ietf-nntp OVER extension

[Russ Allbery]

Clive D W Feather <clive at on-the-train.demon.co.uk> writes:
> Russ Allbery <rra at stanford.edu> writes

>> Why does this need to be optional?  It's a new command; we can require
>> that this be implemented.

> Some people were claiming that it would be too hard to add to their
> implementations.

I think that Andrew's point was well-made and that maybe we shouldn't add
this just for the sake of consistency when it seems rather unlikely that
any client will bother issuing such a command.

>> I'd like to see :lines and :bytes here instead.

> Any particular reason? I was aiming for something that was more clearly
> not a header.

> In fact, wouldn't "body:lines" and "article:bytes" be even better names?

Having them clearly not be a header is the reason why I'd prefer :lines
and :bytes.  The leading colon serves as an immediate signal that this is
something different, whereas with line:count or body:lines I have to scan
forward to see the colon and for some reason it just looks odd to me.

Maybe I'm too much of a programmer, but I'm very used to leading
punctuation to indicate something that's in a special or reserved
namespace (_variable in C, *variable* in Lisp, @FOO@ in autoconf
substitutions, $SHELL variables, etc.).

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>