I am working on a new protocol to be published on the unprotocols.org site. I 
am following the template as given by the majordomo protocol [1]. One thing 
about the protocol definition bothers me.

All of the message types define frame0 as a null message part. The assumption 
is that this protocol only makes sense with REQ/XREQ sockets which always send 
a null message part first. So the part that bothers me is that the *socket 
framing* is part of the majordomo protocol definition.

Why is that necessary? If an implementer is using a REQ socket, they get that 
frame "for free" and never even see it. If they use XREQ, then they must know 
to prepend all of their messages with a null part anyway.

Confused...

cr

[1] http://rfc.zeromq.org/spec:7
_______________________________________________
zeromq-dev mailing list
[email protected]
http://lists.zeromq.org/mailman/listinfo/zeromq-dev

Reply via email to