Hello gang,
	I signed up with the listserv just a week ago and I've been watching the
discussions about the new numbering.  Although, I doubt my opinion counts
for anything.. I'll give it anyway.  I haven't really researched this topic
yet, so I apologize in advance if these points have already been made.  So,
as the song goes: "I will dispense this advice, now:"

I do development work, and often need to come up with identifier schemas
that maximize the amount of possibilities as well as maintain a certain
level of knowing what the item refers to, without having to know its full
description.

One possibility (type based identifier):  I would reserve the first
character for type of cache.  Here are some examples: T=Traditional,
L=Locationless, V=Virtual, M=Multi, U=Unkown etc.  Then the remainder 5
characters could be used for ID.  (Of course, you could dramatically
increase the number of ID's by going with a base 31 as has been discussed.)
If we later run out of space for T(raditional) cache numbers, you can simply
assign a secondary character, say B.  This would be similar how we used to
only have 1-800 numbers, but we ran out, so now the prefix could be 800,
888, etc.

Another possibility (location based identifier): Reserve the first 2
characters as location indicators: AZ, PA, TX (states)  and CH, IL, DU, RU
(countries, etc.)  This does however unfairly limit the number of caches
within a state or country.  As we all know, AZ would probably need more
numbers than RI. ;)

Thanks,
TheWebbman
Webb Pickersgill