Re: [ISSUE-34] Problem with mandatory attributes for quality

I would see a requirement to support multiple annotations per element but 
only independent ones. That is, Arle's second example would seem required 
as it allows multiple, independent agents to add markup. For the first 
example, the parent entity seems redundant.

Felix, what would prohibit the application of CSS/DOM processing to the 
multiple records?

Phil.





From:   Felix Sasaki <fsasaki@w3.org>
To:     Arle Lommel <arle.lommel@dfki.de>, 
Cc:     Multilingual Web LT Public List <public-multilingualweb-lt@w3.org>
Date:   07/08/2012 15:05
Subject:        Re: [ISSUE-34] Problem with mandatory attributes for 
quality



Hi Arle, all,

as Yves has pointed out in a separate mail, inline markup has its limits. 
But is there a need to have "multiple quality records" inline? No matter 
how you represent the multiple records, you won't be able to apply the CSS 
styling to it anymore, or any other dom based processing in the browser. 

So I would propose to leave the inline proposal as is, with Yves' 
suggestions at
http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Aug/0107.html

and leave further discussions to the standoff / reference scenario. The 
main point is that we align the metadata available inline and offline, but 
having the multiple quality records just offline sounds reasonable to me.

(Btw., a conversion to NIF might be another solution to the offline 
scenario, but I won't go to far here ...)

Best,

Felix

2012/8/7 Arle Lommel <arle.lommel@dfki.de>
Going further, I am not sure what it would mean, however, if a 
locQualityScore value were declared inside the context of another one. 
Under out normal precedence rules it would be as if the "outer" one did 
not exist, but it could be that someone it expressing partial scores 
(e.g., the overall score is 85, but this <div> has a score of 98 and that 
<div> is a 42. In this case it is not  like translate where the meaning is 
clear: a translate="yes" bit nested in a translate="no" bit completely 
overrides the translation="no" intent. But here the semantics are not so 
clear. Maybe it would solve it to state that the attribute does NOT apply 
to daughter elements. It's clear that semantically it does, but 
syntactically, we cannot see this as normal inheritance if the value from 
the whole cannot be seen as applying to the parts

And I see I contradicted myself. First I said that locQualityScore doesn't 
make sense going down to daughter elements, which is true in general, but 
at the same time it isn't true that if I have some structure like this:

<div id="1" its-loc-quality-score="80">
<div id="2" its-loc-quality-score="60">Some bad <span id="A" 
its-loc-quality-type="misspelling" 
its-loc-quality-code="ABC:SPELLING">kontint</span></div>
<div id="3" its-loc-quality-score="100">Some good content</div>
</div>

that there is no relation between the daughters and the parent.

Rather, there is a contributory (i.e., bottom-up) relationship at work 
here, not one of normal inheritance: the 80 applies to the whole, not to 
the constituent parts. The values for divs 2 and 3 contribute to the value 
of div 1.

And going down further, the same applies, the value of 60 applies only to 
the div as a whole, not to any part of it. If anything "Some bad" would 
have a score of 100 and "kontint" might have a value of 0.

On the other hand, it is not as if the value associated with divs 2 and 3 
completely replace the value of div 1. If they did there would be no text 
to which the value from div 1 applies, which clearly is not the case: it 
applies to all of the text in div 1, taken as an intact whole.

The more I look at the score, the tougher it gets to unify with the other 
span-ish (as opposed to Spanish) attributes used to identify specific 
errors.

LocQualityProfile is maybe a bit easier to unify with them, but what I'd 
really like to see is something like this:

<html lang="en">
   <head>
      <meta charset="utf-8" />
      <meta its-loc-quality-profile="XYZ:xyz.com/fake" />
      <meta its-loc-quality-profile="PQR:pqr.com/alsofake" />
      <title>Some junk in a document</title>
   </head>
   <body>
      <p class="segment" id="s0001"><span
          its-loc-quality-type="inconsistent-entities"
          its-loc-quality-code="XYZ:wrong_name">Christian Bale</span>
      (<span its-loc-quality-type="numbers"
          its-loc-quality-code="PQR:numbers_do_not_match"
          its-loc-quality-comment="Should be 1867">2067</span>–1934) 
conceived of an
      instrument that could transmit its sound from a power plant for 
hundreds of miles
      to listeners over telegraph wiring.</p>
   </body>
</html>

Here the two quality profiles exist side by side and the prefix in 
loc-quality-code selects which one applies to which piece of content.

So the question is if we can do this using the ITS inheritance model and 
scoping model or not. If we cannot than we run into having to redeclare 
the profile for each and every place we want to use the category.

-Arle



-- 
Felix Sasaki
DFKI / W3C Fellow



************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the sender immediately by e-mail.

www.vistatec.com
************************************************************

Received on Tuesday, 7 August 2012 14:15:46 UTC