Re: [Action-31] Clarify definition of implementations as consumers and producers as basis to defining the requirements for testing two implementations

Hi all, again re-posting for Dave, who is having e-mail authorization
issues:
----------------------

Hi,
Some clarification on this issue. We intend to follow the good practice
of ITS1.0 and have a clear conformance section such
as;http://www.w3.org/TR/its/#conformance

This section includes both statements about conformance to specific data
categories and statements about the defaults / inheritance / overriding
characteristics of a conformant implementation.

However, in addition it may be the case, especially as the processing
and stakeholder of MLW-LT is wider than ITS1.0, the context in which
specific data categories are used and interpreted will not be the same
for all data categories. for instance some may be relevant when pass
content from creator/authors to localisaters, some for passing content
from localisers back to publishers, plus other relationships e.g.
to/from LT functions and workflow/QA management functions.

Therefore we may need a clear conceptual framework that classifies the
type of consumer and producer so that the semantic of the data category
can be unambiguously understood by the anticipated wider range of
implementors, but also in a way that does not require a lot of
additional information to be absorbed by someone interested in
implementing only one data category. For example, the end-to-end process
flow in the current requirements document
(http://www.w3.org/International/multilingualweb/lt/wiki/Requirements#Support_an_End-to-End_Use_Case)
goes some way to communicating this, but obviously is presented as a
specific, albeit fairly general purpose workflow.

The issue is therefore, what level of such contextual information needs
to be included in the specification and to what level should it be
normative or informative? More specifically, would defining a minimal
set of abstract produce and consumer roles for the processing of content
with MLW-LT compliant meta-data address this issue.

This was discussed at the break out session in Luxembourg and the
concensus was that while this might be useful in situations where the
workflow was entirely internal, for workflow crossing organizational
boundaries it would be problematic. This is because the WG could end up
in the position of trying to impose business roles on parties involved
in commercial negotiations e.g. LSPs and their clients, which would be
difficult to achieve consensus upon. So such consumer/producer role
would be useful in some circumstance, but could not be made normative
compliance features, just informative, supported by best practice.

Regards,
Dave

------------------------


Dr. David Filip
=======================
LRC | CNGL | LT-Web | CSIS
University of Limerick, Ireland
telephone: +353-6120-2781
*cellphone: +353-86-0222-158*
facsimile: +353-6120-2734
mailto: david.filip@ul.ie



On Thu, Mar 29, 2012 at 12:30, Dr. David Filip <David.Filip@ul.ie> wrote:

> Hi Dave, I understand that you are swamped, still this one seems quite
> important with loads of dependencies and potentially big impact, In case
> you cannot start working on this one this week, please propose another
> owner.
> If anyone is interested in owning this one, please contact Dave and take
> over from him.
>
> Thanks
> dF
>
> Dr. David Filip
> =======================
> LRC | CNGL | LT-Web | CSIS
> University of Limerick, Ireland
> telephone: +353-6120-2781
> *cellphone: +353-86-0222-158*
> facsimile: +353-6120-2734
> mailto: david.filip@ul.ie
>
>