See also: IRC log
<fsasaki> checking attendance ...
<fsasaki> agenda here http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Dec/0075.html
felix: today was deadline for test milestone M2, to have each partners attempted at least one of each type of feature test they are planning
felix: thanks for everyone on working to progress in advancing to the test suite
dom: show screen with progress
... reviews status, noting there are some short delays which have been agreed
... des and tidle have delays and need a revised M2 date to be agreed with co-chairs
... ruby being done in TCD, seems OK
... milan has delay, tadej has a delay
... need to discuss local filter with DCU
... prov test is still shakey, leroy is checking
... logrus with locqualissue need to be resolved
... mt confidence OK
... cocomore need to complete allowed char
... overall happy that progress is good, but need to communicate with outstanding tests tomorrow.
Felix; agree, keep cochair posted
Dom: highlighted that the complete test deadline in end Jan, so really not far off
felix: we should allocate plenty of time at prague meeting for test suite
dom: also note at march EU review we should have 80% conformance in actual implementations
felix: as yves pointed out, should have some way of comparing results on web site
<pnietoca> sorry my microphone is not not working again, I'll try to use Mauricio's
dom: for M2 and M3 we need people to perform their own check, but we will need to put up all the outputs and do authoritative, verifiable test
felix: online test would be really good
leroy; we can provide online test
dom: so we will deliver in new year
<scribe> ACTION: leroy to provide online diff service so posted output can be checked [recorded in http://www.w3.org/2012/12/17-mlw-lt-minutes.html#action01]
<trackbot> Created ACTION-373 - Provide online diff service so posted output can be checked [on Leroy Finn - due 2012-12-24].
Dom: Dom and Leroy will be unavailable for two weeks over chirstmas
<DomJones1> Dom and Leroy away from 22nd December to the 7th of January
yves: idea is that final test page was generated automatically for ITS1.0, so that would be good for 2.0
yves: i will look at this over holidays
dom: does this just diff the input and output?
felix; yes it just compared files and provide a green flag if it works
scribe: so this could be done with the diff service mentioned earlier as a batch over all current test input files and submitted output files
Dom; we will look at this as part of last action itme
yves: during testing looked at
wider coverage of domain test, with different
... should these be different tests?
yves: resolution to keep tests
with and without capitalisation
... there are these test cases, and some others missing from coverage.
... should we just let Leroy know?
Leroy: just let me know once the test input is done, and i will update the parser
<fsasaki> scribe: fsasaki
daveL: are we still missing test cases with the MUST statements?
<scribe> scribe: daveL
felix: many of these tests still
need to be added, e.g. with annotation and locqualissue
... so we definitely need these tests
... and people need to make sure they allocate implementer time for these in Jan and feb
phil: has a look at locqulissue
one, and found useful doc on tests for MUSTs
... but it was difficult to see how manual tests could be formulates
action; felix to distribute action items to define these tests and to provide guideance of how to formulate these tests against rcf2119 table
<scribe> ACTION: felix to distribute action items to define these tests and to provide guideance of how to formulate these tests against rcf2119 table [recorded in http://www.w3.org/2012/12/17-mlw-lt-minutes.html#action02]
<trackbot> Created ACTION-374 - Distribute action items to define these tests and to provide guideance of how to formulate these tests against rcf2119 table [on Felix Sasaki - due 2012-12-24].
felix: thanks for finding people
... but we won't share lists since some contacts are confidential
... asks Yves to handle LC reviews to OKAPI community
dave: I will handle this with andy zydron
felix; will des lead contact with systrans
felix: please let neives know if you receive specific committments to review
des: is there a proforma for requesting
felix: posts proforma, and arle
did some also.
... david was doing a more business oriented one, but is off ill
pedro: will nieves contact people directly?
felix; nieves will NOT contact there people, only tracking who we are contacting
scribe: the best if people contact them directly.
felix: I'm happy to contact the specific people whose details pedro provided
pedro: had contacted them already, but they just need the details of how to submit
felix: the email pasted above, has details of email address where to send comments
felix: these issues raised
already by people within the WG
... issue-58, an example to be fixed in a version value
felix: issues-59, a sentence
about pointing and adding, resolved by clarifying a sentence
circumstance when it is mutually exclusive
... in two section as per above emial
felix: issue-60 loc qual type, with resolution as per above email -which didn't require new type
felix: jorg followed up with
clarification for labelling of the type value table
... which is now changed.
felix: issue-62 standoff markup for global rules
felix: could use standoff and
global attributes, which doesn't make sense since this isn't
the case for local
... this has now been changed in the schema, though it didn't need a change to the examples
... any further question on these, or can they be closed?
<fsasaki> 58, 59, 60, 62
felix: issue-63 on loc qual type conformance
phil: on this, in cngl we worked
on some automated linguistic target qa, based on comparison to
... so it is different to the distinct techniques for specific errors as per other types
... conformance is a term that gives a numeric value to compare target against a reference corpus
... so it may be useful to have
yves: how is this different to loc quality rating?
phil: rating is more formulaic way of computing a rating from other issues, whereas the conformance quality is very useful at segment level
yves: where would the value go, in the severity
felix: would like to delay the decision unitl after chistma
<scribe> ACTION: felix to raise issue-63 again in the 7th jan call [recorded in http://www.w3.org/2012/12/17-mlw-lt-minutes.html#action03]
<trackbot> Created ACTION-375 - Raise issue-63 again in the 7th jan call [on Felix Sasaki - due 2012-12-24].
felix: discussed this last
... both how to cover section 1 and 2 and also on best practices
... are people happy for christian to edit quite freely
<fsasaki> scribe: fsasaki
daveL: would be happy to know
what the time frame the rewriting would be on
... david and I wanted to look into this too
... need to coordinate things
<daveL> christian: what time frame would fit best.
daveL: for me after january 15
would be good
... if you want to work on it now, and we hand it over
<daveL> ACTION: chriLi to provide rewrite of section 1and 2 by mid Jan [recorded in http://www.w3.org/2012/12/17-mlw-lt-minutes.html#action04]
This is now action-377
<daveL> felix: best practice draft and high level use cases
<daveL> .. if anyone want to work on this over christmas, otherwise i will try and pull these together
<daveL> ACTION: felix to pull this material on best practice together onto wiki for people to comment on [recorded in http://www.w3.org/2012/12/17-mlw-lt-minutes.html#action05]
<trackbot> Created ACTION-376 - Pull this material on best practice together onto wiki for people to comment on [on Felix Sasaki - due 2012-12-24].
<daveL> felix: prague face to face planning will be mostly for testing and last minute comments, planning for later in year will be discussed in Rome
<daveL> felix: I wish those leaving now a very Merry Christmas!
<daveL> fleix; arle, who do we still need input from
<daveL> arle: still need to assemble what is still missing
<daveL> felix: we need to compile this by wednesday
<daveL> ... so please everyone check if they are missing some information from actions.
<omstefanov> A Merry Christmas and a "Guten Rutsch" to you, Felix and to the other chairs!
<pnietoca> thank you! Merry Chritsmas to all!
<omstefanov> Merry Christmas to all!
you too, olaf!