Merit Network
Can't find what you're looking for? Search the Mail Archives.
  About Merit   Services   Network   Resources & Support   Network Research   News   Events   Home

Discussion Communities: Merit Network Email List Archives

North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Namespaces (was: yet another dns namespace overlay play)

  • From: Joshua Goodall
  • Date: Thu Mar 08 04:25:42 2001

On Wed, 7 Mar 2001, Vadim Antonov wrote:

> I'll register
> The whole idea of unique human-readable names is broken (I would go as far
> as to say that the idea of any global name space is silly :)

It shouldn't be. The great mistake with DNS was allowing a hierarchical
network engineering convenience to *become* a flat namespace used as a
globally-unique identifier for bodies of data.

Historically we could pin this on a) the formulation and standardisation
of the URL and b) the existence of gTLDs.

A better method for addressing data would be based on source-brokered,
signed, distributed caches of keywords that can be search and, more
importantly, bookmarked in the context of each signer.

Thus removing the visibility of the server domain and eliminating the
fistfights over the abused body of the gTLD namespace, relegated DNS to
where it belongs - a name-to-IP-address mapping.

> The only real solutiuon to the present and future DNS woes is to replace
> it with the hyperlinks, portals, address books and search engines - and
> _no_ human-readable names.  This effectively creates as many "roots" as
> there are users.  My "John Doe" is not the same as your "John Doe" :)

A similar suggestion, I think - but turn it around so that sites are
self-brokered. Portal sites then become indexers of indices. The search
engines are, in a way, creators of meta-data.

> PS	I know, I know, it is politically impossible to abolish DNS
> 	wholesale.  Any progress is politically impossible.

Nothing is politically impossible. The days of neutral community
innovation are not over. If you say it over and over again, you can even
believe it.

We raised and developed these ideas at the last RIPE meeting with a few
colleagues & friends (under the pretext, yes, of "DNS Sucks"). I think the
concepts are sound but acceptance is hard; one can't accomplish this with
a proprietary, licensed, patented product.

- Joshua

-[ Joshua Goodall ]-----------------------------------------------
-[ Chief Systems Architect, IP R&D ]----- Cook, Geek, Lover ------
-[ ]--------------- --

Discussion Communities

About Merit | Services | Network | Resources & Support | Network Research
News | Events | Contact | Site Map | Merit Network Home

Merit Network, Inc.