See also: IRC log
felix: Carl & Frederick gave
a presentation in XML Prague
... topic of the presentation - building more connectors
... connectors between ITS 2.0 supported systems (from XML
editor, e.g., to OKAPI, etc.)
... Good progress in implementation work
... If there are resources left in the second half, maybe there
could be implementations on connectors between tools
>> second half -> second half of 2013
felix: by Feb 15th people need to update Wiki
felix: There was no clear outcome
of the discussion (in previous call)
... Is there something more needed?
Dave: Last Mondays call, it came
down to change wording in the specification - the general issue
where ITS applies in HTML where it is XML
... We discussed in the call that there should be a note
... Or a best practice
... The wording refers to when you work with XHTML in Web
... There was no solution if we would use XHTML as an internal
format
Felix: talked to Jirka, he said that easy guidance could be that if it is XHTML you handle it as XML, but if you need to publish it, it could be changed to "its-"
Dave: Could be good clarification
Felix: Jirka not here, so we cannot create the note
<fsasaki> ACTION: jirka to draft an explanatory note on issue-110 - where ITS applies in HTML where it is XML, including a workflow on the web and outside the web [recorded in http://www.w3.org/2013/02/11-mlw-lt-minutes.html#action01]
<trackbot> Created ACTION-444 - Draft an explanatory note on issue-110 - where ITS applies in HTML where it is XML, including a workflow on the web and outside the web [on Jirka Kosek - due 2013-02-18].
Felix: Could be useful to Linguaserve
<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt-comments/2013Jan/0015.html
Felix: waiting for Christian to
get back on us
... Later in the call, term+disamb
... Could help resolve the Nif comment
Dave: It has come up with the
Indic languages+disambig.
... Also possibly Loc.note, etc.
<fsasaki> https://www.w3.org/International/multilingualweb/lt/track/issues/109
Dave: Got a response on NIF (in e-mail
Felix: Is there someone writing / do we have the manpower
<fsasaki> ACTION: felix to write a note about NIF role for standoff scenarios (might include issue-68 and issue-109 [recorded in http://www.w3.org/2013/02/11-mlw-lt-minutes.html#action02]
<trackbot> Created ACTION-445 - Write a note about NIF role for standoff scenarios (might include issue-68 and issue-109 [on Felix Sasaki - due 2013-02-18].
Felix: Writing himself an action on Issue 68 and 109
Dave: The note first and then we
will see
... Working with Leroy on best practice where NIF predicates
could be used
... but there are no examples yet
Felix: wants to do dreaming
around
... dreams about standardising NIF through W3C
... Could it be a topic to discuss?
Dave: The onto-lexica group maybe
is interested
... maybe also in CNGL
... It could fit well - could be interesting (also to
DCU)
... There hasn't been much activity on the onto-lexica
group
... But we could put the idea on the table
Could someone turn the music off
<fsasaki> done :)
Dave: NIF plays a big role in [some events]
<fsasaki> = european data forum etc.
<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Feb/0049.html
Shon: Minor issue, we have no
spec. that states in the loc note element for space
normalisation.
... not clear whether it should be normalised
... seems clear that everyone does it
... Should we get into details how it should be normalised?
<fsasaki> this is now https://www.w3.org/International/multilingualweb/lt/track/issues/116
Shon: Difficult to specify ...
Felix: Test Suite output won't be
influenced
... test suite removes empty spaces
... not perfect, but catches most errors
... In output we have a description where there is space/tab
character
Shon: I'm working with test files, othis is how the issue got apparent
(not sure I got this right - there is someone breathing really loud - I barely hear Shon)
Shon: But not sure whether this is really a big issue
Shon/Felix: suggest to leave as is
Felix: We come back next week and close it next week
Felix: By Shaun
<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Feb/0048.html
Shaun: tried to give an example
for a selector that would not work if you put it in XSLT
... Example is not a valid path as it is missing a leading
slash
... example in ITS spec...
<fsasaki> now https://www.w3.org/International/multilingualweb/lt/track/issues/117
Shaun: Seems to be ok if you cannot handle absolute paths then it is not conforming to the spec.
Felix: ITS 2.0 there was a change in late december - it is just a location path and not an abs. loc. path
<fsasaki> http://www.w3.org/TR/2012/WD-its20-20121206/#d0e2063
Felix: It is also possible to use relative paths
Shaun: Language should be cleaned up in the example
Felix: Pandora box 6.1?
<fsasaki> section 1-2
Felix: Not 6.1, but 1-2
... These are not normative, but have a lot of issues
... We do not need to resolve this for the last call, but the
sections will have to be updated
... Proposes Shaun to rewrite the example
Shaun: Yes
<fsasaki> ACTION: shaun to re-write example mentioned in issue-117 [recorded in http://www.w3.org/2013/02/11-mlw-lt-minutes.html#action03]
<trackbot> Created ACTION-446 - Re-write example mentioned in issue-117 [on Shaun McCance - due 2013-02-18].
<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Feb/0028.html
Felix: Wiki changes, wants for
people to be aware
... Thx. to people who already made changes
... Will make changes on Friday
<fsasaki> https://github.com/finnle/ITS-2.0-Testsuite/tree/master/its2.0/xliffsamples
Felix: XLIFF samples, various
people+Marcis asked
... This is not part of conformance testing
... People need this, but this is not for the stand.
process
Dave: finding it useful
... useful for use-cases for Rome
... Giving us means to figure out the XLIFF mapping
<fsasaki> yves' mail on the mapping at http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Feb/0046.html
Dave: Using for testing of
mapping for provenance
... Will put more examples later this week
... Had a talk with Ankit/Phil to see whether XLIFF mapping for
[..] could be included as well
<fsasaki> Marcis: thanks a lot, we were looking for such examples
<fsasaki> .. we are looking forward to see more, covering other data categories
<fsasaki> .. looked at the biiiig XLIFF file
<fsasaki> .. we are looking for particularily elements that cover data categories that we will implement
<fsasaki> .. terminology, elements within text, ...
<fsasaki> .. haven't had the change whether the examples cover everthing
<fsasaki> felix: do you need several data categories at the same time in one file?
<fsasaki> marcis: no, even if we have one data category per file that is enough
<fsasaki> dave: we are also looking into how to introduce tan and terminology into a use case
<fsasaki> .. we have taken content out of xliff, run it through enrycher and what the output would be
<fsasaki> .. not sure what the values would be, but we start with suggestions
<fsasaki> marcis: the more test files we have the better
<fsasaki> dave: esp. real life translated
<fsasaki> ACTION: yves to make a batch transformation of the test suite to xliff [recorded in http://www.w3.org/2013/02/11-mlw-lt-minutes.html#action04]
<trackbot> Created ACTION-447 - Make a batch transformation of the test suite to xliff [on Yves Savourel - due 2013-02-18].
<fsasaki> covered by action to jirka, see above
Felix: already gave action item to Jurka
Felix: As said before call, me,
Tadej, Marcis, and Christian took this off the call
... Outcome: some issues are resolved
... Differences came clear
... confidence of term relates to term=yes/no, but not the
ref.
... For Disambig. it relates to the complete annotation
... Maybe after Wednesday we will come up with ways how to make
differences clearer
... does not want this to be secret discussion
... Offline calls meant just to move discussion forward
... anyone can join
Dave: to clarify - the value for terminology confidence relates to term=yes/no?
Felix: Yes
... this is why here merging does not fit
... A guess for the outcome - we probably won't merge, but will
have progress on describing relations and giving guidance on
usefulness
<fsasaki> tadej: one solution would be to have two different confidence values, one for term flag, one for identifier
<fsasaki> .. but there is a lot of orthnogality here, no clean way to merge things
<fsasaki> ..we also decided to call this "text analyis", it has a "ta" prefix instead of a "disambig" prefix
<fsasaki> marcis: nothing to add
Felix: Some new small issues, the larger on term+disambig
<fsasaki> https://www.w3.org/International/multilingualweb/lt/track/actions/open
Felix: Have we forgotten
anything?
... knows person who has the most actions - he is called
Felix
I hear only noise
Dave: said last week that there
were people for shephearding
... If they could send out a message to commenters leaving one
week to reply whether answers are satisfactory?!
Felix: Agrees
Felix: Rome workshop ahead, call
next week, Feb. 20
... Asked implementers to prepare poster drafts
... Will be public
Felix: Any other questions?
... thanks a lot
... good that we have run out of discussion topics
... We could have also shorter calls in the future