"The other DCCP" [was: Re: [dccp] dccp spec review (Rescorla)]

[Simon Leinen <simon@limmat.switch.ch>]

Eddie Kohler writes:
> [Eric Rescorla:]
>> 2.3 Confusing Terminology
>> -------------------------
>> The authors have a tendency to use "DCCP" to mean "DCCP
>> implementation", as in "The other DCCP will send an Identification
>> option to resync.".  This strikes me as pretty confusing. I'd
>> prefer to see a distinction between the protocol and the stack.

> We should fix this up. We've alternated between terms like "DCCP",
> "endpoint", and "implementation". I'd prefer "endpoint", as in: "The
> other endpoint will send an Identification option...." Any other
> preference?

Frankly, this kind of ambiguity wouldn't bother me at all, because it
is very easy to derive from the context whether "DCCP" refers to the
protocol or an endpoint.  This use of language has been common in the
past when talking about TCP, and many of us find it quite natural.

I'm afraid you're solving a non-problem here, at the cost of actually
making the specification language *more* complicated.

And now I'll get back in my box...
-- 
Simon.
(who likes to program in a language that allows him to give the same
name to a variable, a function, a class... *shudder* how ambiguous :-)

_______________________________________________
dccp IETF mailing list: dccp@ietf.org
list info:  https://www1.ietf.org/mailman/listinfo/dccp
wg charter: http://www.ietf.org/html.charters/dccp-charter.html