RE: Notary services requirements -- directions?

["Carl Wallace" <cwallace@xxxxxxxxxxxx>]

> But, well, before making a title, it seems to me that we may 
> first come to an agreement  about *what* protocol we are 
> doing, i.e., the ones between machines that 'certify' 
> something, my view:
> 
> - act as a more or less automatic tool to perform some defined
>   validation activity
> - a to 'certify' the result, i.e. authenticating the data.
> - used in workflows.  
> 
> Note that protocol is meant in a large sense, not just the 
> 'access protocol', there are components for auditable 
> operational requirements, i.e. a service must play the game 
> in the right way, and cannot just pretend to do something. 

It seems like we are settling on a few avenues (with existing protocol
correlations in parentheses):

	- long-term signature formats (e.g. ERS)
	- long-term signature format protocols (e.g. machine-to-machine a la
WebDAV or DVCS variant)
	- server-based data certification protocols (e.g. machine-to-machine
a la DVCS)

Is there an additional need to define means for representing notarial acts
performed by humans?

Has anyone been following the OASIS work that was discussed during the San
Diego meeting?