Re: [ISSUE-22] Provenance and Agents

Comment below on 'items'

On 25/10/2012 23:36, Felix Sasaki wrote:
> Hi Dave, all,
>
> 2012/10/26 Dave Lewis <dave.lewis@cs.tcd.ie <mailto:dave.lewis@cs.tcd.ie>>
>
>     Hi Yves,
>     I don't see immediately why not. 'Items' is nice and neutral
>
>     But would it not be easier to retain the item element name from
>     the different data categories? e.g.
>
>     <its:items id="1">
>          <its:locQualityIssue locQualityIssueType="misspelling"
>                 locQualityIssueComment="'c'es' is unknown. Could be
>     'c'est'"
>                 locQualityIssueSeverity="50"/>
>          <its:translationProvenanceRecord
>
>                
>     transToolRef="http://www.onlinemtex.com/2012/7/25/wsdl/"
>     <http://www.onlinemtex.com/2012/7/25/wsdl/>
>                 transOrg="acme-CAT-v2.3"
>                 transRevToolRef="http://www.mycat.com/v1.0/download"
>     <http://www.mycat.com/v1.0/download>
>                 transRevOrg="acme-CAT-v2.3"
>     provRef="http://www.examplelsp.com/excontent987/productio/prov/e6354"
>     <http://www.examplelsp.com/excontent987/productio/prov/e6354>/>
>     </its:items>
>
>
>
> Yves wrote at
> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Oct/0296.html
> "We said we should have a single root element in each script. So you 
> would just have several scripts if you have several stand-off references."
> To resolve my concern
> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Oct/0294.html
> See "Could this create a conflict".
>
> What you do above does create the conflict, in terms of
>
> - conformance (does an implementation just implementing locquality 
> issue need to pass all information, 
> including its:translationProvenanceRecord, or throw it away?), or in 
> terms of
>
> - precedence ("I am implementing locquality issue globally, 
> and translationProvenance locally - what do I do if there is in one 
> script element both information?).
>
> So I think we need to keep separate items elements.
OK, so to confirm, do we go with:

"An element|items|(or|<span items>|in HTML) which contains:

  *

    One or more elementsitem(or|<span its-item>|in HTML), each of which
    contains:

      o

        At least one of the following attributes:

          +

            A|locQualityIssueType|attribute that implements thetype
            information
            <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#lqissueDefs>.

          +

            A|locQualityIssueComment|attribute that implements
            thecomment information
            <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#lqissueDefs>.

      o

        An optional|locQualityIssueSeverity|attribute that implements
        theseverity information
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#lqissueDefs>.

      o

        An optional|locQualityIssueProfileRef|attribute that implements
        theprofile reference information
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#lqissueDefs>.

      o

        An optional|locQualityIssueEnabled|attribute that implements
        theenabled information
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#lqissueDefs>."

    Do you think we need then an explicit statement ruling out mixing
    data categories in the same item?

    Also, I'm not sure if using its:locQualityIssue or
    its:translationProvenanceRecord instead of its:item would help
    address Yves' point
    in:http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Oct/0296.html


"The only drawback (if you can call that one) is that you would not know 
in a single operation what data category the <its:items/its:records> 
element holds if you are looking at it without looking from the content 
referencing them. But that's probably not a big problem"?

I'm asking primarily to nail this down for the provenance data category 
also.
cheers,
Dave