Re: smashing down callsigns into 32 bits
- To: "Brandon S. Allbery" <email@example.com>
- Subject: Re: smashing down callsigns into 32 bits
- From: "Louis A. Mamakos" <firstname.lastname@example.org>
- Date: Sat, 03 Dec 1994 09:25:03 -0500
- Cc: email@example.com
- In-reply-to: Your message of "Fri, 02 Dec 1994 17:14:18 EST." <m0rDgF9-0002iMC@kf8nh.wariat.org>
- Sender: firstname.lastname@example.org
> In your message of Fri, 02 Dec 1994 11:49:28 EST, you write:
> | determine permutations. The proper value of a field AA9AAA is
> | 26^5 * 10, noting of course other formats too so more likley 27^5
> | and when you add non-US signs like C21 and 4N7 it gets a bit bigger...
> First approximation: 36*37*10*(27**3) --- you can save a little by creative
> handling of shorter callsigns, perhaps, and/or replace 27**3 with 26*(27**2)
> at the risk of offending JY1 :-)
Of course, since the IP addresses have topological significance, you
can't just come up with a transform into an opaque address. So it
really doesn't matter what type of hash function you use, or how
efficient it is since you can't use the result in the real world.
This parallels the usual discussion about how IP network numbers are
in short supply, and how inefficiently they are being used. That's
because they're classful addresss (A/B/C) rather than having an
explicit mask, and the fact that you route on them so you can't use
each and every address sequentially.
Louis A. Mamakos, WA3YMH email@example.com
Backbone Architecture & Engineering Guy uunet!louie
AlterNet / UUNET Technologies, Inc.
3110 Fairview Park Drive., Suite 570 Voice: +1 703 204 8023
Falls Church, Va 22042 Fax: +1 703 204 8001