[OpenID] W3C TAG recommends against XRI

[Drummond Reed]

Good point, Peter. I'm sitting in a room of LDAP and AD architects at
Microsoft right now and your theory is right on - because of their directory
and X.500 background, they get the value proposition for XRI in seconds.
This proves out the XRI Technical Committee's experience that, while Simon
is right that we have yet to come up with a compelling way of getting
developers _who do not have one of the problems XRI solves_ to easily
understand the value proposition, the converse is also true: if a
developer_has_ one of those problems, then they can turn into an XRI
evangelist very quickly (like the Boeing directory team that is now very
active on the XRI TC).

 

Speaking of persistence, I think it's widely recognized the primary reason
XRI was included in OpenID Authentication 2.0 (besides XRDS for discovery)
was for its deep structural solution to the OpenID recycling problem. Yes,
once the seriousness of this problem was recognized, the hash URIs solution
was developed so that OpenID wouldn't be completely dependent on XRI
architecture to solve the problem. But most OpenID architects I know
acknowledge that XRI i-name/i-number synonym architecture is a cleaner and
stronger solution - for example it enables a OpenID user to completely
abandon an XRI i-name (or a URL) while still retaining full control over the
XRI i-number, for the XRI i-number to be portable and fully and
independently resolvable from the i-name, neither of which you can do with
hash URLs.

 

It's no secret that other Internet scale identity frameworks also need these
same capabilities, which is why XRI and XRDS is already the core identifier
and discovery architecture for the Higgins project, and discussions are
active with the information card folks.

 

In short, us XRI folks are not crazy (we hope). We have been slow to figure
out how to communicate about XRI technology (hell, we were plenty slow in
developing it ;-), but we sure know why were are so committed to it.

 

We invite the OpenID community's help in how we can communicate better about
it.

 

=Drummond 

 

  _____  

From: general-bounces at openid.net [mailto:general-bounces at openid.net] On
Behalf Of Peter Williams
Sent: Friday, May 23, 2008 11:15 AM
To: Simon Willison
Cc: OpenID List
Subject: Re: [OpenID] W3C TAG recommends against XRI

 

I have to be amongst the most stupid people on this list (if not the most
stupid), and I get XRI. How can this be? I have to assume Edd is 10,000*
better than I am at web stuff, given the intro he gets.

 

What's the difference between XRI and ldap? Precious little. 

 

Does Edd understand LDAP? If he does, tell him XRI is ..."LDAP updated" for
web culture.

 

Period. 

 

Actually, no! its not period. Ive no idea who Edd is or what XTech
conference is about. Is he an authority?

 

Perhaps, we need a directory of authorities, so we can distinguish between
blog blather and citation-grade statements that one might rely upon. Hmm.
There is that technology that does that.. Its called XRI... its an updated
form of LDAP, wouldn't you just know it. Its made for the web, rather than
the 1980s telco world (L)DAP was conceived for.

 

_________________________
Peter Williams

  _____  

From: Simon Willison
Sent: Fri 5/23/2008 10:09 AM
To: Andy Powell
Cc: OpenID List
Subject: Re: [OpenID] W3C TAG recommends against XRI

If Edd Dumbill (ex-managing editor of xml.com and long-running chair
of the XTech conference) can't get his head around XRIs then what hope
for everyone else?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-general/attachments/20080523/0321ac5f/attachment-0002.htm>