[OpenID] Wiki page: Attempting to document the "Email Address asOpenId" debate.

Hallam-Baker, Phillip pbaker at verisign.com
Wed Feb 14 17:03:43 UTC 2007


The mailto: scheme was not our finest hour.

When it was designed none of us really understood fully that we were conflating the method and the identifier. In particular we only defined methods for HTTP, not for other protocols.

In that case of mailto: the identifier apparently conflates the method (ASYNCHRONOUS-POST) and the resource identifier (alice at example.com).  If we had been thinking a bit further we might have defined an mechanism RFC822:alice at example.com and a set of methods (asynchronous-POST, SYNCHRONOUS-CONNECT etc.).


We can always define our own URI scheme: OPENID:alice at example.com

This can then map to multiple resolution protocols at the option of the registry providing the service as described through configuration data expressed through the DNS.


In this way OPENID becomes the lynchpin of the next generation Internet identity system embracing and extending SMTP, JABBER and Blogging while at the same time preserving current human factors.

> -----Original Message-----
> From: specs-bounces at openid.net 
> [mailto:specs-bounces at openid.net] On Behalf Of Robert Yates
> Sent: Wednesday, February 14, 2007 11:54 AM
> To: Stephane Bortzmeyer
> Cc: Claus Färber; specs at openid.net; general at openid.net
> Subject: Re: [OpenID] Wiki page: Attempting to document the 
> "Email Address asOpenId" debate.
> 
> On 2/14/07, Stephane Bortzmeyer <bortzmeyer at nic.fr> wrote:
> >
> > A real-world example is NAI (Network Access Identifiers) in 
> RFC 4282.
> >
> > Mine is bortzmeyer at net2.nerim.nerim. Certainly not an email address.
> 
> Is a URI scheme defined for RFC 4282?, I took a quick look 
> and couldn't find one.
> 
> As Phillip keeps pointing out [1] [2] there is an important 
> distinction between what the user is expected to type and the 
> canonical representation used by the machines. IMO, this wiki page
> *should* be about mailto: URI's as openids and not other 
> machine representations that happen to use the same user format.
> 
> Rob
> 
> [1] http://openid.net/pipermail/general/2007-February/001705.html
> [2] http://openid.net/pipermail/specs/2006-November/000811.html
> _______________________________________________
> specs mailing list
> specs at openid.net
> http://openid.net/mailman/listinfo/specs
> 



More information about the general mailing list