On the wave of SPAM...
The XEP looks ok, but I see problems with section 4 ("Use with the
Blocking Command"). It looks like some sort of traffic optimization, no?
This creates some module dependencies which is something I don't like
very much. For example, the XEP-0377 might be implemented by a
third-party developer while it requires some hooks or callbacks in
XEP-0191 (which might also be implemented by third-party). This will
create burden for admins because of server-module version mismatch.
Another issue is that it's hard to process errors from such request.
For example, if a client receives some server processing error, does
that mean the error relates to XEP-0377 or to XEP-0191? Should a client
split the request in two parts and resend? Other errors might be
defined in the future specific to the XEPs which is also hard to
separate. So I don't think we gather any advantages with such sort of
"stanza optimization", but we will likely face the discussed troubles.
_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________
