Re: [xliff] Proposal to add REQUIRED id (NMTOKEN) to <metadata> and <metaGroup>

["Dr. David Filip" <David.Filip@ul.ie>]

Yves, it is the same old story for all terminal elements, it is not as bad as missing ids on the group element. Still you are forcing people who want just to to reference to enrich to be able to do it.

Re mda usefulness, we are using mda for all sorts of stuff that we had before in PIs and in phases that got killed, for instance workflow metadata.

Rgds

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

http://www.cngl.ie/profile/?i=452

mailto: david.filip@ul.ie

On Tue, Jan 14, 2014 at 3:18 PM, Yves Savourel <ysavourel@enlaso.com> wrote:

Hi Bryan, all,

 

The problem I have with this proposal is to have the id being required.

 

In a URI fragment the references to those item are ‘terminal’ (leaf) parts and not ‘container’ (like group), so not having Ids is not problematic because they are used only when they exist.

 

If a tool uses pointers to those elements it is very likely because it created them in the first place so it can put the id if it needs the id.

 

Another aspect (and one probably more worrisome):

We have an appliesTo attribute for those elements. So things are becoming quite confusing: You can have an <mrk> in the target only that points to a metadata and the appliesTo for that metadata be set to ‘source’… it’s contradictory.

 

And please, let’s not add PR or constraints to tighten up things: 2.0 is becoming extremely difficult to validate. Checking that appliesTo is set only when something is not pointing to its id for example would be just about impossible to verify.

 

Do we have use cases for this module? I mean actual real life implementations doing something useful with it? So far the examples I’ve seen look a lot like skeleton-type data (which would go against the PR #2 of the section 4.9.3).

 

But I’m getting off-topic. For this proposal I’d rather have the id be optional.

 

Cheers,

-yves

 

 

From: xliff@lists.oasis-open.org [mailto:xliff@lists.oasis-open.org] On Behalf Of Dr. David Filip
Sent: Tuesday, January 14, 2014 3:17 AM
To: Schnabel, Bryan S
Cc: xliff@lists.oasis-open.org
Subject: Re: [xliff] Proposal to add REQUIRED id (NMTOKEN) to <metadata> and <metaGroup>

 

I second this proposal, it seems uncontroversial to me.

No referencing of the mda data would be possible w/o a wildcard mechanism that we discarded as an option in the last meeting.

 

Cheers

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

http://www.cngl.ie/profile/?i=452

mailto: david.filip@ul.ie

 

On Mon, Jan 13, 2014 at 11:28 PM, Schnabel, Bryan S <bryan.s.schnabel@tektronix.com> wrote:

As a result of our Fragment Identifier resolution, I propose we a add REQUIRED id (NMTOKEN) attribute to <metadata> and <metaGroup>.

 

Please let me know if you disagree. If this turns out to be noncontroversial, I will ask for a second at the meeting tomorrow.

 

Thanks,

 

Bryan