Christian Huitema writes:
> We did in fact have this discussion in depth 2 years ago, when we
> submitted the draft to the working group.
I'm skeptical. I can't find server-side indirection discussed anywhere
in the archives. Please provide a verifiable reference.
> The motivation for A6 is the key database paradigm that atomic
> information (e.g. a site prefix) should be stored exactly once.
Not just a ``paradigm,'' but a ``key paradigm''! Wow. I see the light!
We have to abolish PTR records, because they repeat information from A
records. Clients that want reverse lookups will have to try forward
lookups on random names. Information should be stored exactly once!
Perhaps you are unaware that there are entire books on the design of
serious databases: credit-card databases, airline reservation databases,
etc. All the designs violate your ``key paradigm'' in a huge number of
ways. Information is stored more than once for the sake of security
(e.g., double-entry bookkeeping for auditors), reliability (e.g.,
transaction logs), speed (e.g., caching, reverse indices), etc.
> This has major consequences on ease of update, as well as caching.
These are certainly issues to consider. They deserve rational analysis,
not ignorant religious nonsense like ``storing atomic information
exactly once is a basic tenet of database theory.''
---Dan
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
