Re: [tcpm] Fwd: New Version Notification for draft-gont-tcpm-urgent-data-00

On Sunday 09 November 2008, Joe Touch wrote:
> > Man pages? We are talking about what is sent on the wire. We're talking
> > about the semantics of the urgent pointer. Why should, e.g. a UNIX man
> > page state whether the urgent pointer points to the last byte of urgent
> > data vs. the byte following the last byte of urgent data??
>
> The UNIX man page should explain where its implementation differs from
> the spec. The BSD source code does, FWIW.

What Fernando says (as far as i understand) is that we have a reality here 
where everyone does the X thing (where X may be "wrong"). This is a reality 
and we have to accept it.

After accepting that this actually happens, we must either (a) persuade 
everyone to change their implementation or (b) update the RFC.

It seems that (a) isn't possible since it will introduce a backward 
incompatibility and practically invalidate the urgent pointer as a whole.

Since everyone does the X thing we can easily update the RFC without 
practically changing the actual meaning (or interpretation since everyone 
does it already the other way) of UP.

RFCs MUST be at sync with the real world. Having RFCs that don't represent 
reality makes RFCs as a whole (or at least TCP related) less valuable and 
less respectable. This will also save time and pain from all current and 
future TCP implementors.
_______________________________________________
tcpm mailing list
tcpm@ietf.org
https://www.ietf.org/mailman/listinfo/tcpm