Re: Dead Internet Theory

Lawrence D'Oliveiro <ldo@nz.invalid> writes:

Richard Kettlewell wrote:

If you want globally unique names then a central naming authority is an
easy way to do it.

>
I’m not sure you could describe the DNS as a “central naming authority”.
Look at the proliferation of registrars, and the opening up of all kinds
of random new TLDs, obviously as a revenue-generating exercise.
>
Once you have a domain name registered, you are free to use the entire
namespace under it for your own purposes.

I did allude to this briefly in the previous post, but it can be
expanded upon. The authority is delegated, but it is delegated from a
central point. If the people who ultimately control the IANA root don’t
like your face, they can (with some inconvenience) stop you using your
chosen names.

You can certainly say it’s a social limitation rather than a technical
limitation: you can set up your own roots if you like. But approximately
nobody will pay any attention if you do. The social limitation is no
less real than a comparable technical limitation.

In contrast the PGP web of trust doesn’t have this property. There’s no
single point of control that can deny you use of the system. Certainly a
coordinating gorup of key server operators could make your life harder,
but your neighbours’ signatures on your keys still exist.

UUCP bang paths had a weaker (totally unsecured) form of the same
property.

The nearest you can get in PGP to a single centralized point of control
is the software implementations. But there are several and even though
one is fairly dominant, its open source nature means that any attempt to
exploit its position would be highly visible. So again there is a social
vs technical point here: the GnuPG authors could, technologically and with
incomplete but nevertheless wide coverage, ban you from using your keys;
but for social reasons this is extremely unlikely to happen.

--
https://www.greenend.org.uk/rjk/