[OpenID] Interop

[Chris Drake]

Hi Martin,

Friday, June 8, 2007, 6:26:16 PM, you wrote:

MA> XRI resolution is done over HTTP, and so depends on the DNS albeit
MA> in a purely internal way.

There's more than one way to resolve XRI's - one of them does work
over HTTP, but there's others that work different ways (eg: EPP), and
I imagine there will be more in future as well (if there's not already
more that I don't know about).  Browsers will support XRI natively in
future, so it remains to be seen how they implement the resolution.
With all the fuss everyone's made about DNSSec, it seems likely to me
they're probably going to pick something that's more secure than plain
DNS/HTTP when it comes out.

Kind Regards,
Chris Drake


Friday, June 8, 2007, 6:26:16 PM, you wrote:

MA> Guido Sohne wrote:
>> 
>> It looks more and more like XRI could be a replacement for DNS itself?
>> If DNS information were subsumed into the XRI databases, you could
>> simply toss the DNS system in the bin ...

MA> XRI resolution is done over HTTP, and so depends on the DNS albeit in a
MA> purely internal way.

>> And the same problems are cropping up again. I believe what we are
>> discussing is akin to the multiple roots issues (regular DNS versus
>> OpenDNS) seems strikingly similar to (xri.net versus 'some other'
>> provider).
>> 

MA> Perhaps, but XRI at least has syntax to accomodate alternative "roots".
MA> The result doesn't look as pretty as an i-name, but it is unabiguous.
MA> The same cannot be said for alternative DNS roots, which can and have
MA> had namespace collisions.

MA> You could argue that using DNS names as disambiguators in a non-iname
MA> XRI still has DNS's problem of alternative roots, but XRI itself is not
MA> making this problem any worse.


MA> _______________________________________________
MA> general mailing list
MA> general at openid.net
MA> http://openid.net/mailman/listinfo/general