El Friday 22 August 2008 07:23:28 Paul Kyzivat escribió: > Clearly there was a decision to model it after email.
I understand it, of course, but some times when complaining about SIP BNF I've received replies arguing that it's not a problem of SIP, but a problem of SMTP or HTTP in which SIP is based. This argue is not valid for me, since SIP designers could the oportunity of basing SIP in other protocols (or making it from scratch). Anyway, even if SIP is based in email, it could be done more conservative. For example headers like "From" are a pain since basically *all* is valid: From: DisplayName <sip:[EMAIL PROTECTED];uri_param=qq>;hdr_param=kk From: "DisplayName" <sip:[EMAIL PROTECTED];uri_param=qq>;hdr_param=kk From: <sip:[EMAIL PROTECTED];uri_param=qq>;hdr_param=kk From: sip:[EMAIL PROTECTED];hdr_param=kk <-- URI param not allowed here And also the header field name: From FROM from f F frOM Instead of allowing all these stuff, allow just: From: "DisplayName" <sip:[EMAIL PROTECTED];uri_param=qq>;hdr_param=kk From: <sip:[EMAIL PROTECTED];uri_param=qq>;hdr_param=kk From f People would be happier and the protocol would also be based in STMP and HTTP. :) Best regards. -- Iñaki Baz Castillo [EMAIL PROTECTED] _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
