information about the Standards mailing list