November 30, 2010

Announcing Project IDONS: Internet Distributed Open Name System

IDONS Discussion Forums

Blog Update (December 9, 2010): IDONS (Internet Distributed Open Name System) Discussion Forums Available - Your Participation is Invited

Greetings. As regular readers know, I am a vocal critic of ICANN's plan for a massive expansion of new Top Level Domains (TLDs), and of the increasing abuse of the domain name system both for the profit of the "domain-industrial complex" and for Internet censorship and control purposes -- among other related concerns.

Recently, in Take a Tiny First Step Toward Controlling Your Internet Addressing Destiny, I noted:

"An alternative Internet name to address mapping system -- fully distributed, open source, fault-tolerant, secure, flexible, and not subject to centralized constraints, meddling, and censorship -- will take significant time to develop, and a long transition period for deployment."

And I asked readers to take part in some initial experimental activities (thanks to all who responded).

What I didn't say then -- in the hope of having this project a bit farther along before any public announcements -- is that I've been talking to my colleagues and others about this issue for quite some time, and I believe it's fair to say that we've agreed that it is not only necessary to move beyond the current DNS and naming environment, but that such a project is entirely practical -- if managed in an organized and reasonable manner.

Ad hoc attempts to bypass the existing system (such as those newly proposed by Pirate Bay) are likely to create fragmentation and confusion, and therefore ironically tend to further entrench the existing system.

But, even though it may seem on its face like suggesting that the electric grid move from A.C. to D.C. throughout, a fundamental evolution in the way that we handle names and addresses on the Internet is an idea whose time has come.

The scope of the project on which I've been working, which I call IDONS - Internet Distributed Open Name System -- is in early stages, but would ultimately be significant both in terms of technology and time. It may perhaps be reasonably compared with the scale of IPv6 deployment in some ways.

For reference only, since working documents have evolved beyond this, the original set of objectives for IDONS included:

- Fully distributed

- No centralized control

- Fault tolerant

- Open source

- All communications encrypted

- Local databases optionally encrypted

- Extensible and expandable to any foreseeable size and loads

- Finite, deterministic time for “settlement” of new name/identifier registration requests across the entire IDONS space

- Gateways during extended transition phase for interoperation with non-upgraded network segments, clients, etc.

- Decoupling of locally-defined (e.g. human-meaningful) names and related addressing elements (aliases, etc.), from actual “random string” network identifier objects and constructs

- Minimal constraints on name selections and changes

- No central registries

- No registrars

- No fees nor charges necessary for any name or address operations across IDONS

There's a whole lot more involved of course -- both technical and nontechnical. Nor do I believe that such a project can succeed without serious community support and significant funding. Like I said above, ad hoc won't fly for this.

Enough for now.

Interested parties are invited to contact



IDONS Discussion Forums

Blog Update (December 9, 2010): IDONS (Internet Distributed Open Name System) Discussion Forums Available - Your Participation is Invited

Posted by Lauren at November 30, 2010 03:58 PM | Permalink
Twitter: @laurenweinstein
Google+: Lauren Weinstein