Re: issue-51 too many global rules

Hi Dave, all,

2012/10/23 Dave Lewis <dave.lewis@cs.tcd.ie>

>  On 23/10/2012 14:52, Felix Sasaki wrote:
>
> I know - my point is not about pointers, but about "adding information
> *without fixed values* to attributes or elements". I just can't imagine
> people writing rules like this
>
>  <its:locQualityIssueRule selector="//span[@id='q1']"
> locQualityIssueType="typographical" locQualityIssueComent="Sentence without
> capitalization" locQualityIssueSeverity="50"/>
>
>
>  That is, tailored to one "span" element. Am I wrong?
>
>
> Hi Felix,
> I think you are right on the point that authoring global rules doesn't
> scale for individual spans. You make the same point well in relation to
> provenance in:
>
> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Oct/0093.html
>
> That was in response to my suggestion that it may be 'convenient' to write
> global rules in certain circumstances. To give a more concrete example,
> look at the source view on a microsoft support page, e.g.:
>
> http://office.microsoft.com/en-us/word-help/crop-a-picture-or-place-it-in-a-shape-HA102749423.aspx?CTT=1
>
> There you would see fragments like:
> <li>Click <b class="ui">Picture Tools</b> &gt; <b class="ui">Format</b>&gt; <b
> class="ui">Size</b> and click the arrow under <b class="ui">Crop</b>.</li>
> here's where a class attribute identifies a type of content likely to be
> treated differently during localization. So having a global rule with a
> selector="//b[@class='ui'] would be more convenient for overriding, e.g.
> the provenance of translation by software translators for these nodes over
> provenance recording documentation translators for the rest of the page.
>
> It would be interesting to hear from out Microsoft friends on how they do
> this in practice. Presumably this is generated from some source content and
> they may have tooling that can easily manage the insertion of local mark-up
> - but for other trying to address the same problem more manually, one can
> imagine that global rules, especially related to class attributes or <code>
> elements, would indeed be a convenience.
>
> It would be good to hear from some of the industrial implementers on
> whether this really is a convenience or not for them. If not, i would not
> strongly feel the need to retain global rules purely for the convenience of
> avoiding local mark-up in these cases.
>

Great, I agree. So let's see if we get feedback by the TPAC meeting, and if
not, let's drop the global rules in question.


>
> However, I think Yves' point that we would loose the ability to mark-up
> attributes without global rules is a much stronger argument for retaining
> them - that feels like a real loss of expressiveness that is already
> established in ITS1.0.
>


If nobody uses the expressiveness, we don't need to add it to new data
categories in ITS 2.0. I still get nightmares from rubyPointer .... :)
In ITS 1.0 the expressiveness was mostly used on a per format basis, e.g.
saying "all 'alt' attributes at HTML 'img' should be translated. I don't
see the "per document format" or even "per template" use case for

QualityIssue, Quality Precis, Disambiguation, mtConfidence, text analysis
annotation, translation provenance.

So for these the "pointer attributes" (or even reference pointer only)
might be sufficient.

Best,

Felix


>
> cheers,
> Dave
>
>
>
>
>
>
>
>


-- 
Felix Sasaki
DFKI / W3C Fellow