[105614] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

Re: what problem are we solving? (was Re: ICANN opens up Pandora's

daemon@ATHENA.MIT.EDU (David Conrad)
Fri Jun 27 17:21:16 2008

From: David Conrad <drc@virtualized.org>
To: "Scott Francis" <darkuncle@gmail.com>
In-Reply-To: <171423de0806271402p5e9e9be9l283233fad83cac75@mail.gmail.com>
Date: Fri, 27 Jun 2008 14:21:08 -0700
Cc: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org

On Jun 27, 2008, at 2:02 PM, Scott Francis wrote:
> what little assurance we have that e.g. bankofamerica.com is the
> legitimate (or should I say, _a_ legitimate) site for the financial
> institution of the same name becomes less certain when we have e.g.
> bank.of.america, www.bankofamerica.bank, www.bankofamerica, www.bofa,
> and other variants.

I agree, but we already face that problem now.  Is bankofamerica. 
{org,net,us} the same thing as bankofamerica.com?  I would agree that  
a flood of new TLDs would exacerbate the problem, but I suspect the  
difference is between a run over on a two lane street versus being run  
over on a five lane highway.  In both cases, you're road pizza....

> Perhaps the solution is to devalue names (through the introduction of
> some theoretically unlimited number of variants) to the point that
> users come to rely upon reputation-based systems (e.g. PageRank)
> exclusively.

I suspect the right answer is to rely not on reputation or labels, but  
rather stronger security credentials, e.g., valid X.509 certs, PGP/GPG  
signatures, etc.  Of course, that's been true for a while now.

Regards,
-drc



home help back first fref pref prev next nref lref last post