On Wed, 22 Nov 2000, Richard Archer wrote:

> The point is that the DNS fails miserably as a directory service.
> 

Hi Richard,

I'm not sure that this is the point. In anycase
imagine for a moment the Internet was run
as an Novel Directory Service, Active Directory Service,
X500 directory service or any supposedly 'state of the
art' purpose build directory service you can name.

I can hear people making the same point you
make. Meltdown - total failure, doesn't scale,
can't replicate or whatever.

Changing DNS resolver API calls ("gethostbyname" or
equivalent or adding some pre-resolution preparation
mechanism) in almost all Internet applications
(as suggested in the draft) just ain't going to
happen in my opinion - this is a bigger can of worms than
dealing with the DNS issues in the first place - not to
mention the actual design of a directory service that
people would agree on in the first place.


and :

wrote: Ian Johnston <ian.johnston&#167;setel.com.au>

> Domain name policies have limited the ability of users to
> locate information using generic and geographic names
> through the use of URLs based on words
> and names of places, which are easily remembered, intuitive,
> meaningful, well known or easily recognisable.

So for *some* things the current DNS system actually
works beautifully as a directory service. What could
possibly be easier than entering microsoft.com
or oracle.com to get in contact with those company's.

Ok I know that doesn't help the rest of us and we all
know about better directory services but while there is
so much utility (either by design or good luck) in the
DNS system I can assure you it isn't a total failure
as a directory service - and hence the premium for
prefered domain names.

I guess we have to agree to disagree on this - back to
the wall!

regards doug