Re: [ISSUE-6] Re: [ACTION-123]

(co-chair hat on)

Dave, all, I know that there is interest in the group about the process
model topic, but we shouldn't do step two before we are sure about step
one: There are lot's of missing pieces in the current ITS 2.0 draft, see
http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jun/0133.html
and there are lots of data categories that need input in the upcoming
weeks, see
http://www.w3.org/International/multilingualweb/lt/wiki/Implementation_Commitments

Please, use your time (I know that everybody is busy before the summer
break) to fill these gaps asap, before moving other (even non normative)
discussions forward. I esp. disagree with putting effort into
"To go forward in firming up these definitions, I propose we add some input
and output plus some conditions on these", before the gaps mentioned above
are filled.

Thanks,

Felix

2012/6/25 Dave Lewis <dave.lewis@cs.tcd.ie>

>  Hi Moritz,
> Thanks for following up on this action, which addresses the ongoing
> ISSUE-6 - let keep discussion under this thread.
>
> First we should remember the role such a model should play in ITS2.0,
> specifically we need process values for both the readiness and the progress
> data categories. These values should also be used in a consistent way for
> the value of the activity object in the PROV WG proposed for stand-off
> provenance mark-up.
>
> However, as discussed in Dublin, this is a complex topic with lots of
> different views, so we are not aiming for a normative outcome, but some
> best practice document to accompany ITS2.0.
>
> To facilitate this I suggest we capture process definition separately  and
> hopefully agree a common set of process names using wiki page:
>
> http://www.w3.org/International/multilingualweb/lt/wiki/Metadata_and_workflow_comparison
>
> but do this in parallel to, and therefore not holding up, the normative
> data category definitions.
>
> I've moved Pedro's suggested process names onto this page but reworked
> them to:
> - provide names the might be more broadly accessible, and in the style of
> data categories (i.e. with hyphens rather than camel case)
> - provide a little more in terms of process definition
> - add a few processes I thought were missing
> - group under some headings, again to help communicate between different
> viewpoints, and provided some further structure through suggested subclass
> relationship between some process definitions.
>
> To go forward in firming up these definitions, I propose we add some input
> and output plus some conditions on these. Then we can rerun the exercise
> Arle started with the included table to cross referencing these again CRUD
> usage of data categories.
>
> At this point then, I'd asked interested people to review this page and
> either provide comments/revisions to existing processes or suggest other
> one we need.
>
> cheers,
> Dave
>
>
>
> On 22/06/2012 17:17, Moritz Hellwig wrote:
>
>  Hello,****
>
> ** **
>
> in Dublin we talked about processes and whether we can find a common
> (basic) set of processes that we can use for ITS 2.0. So I’d like to
> trigger a discussion to find if we can agree on such an ontology. I think
> it was Pedro who was kind enough to provide two models and a set of
> processes we can discuss:
> http://www.w3.org/International/multilingualweb/lt/wiki/Requirements#Process_Model
> ****
>
> ** **
>
> I favour Model 1. And I think there definitely should be a difference
> between processes that apply to the target, the source or both.****
>
> ** **
>
> Could we take the models and list as a model and collect which processes
> are needed from your side? Or can maybe be omitted?****
>
> ** **
>
> Cheers,
> Moritz****
>
> ** **
>
>
>
>


-- 
Felix Sasaki
DFKI / W3C Fellow