W3C home > Mailing lists > Public > public-multilingualweb-lt@w3.org > November 2012

RE: [All] editing call minutes and two questions

From: Mārcis Pinnis <marcis.pinnis@Tilde.lv>
Date: Fri, 30 Nov 2012 13:57:31 +0200
To: Felix Sasaki <fsasaki@w3.org>, "public-multilingualweb-lt@w3.org" <public-multilingualweb-lt@w3.org>, "tadej.stajner@ijs.si" <tadej.stajner@ijs.si>
Message-ID: <AC6FD4BB9BB02540AC7322091A6C3B5472B025B000@postal.Tilde.lv>
Hi Felix,

If I understand correctly the question is whether we need to add global rules for Term Confidence?


1)      It depends on whether there is a way to mark terms in the scenarios that require a global MT Confidence? For instance, can we add global rules referring to subparts of texts where tags or attributes are not allowed? MT confidence would indicate to the whole tag or attribute, but not the subparts. If we cannot annotate the subparts as terms, then Term Confidence in global rules would not make sense as such fragments would seldom be terms anyway (they would probably be larger expressions).


2)      On the other hand, Terminology may be defined also globally (example 40), but it refers to whole tags (or attributes ...). For such cases it would be reasonable to add a global rule that would identify a single tag or attribute in the case if a local attribute is not possible. However, such an implementation would require identifying each separate element or attribute with its own xPath and that would cause complexity in editing a document and I would not see it useful (the troubles are from a development point of view not worth it... – having a dictionary apart from the document then would make much more sense, but that is not what ITS is about...)

Personally, I do not like the idea of having a global indicator to precise elements/attributes at all (I think that such structures cause huge problems when data needs to be updated), but if there is already a mechanism for MTConfidence, then someone has already gone through the trouble?! And from that point of view, it may be worth considering also such an option for other confidence scores. Also from a functional perspective (whether we can annotate anything anywhere with the help of ITS 2.0) we would need to add it, I suppose.

Thoughts?

Best regards,
Mārcis ;o)

From: Felix Sasaki [mailto:fsasaki@w3.org]
Sent: Thursday, November 29, 2012 6:28 PM
To: public-multilingualweb-lt@w3.org; Mārcis Pinnis; tadej.stajner@ijs.si
Subject: [All] editing call minutes and two questions

Hi all, esp. Mārcis and Tadej (since there are questions for you below),

see the minutes from today's editing call here

http://www.w3.org/2012/11/29-mlw-lt-minutes.html


and below. Two important questions

1) Esp. for Mārcis: can you have a look at
http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Nov/0254.html

question no 2), and tell us what you think?

2) Esp. for Tadej: you said that in the global section of disambiguation we should add disambigSource
http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Nov/0234.html

but we realized that we don't know how. In the global section there is no mention of "mode 1" and "mode ".

Can you re-write the definition of global in disambiguation

[

·         A required selector attribute that contains an absolute selector<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors> which selects the nodes to which this rule applies.

·         An optional disambigGranularity attribute that contains a string, specifying the granularity level of the disambiguation. The value MUST<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#rfc2119> be one of the following identifiers: "lexicalConcept", "ontologyConcept", or "entity". The default value is "entity".

·         At least one of the following:

o    To specify the target type class, exactly one of the following:

§  A disambigClassPointer attribute that contains a relative selector<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors> pointing to a node specifying the type of entity or concept class behind the selector.

§  A disambigClassRefPointer attribute that contains a relative selector<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors> pointing to a node that holds an IRI that specifies the type of entity or concept class behind the selector.

o    To specify the target identity, exactly one of the following:

§  A disambigIdentPointer attribute that contains a relative selector<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors> pointing to a node that represents a unique identifier for the disambiguation target.

§  A disambigIdentRefPointer attribute that contains a relative selector<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors> pointing to a node that holds an IRI that represents a unique identifier for the disambiguation target.

]

so that the two modes and disambigSource are in here?

Having that by Friday end of business day would be great. We then would be finished with the normative parts of the draft.

Thanks,

Felix


   [1]W3C



      [1] http://www.w3.org/




                               - DRAFT -



                          MLW-LT editing call



29 Nov 2012



   See also: [2]IRC log



      [2] http://www.w3.org/2012/11/29-mlw-lt-irc




Attendees



   Present

          dave, felix, jirka, tadej, karl, Yves_



   Regrets

   Chair

          felix



   Scribe

          fsasaki



Contents



     * [3]Topics

         1. [4]questions from yesterday

         2. [5]8.13 External Resource

         3. [6]8.14 Target Pointer

         4. [7]8.15 Id Value

         5. [8]8.16 Preserve Space

         6. [9]8.17 Localization Quality Issue

         7. [10]8.18 Localization Quality Rating

         8. [11]validation of example files

         9. [12]8.19 MT Confidence

        10. [13]back to rating data category

        11. [14]allowed characters + storage size (again)

        12. [15]going back to rating value

        13. [16]spec editing topics

        14. [17]disambigSourcePointer

     * [18]Summary of Action Items

     __________________________________________________________



   <scribe> scribe: fsasaki



questions from yesterday



   dave sent out mail to confirm confidence score for terminology,

   as discussed yesterday



   <Yves_> FWIW: we'll implement confidence it in Okapi in needed



   now on action-339



   provRef as single URI or list of URI?



   "Reference to external provenance information"



   "AN IRI" > "A space separated list of IRIs"



   <Yves_> then the examples need to be changed too



   <Yves_> I reduced them all to a single IRI



   "A space (U+0020) separated list of IRIs"



   Yves_, currently changing just one example -

   EX-provenance-local-1.xml, to have a space separated list



8.13 External Resource



   nothing on external resource, it seems



8.14 Target Pointer



   nothing on target pointer, it seems



8.15 Id Value



   mail from fredrik

   [19]http://lists.w3.org/Archives/Public/public-multilingualweb-


   lt/2012Nov/0250.html



     [19] http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Nov/0250.html




   <Yves_> +1 for Jirka's comment



   changing the id value section to say: "If the attribute xml:id

   is present or id in HTML for the selected node, the value of

   the xml:id attribute or id in HTML MUST take precedence over

   the idValue value."



   that's everyting on id value section, it seems



   <Yves_> One note for later (Storage Size section): "Dominican

   Replubic" should Dominican Republic" in example

   EX-storageSize-html5-local-1.html



   Applying the id value data category to xml:id (in XML) or id

   (in HTML) attributes in global rules is not necessary since

   those attributes are ...



   Yves_, thanks, we will take that up later, now 8.16 Preserve

   Space



8.16 Preserve Space



   added a link from "xml:space" to

   [20]http://www.w3.org/TR/2008/REC-xml-20081126/#sec-white-space




     [20] http://www.w3.org/TR/2008/REC-xml-20081126/#sec-white-space




8.17 Localization Quality Issue



   "locQualityIssuesRefPointer, do not apply to HTML" >

   locQualityIssuesRefPointer does not apply to HTML



   "ITS 2.0-compliant tools that use these categories" > "ITS

   2.0-compliant tools that use these types"



   "their internal categories" > "their internal types"



   replacing other occurences of "categories" with "types"



8.18 Localization Quality Rating



   felix: looks fine



   dave: changed after yesterday's discussion



   <Yves_> I have to leave for the office now. I may be back

   online in 20/30 mn (depending on how many coyotes I have to

   fend off on the bike path)



   thanks, bye, Yves_



validation of example files



   jirka: did add a target for validation

   ... need someone to fix the errors



   <scribe> ACTION: felix to check examples found by relax ng

   validation [recorded in

   [21]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action01]



   <trackbot> Created ACTION-340 - Check examples found by relax

   ng validation [on Felix Sasaki - due 2012-12-06].



   <scribe> ACTION: felix to check syntax highlighting of html

   examples with focus on quotes - due 31 december [recorded in

   [22]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action02]



   <trackbot> Created ACTION-341 - check syntax highlighting of

   html examples with focus on quotes [on Felix Sasaki - due

   2012-12-31].



   jirka: now have created xml validation, after last call

   publication will work on html validation



8.19 MT Confidence



   going through the three sections 8.19 MT Confidence, 8.20

   Allowed Characters, 8.21 Storage size



   dave: for the example EX-mtConfidence-global-html5-1.html would

   be good to have an example with two different attributes with

   text content, in the same element

   ... e.g. title and alt in an image



   <scribe> ACTION: daveL to create mt confidence score example as

   described in

   [23]http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40 - due 31

   december [recorded in

   [24]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action03]



     [23] http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40




   <trackbot> Created ACTION-342 - create mt confidence score

   example as described in

   [25]http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40 on David

   Lewis - due 2012-12-31].



     [25] http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40




back to rating data category



   jirka: in spec the locQualityRatingScore value is between 0 and

   100

   ... yestedays call we said between 0 and 1

   ... which one is true?



   <scribe> ACTION: felix to check with yves whether

   locQualityRatingScore should be between 0 and 1 or 0 and 100

   [recorded in

   [26]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action04]



   <trackbot> Created ACTION-343 - Check with yves whether

   locQualityRatingScore should be between 0 and 1 or 0 and 100

   [on Felix Sasaki - due 2012-12-06].



   action-343: also relevant for locQualityRatingScoreThreshold



   <trackbot> ACTION-343 Check with yves whether

   locQualityRatingScore should be between 0 and 1 or 0 and 100

   notes added



allowed characters + storage size (again)



   allowed characters looks fine



   "nel for NEXT LINE (U+0085)" - need to check code for nel



   seems to be correct



   went through all data categories, only topic left is

   locQualityRatingScore value, see above



going back to rating value



   yves: phil mentioned that des wanted something between 0-100

   ... from my point of view it doesn't matter

   ... that is, the spec would be right, I didn't change it



spec editing topics



   yve: val and quote have strange behavior



   "A disambigIdentPointer attribute that contains a relative

   selector pointing to a node that represents a unique identifier

   for the disambiguation target." plus disambigSourcePointer



   <scribe> ACTION: felix to write mail today to tadej to confirm

   position of disambigSourcePointer in global rules [recorded in

   [27]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action05]



   <trackbot> Created ACTION-344 - Write mail today to tadej to

   confirm position of disambigSourcePointer in global rules [on

   Felix Sasaki - due 2012-12-06].



disambigSourcePointer



   <fsasaki_> need to check with tadej about this



Summary of Action Items



   [NEW] ACTION: daveL to create mt confidence score example as

   described in

   [28]http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40 - due 31

   december [recorded in

   [29]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action03]

   [NEW] ACTION: felix to check examples found by relax ng

   validation [recorded in

   [30]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action01]

   [NEW] ACTION: felix to check syntax highlighting of html

   examples with focus on quotes - due 31 december [recorded in

   [31]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action02]

   [NEW] ACTION: felix to check with yves whether

   locQualityRatingScore should be between 0 and 1 or 0 and 100

   [recorded in

   [32]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action04]

   [NEW] ACTION: felix to write mail today to tadej to confirm

   position of disambigSourcePointer in global rules [recorded in

   [33]http://www.w3.org/2012/11/29-mlw-lt-minutes.html#action05]



     [28] http://www.w3.org/2012/11/29-mlw-lt-irc#T14-50-40




   [End of minutes]

     __________________________________________________________





    Minutes formatted by David Booth's [34]scribe.perl version

    1.137 ([35]CVS log)

    $Date: 2012-11-29 15:31:06 $

     __________________________________________________________



     [34] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm


     [35] http://dev.w3.org/cvsweb/2002/scribe/




Scribe.perl diagnostic output



   [Delete this section before finalizing the minutes.]

This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01

Check for newer version at [36]http://dev.w3.org/cvsweb/~checkout~/2002/


scribe/



     [36] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/




Guessing input format: RRSAgent_Text_Format (score 1.00)



Succeeded: s/0-1/0-100/

Found Scribe: fsasaki

Inferring ScribeNick: fsasaki

Present: dave felix jirka tadej karl Yves_

Got date from IRC log name: 29 Nov 2012

Guessing minutes URL: [37]http://www.w3.org/2012/11/29-mlw-lt-minutes.ht


ml

People with action items: davel felix



     [37] http://www.w3.org/2012/11/29-mlw-lt-minutes.html






   End of [38]scribe.perl diagnostic output]



     [38] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm




Received on Friday, 30 November 2012 11:58:10 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 9 June 2013 00:25:03 UTC