On Tue, 26 Nov 2002, Charles E. Perkins wrote: > Pekka Savola wrote: > > I wouldn't call one organization _assigning to end-users_ prefixes > > from the space of about 2^28 perfectly good, not even _good_. > > > > Quite the contrary. > > You forgot to say why. Plus, depending on the constraints, the > exponent could conceivably be selected to be a number larger than > 28, depending on how many subnets a site-local allocation is > expected to have by default. Evan at 2^28, with an average of 10 > seconds between allocations, that would last over 80 years.
I thought the "why" for no to flat-space end-site allocations by IANA was self-evident. How do you do reverses in an automatic manner (that's about the most significant reason for totally unique, and that's by a long stretch)? What data would you require for an assignment? What do you do when everyone requests assignments? What to do when someone sets up a fancy script which requests assignments at the pace of 1000 per second, just for fun? "No infrastructure" is just so much better. -- Pekka Savola "Tell me of difficulties surmounted, Netcore Oy not those you stumble over and fall" Systems. Networks. Security. -- Robert Jordan: A Crown of Swords -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
