Re: Registration of acct: as a URI scheme has been requested

On 6/23/12 8:24 AM, Michiel de Jong wrote:
> It's a premise of webfinger that we resolve a human-memorable string
> of the form 'user@host' to accounts.

Yes, this is the crux of the matter.

Historically, there's always been a tension between natural keys (human 
readable) and system generated (non human readable) keys. As the Web 
evolves into a more structured data space comprised of data objects, the 
same tensions are resurfacing.

Opaque de-referencable keys (URIs) [1] require resolvers. For http: its 
in-built and widely adopted. For others, a resolver protocol is 
required. Webfinger [2] and Fingerpoint [3] .

What applies to acct: scheme URIs also applies to doi: lsid: etc.. The 
only issue is the cost of making a resolver that's adopted (primarily) 
by user agents of type: Web Browser.


Links:

1. http://www.w3.org/TR/webarch/#pr-uri-opacity -- URI opacity
2. http://hueniverse.com/2009/08/introducing-webfinger/ -- Webfinger 
(ignore the "account" and "subject identity" conflation)
3. http://buzzword.org.uk/2009/fingerpoint/spec-20090822.html -- 
Fingerpoint .

-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen