RE: [xliff] Extensibility in Modules

Thanks David for the reply. Of course, you are right on all points. Playing devil’s advocate, I just think the spec itself can be interpreted differently since it does not clearly state what you just stated below. *Simple* and *Glossary* can be interpreted quite different from *complex* and *terminology* in my dataset. In any case, we will support <mda:metadata> on <gls:glossary> as an extension in true spirit of standards and interoperability. And I’m not worried about it being removed, although some implementers might. Again, it’s too bad that there isn’t consistency in the spec on how to deal with <mda:metadata> as a module or as an extension overall, not just in <gls:glossary>.

From: Dr. David Filip [mailto:David.Filip@ul.ie]
Sent: Tuesday, May 19, 2015 1:55 PM
To: Ryan King
Cc: Dr. David Filip; Yves Savourel; XLIFF Main List
Subject: Re: [xliff] Extensibility in Modules

Ryan, I don't want to be dogmatic here

But if you want to transfer terminology in your package, you should use the gls as the baseline to promote interoperability. Other people will be able to use at least the simple goodness without building support specifically for your terminology in mda.

If anytime you think are more feature rich than what is in the module, you are going to replace it with a proprietary mda construct, we are going again down the XLIFF 1.2 extensibility route

mda although a module is just an extension mechanism that does not allow for specification of any business logic

No one can really consume your mda based terminology as terminology

gls is tied to the core annotation mechanism and can be used for term lifecycle mgmt while terminology is in the XLIFF facilitated roundtrip

Outside of the XLIFF roundtrip terminology is supposed to live in TBs i.e. outside of XLIFF scope..

I think that the gls has a basic terminology roundtrip logic (tied to core) that gives you a terminology specifc baseline that the mda can never express.

All your structures that don't fit the gls you can transfer from your core extension point to the gls extension points.

If you feel that the data isn't safe there, you can suggest explicitly listing mda at those gls extension points.. I think this would be a minor change that would cause no harm.. would work for any XLIFF 2, and from 2.1 it would be treated as a guaranteed module.

Cheers

Dr. David Filip

=======================

OASIS XLIFF TC Secretary, Editor, and Liaison Officer

LRC | CNGL | 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, May 19, 2015 at 9:24 PM, Ryan King <ryanki@microsoft.com> wrote:

>> Anyways, using mda to completely replace gls violates the high level PR saying that user defined extensibility, no matter if mda or custom namespace based, must not be used instead of a core or module feature.

David, I might argue that with you. The spec states that the <gls:glossary> module is for used for “*Simple* glossaries, consisting of a list of terms with a definition or translation”, my terminology needs are not simple and extend beyond a list of terms with definition and translation.

From: xliff@lists.oasis-open.org [mailto:xliff@lists.oasis-open.org] On Behalf Of Dr. David Filip
Sent: Tuesday, May 19, 2015 12:53 PM
To: Yves Savourel
Cc: XLIFF Main List
Subject: Re: [xliff] Extensibility in Modules

Thanks, Yves, Ryan

On Tue, May 19, 2015 at 8:24 PM, Yves Savourel <ysavourel@enlaso.com> wrote:

The only caveat I can think of is that--except for the PR that protects XLIFF-defined elements--you have to expect its support to be as for an extension rather than a module.

Exactly, that's what I also said in the meeting ;-) [but we had a bad feedback between myself and Ryan, so it might have broken up]

I guess, Ryan, and others might benefit from allowing the mda explicitly in Glossary.

On the other hand, if Ryan and MSFT are just looking into a way to include metadata in the glossary module w/o defining their own namespace for it, the mda is as good as any other namespace.

The question is I guess, if MSFT expect a third party to process and roundtrip the mda based metadata that they put in the glossary module.

It's probably worth looking at the detailed use case to tell if the mda is in need of module handling rather than extension handling at the Glossary extension points..

Anyways, using mda to completely replace gls violates the high level PR saying that user defined extensibility, no matter if mda or custom namespace based, must not be used instead of a core or module feature.

Cheers

dF

Dr. David Filip

=======================

OASIS XLIFF TC Secretary, Editor, and Liaison Officer

LRC | CNGL | 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

xliff message