[NNTP] 64-bit article counter extension strawman

[Forrest J. Cavalier III]

> Saying "there's no limit and you should be prepared for any length of
> article number" says to client authors "you must include a bignum
> library."  That seems a little excessive, and in practice I doubt clients
> will actually do this.  I think we need to be very careful about what we
> say about allowed or recommended client behavior if we really don't
> specify any limit.

Put in a NOTE: that the WG considered 64-bit article numbers to be adequate
to handle X articles per day at 20% growth/year until YYYY.

I suppose we could argue about the growth rate, but really, the use of
the protocol cannot grow faster than available bandwidth or CPU power.

If BIGNUMs weren't so messy to store and load, they would be a good choice.
They are not terribly expensive in code size, execution data size, complexity,
or CPU cycles.   Using them creates portable code, instead of the mish-mosh of
ways to declare integer types in C wider than 32 bits on different platforms
and compilers, not to mention all the different ways sscanf and printf and
friends do it on different platforms.

But my guess is that with a note like the above, most implementors would
do 64-bits in the likelihood that they would be deceased before the limit were
reached, modulo any successful cryogenic or life-extending technologies.