From line has ()

[Jethro R Binks]

On Fri, 17 Mar 2006, James Gray wrote:

> > It doesn't.  The RFC advises that you SHOULD do something a particular 
> > way, but does not forbid you from doing it another way if you have 
> > strong reasons for doing so
> 
> <SARCASM>
> Really?  I hadn't realised the RFC's WEREN'T ratified ISO/IETF/IEEE standards.  
> Thanks for pointing that out.  BTW, did you know the sky is blue on a clear 
> day?
> </SARCASM>

We are only speaking of the context of the RFCs, and the Terminology of 
section 2.3 employed by them.  MAY SHOULD MUST etc all have particular 
meanings, and it is that context I say 'forbid'.  SHOULD in the RFC means 
"you really should do this, unless you've really thought about what will 
happen".  Not "the IEEE will spank your ass if you don't".  Amusing though 
that might be.  In summary, I use the term 'forbid' here in the same 
context as you use the term 'violate', below.

> Just so you don't get confused (and start flaming me again): my initial 
> response regarding Exchange's compliance with the RFC's was based on the 
> incorrect assumption the error was in the envelope "MAIL FROM" but 
> wasn't being rejected until after the DATA section.  As you have stated, 
> whilst that behaviour is not "violating" the RFC per se (hard to violate 
> something that isn't necessarily enforceable), it DOES make your MTA 
> quite difficult to communicate with from a client's perspective.
>
> Now before you throw my own contradiction back in my face: I used 
> theterm "violate" in my original to suggest "does not follow the 
> accepted normal MTA behaviour as outlined in the RFC's".  You can't 
> really (literally) "violate" ANY RFC as they are not ratified standards.  
> RFC's may form the basis for an ISO/IETF/IEEE standard, but the RFC 
> itself is not enforceable.  Any admin knows that.

I am not flaming, I am clarifying and correcting (but some might 
say not enough clarity of my own).

If some guidance tells you should do something, and you claim to be 
compliant with it, and then do something it tells you not to do, then it's 
a violation, regardless of how enforceable it is!  If the RFC one claims 
to follow says MUST NOT and one does, then it has been violated.  Not that 
there is much anyone can do about it, except to tell you so and not 
exchange mail with you (in this case).  Hence the existence of DNSBLs that 
list 'non-RFC compliance' of various sorts (to bring this vaguely back to 
relevance to this list).

Jethro.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks
Computing Officer, IT Services
University Of Strathclyde, Glasgow, UK