W3C home > Mailing lists > Public > public-credentials@w3.org > March 2016

Re: Version Control / Cred-Ledger

From: james anderson <james@dydra.com>
Date: Fri, 4 Mar 2016 06:53:03 +0000
Message-ID: <010201534066ab51-564e21f4-f9bd-44ef-a3c7-9e54fb58878d-000000@eu-west-1.amazonses.com>
To: W3C Credentials Community Group <public-credentials@w3.org>
good morning;

> On 2016-03-04, at 07:34, Timothy Holborn <timothy.holborn@gmail.com> wrote:
> 
> Where a credential is updated as a counterpart to a series that may benefit from provenance inspection and/or ledger like functionality, what are the considerations?
> 
> The theory is that each-time the document is updated, perhaps a reference to 'latest' and/or historical may be a better approach vs. invalidating the out of date document without means to look-up and related documents...?
> 
> the other aspect within there is also invalidation of document.  Whether that means it's simply valid or not, or whether it can still be valid for a period of time.
> 
> I think law is a use-case for examining that concept; where acts are passed updating the law, however law as it relates to a historical act may relate to the former version.

these are typical temporal modelling concerns.

do you expect to require anything more than a typical bitemporal model would support?
that is, will it be necessary to say more than
- when an assertion was made
- over which internal was the assertion then said to apply

that any credentials process must be bitemporal is a significant assertion.
the word appears in any thread on this list and google does not believe it appears in any w3c discussion on the credentials topic.

still, if there is nothing to note beyond the two concerns, above, it may be more a functional requirement at the implementation level than a conceptual concern. 

best regards, from berlin,



---
james anderson | james@dydra.com | http://dydra.com
Received on Friday, 4 March 2016 06:53:42 UTC

This archive was generated by hypermail 2.3.1 : Friday, 4 March 2016 06:53:43 UTC