Re: The DNS is not a directory

[peterd at Bunyip.Com (Peter Deutsch)]

[* Boy, I wish I knew who all these lists were, so I could
   trim a couple in confidence! :-) *]


Hi Dave,


[ You wrote: ]

> At 10:18 AM 4/23/98 -0400, Peter Deutsch wrote:
> >DNS is currently used as a directory system. To quote the
> >line from E.T. "this is reality, Greg". You may not like
> 
> Peter,
> 
> The DNS is used less as a directory system than current mythology claims.
> In fact, I claim it isn't used as a directory system at all.
> 
.  .  .
> 2.  Most of the successful pseudo-directory use of the DNS today is
> predicated on having the domain name end in .com...
>.  .  . (There is an IMAP email client, called Simeon, made by
> Esys.  Quick.  Guess their domain name.)

What I would do is try a couple of guesses and if that
fails, I'll get serious and find it some other way, but:

#define personal_annecdote ON

I've tried guessing twice in the past 24 hours and it worked once.
That's just often enough that I'll keep trying.

#define personal_annecdote OFF

More importantly it works often enough that for this and
many other reasons people with entries in DNS want to
choose their own names. If we try to ban them from doing
so they will route around us, just as we all routed around
computing centres ten years ago.

FWIW, from what I can tell I agree with the spirit of ARIN
but the technology of the would-be TLD/registry industry.
Not because I want a TLD/registry industry, but because
they're advocating the Internet way.

FWIW^2, I actually believe that the ARIN approach is fine
for IP addresses (a scarce resource that should be managed
as we manage the radio spectrum) but think we can, should
and must offer far more choice for DNS and (eventually)
real directory services. I hope this isn't schizophrenic
or inconsistent of me.



.  .  .
> ps.  As I've commented to you privately, I think that the whois++ concept
> is quite wonderful, since it allows laying a relational data model on top
> of a hierarchical (and other) storage mechanism.  That is exactly what the
> DNS is, and so it would be a very interesting exercise, indeed, to develop
> a whois++ layer on top of the DNS.  That new layer very much WOULD BE a
> directory service.  Even better, this added functionality can be created
> incrementally, without disturbing the installed base of data or service.
> That's would be an stunning accomplishment and exactly what the Internet
> loves, since we like to protect the installed base.

Thanks Dave. I've tried very hard to not shill for any
specific technology here, but I agree with what you just
wrote and strongly believe we can build real directory
services using Internet concepts. I've also enough faith
in the system that when they're really needed, they will
appear.  Meanwhile, I distinguish between what people
*want* and what they *need*, since we all wants lots of
things but when push comes to shove, tend to pay for what
we need.

And a final thought, if half the energy that's been
deployed in the past two years on DNS fights had been
deployed on Directory Service implementations using
existing technologies, this problem probably wouldn't
exist now. I think the problem is not the answers we're
getting, it's the questions we're asking.

To sum up my position on this thread, at issue here is
whether proposals to limit DNS participation through such
mechanisms as a) limiting the number of TLDs, b) forcing
people into specific TLDs, c) whatever, can be made to
fly. I most strongly believe that doing so would be the
equivalent of violating Bernolli's principle and we should
accept the laws of physics on this and move on.

Wasn't it Milo Medine who said "given enough thrust you
can put a fridge into orbit?" Maybe you can, but at that
point what you're doing is not flying, it's "falling with
style"... :-)


				- peterd



-- 
------------------------------------------------------------------------------
     Peter Deutsch,                                   (514) 875-8611  (phone)
  Bunyip Information Systems Inc.                     (514) 875-8134  (fax)
    <peterd at bunyip.com>                               http://www.bunyip.com

 "This document must be read and contemplated in its entirety in order to
 understand its meaning and intent. This is because many concepts in the
 characterization of SR&ED are interrelated and cannot be applied in isolation.
 Quoting extracts out of context is often inconsistent with a holistic
 interpretation. "
		- Revenue Canada Income Tax Information Circular No. 97-1,
                  "Scientific Research and Experimental Development -
                  Administrative Guidelines for Software Development"
----------------------------------------------------------------------------