[OASIS Issue Tracker] (XLIFF-28) Comments in the draft

[OASIS Issues Tracker <workgroup_mailer@lists.oasis-open.org>]

    [ https://issues.oasis-open.org/browse/XLIFF-28?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=65432#comment-65432 ] 

Yves Savourel commented on XLIFF-28:
------------------------------------

> we have several other Constraints in the spec 
> that are referencing annotatorsRef values

They refer to the attribute annotatorsRef, not to its value (we refer to the "annotatorsRef value" only once in a note in section 5.9.1).
I don't see why referring to the attribute would justify duplicating/editing the description of the value instead to just point to the official description.

Again pointing to the example of xml:lang: there are plenty of references to xml:lang in the spec, but we don't re-write its description. Why its:annotatorsRef would be different?

There are always dangers to duplicating a definition, especially a relatively complex one like annotatorsRef: We may make a typo or an error in the XLIFF document, or the ITS spec may get an errata, etc.
Even worst: In the current spec we actually don't use exactly the same description as in the ITS spec: We introduce a notion of "triple" that doesn't exist in the ITS definition (and may lead developers to get confused with RDF triples). At the very least if the TC decide it's OK to duplicate the definition then we must use the exact same wording as the original.

The only advantage of re-writing the definition in the XLIFF spec is that the reader does not have to click on a link. I understand the positive intend to try to make the reading immediate, but I also think it is outweighed (by far) by the simplicity and the safety of just pointing to the existing information.

> Comments in the draft
> ---------------------
>
>                 Key: XLIFF-28
>                 URL: https://issues.oasis-open.org/browse/XLIFF-28
>             Project: OASIS XML Localisation Interchange File Format (XLIFF) TC
>          Issue Type: Bug
>          Components: ITS Module
>    Affects Versions: 2.1_csprd02
>         Environment: http://markmail.org/thread/jzdqfm5c6hgofjbu
>            Reporter: Yves Savourel
>            Assignee: David Filip
>            Priority: Minor
>              Labels: Proposed, editorial, work_required
>             Fix For: 2.1_csprd03
>
>
> There are quite a few (more than a dozen) embedded comments in this latest
> draft:
> e.g: "COMMENT: MAKE CLEAR THAT THERE ARE TWO HANDLINGS, ONE IS NOT TO REPRESENT
> THE DATA CATEGORY, THE OTHER IS IN THE MODULE. THE
> EXTRACTION IS FOR ITS PROCESSORS GENERATING XLIFF, THE DIRECT REPRESENTATION IS
> FOR XLIFF PROCESSORS."
> Some, like the one above, seem to indicate that there are still relatively
> important parts of the text that needs to be
> edited/added/etc.
> I assume this means there will be a csprd03 regardless of the disposition of the
> comments for csprd02. Is that correct?



--
This message was sent by Atlassian JIRA
(v6.2.2#6258)