11:42:40 RRSAgent has joined #mlw-lt 11:42:40 logging to http://www.w3.org/2013/07/24-mlw-lt-irc 11:42:49 meeting: MLW-LT WG / ITS IG 11:43:08 agenda: http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Jul/0040.html 11:43:36 regrets: Arle, Declan, Dave, Karl 11:43:40 present+ felix 11:43:44 chair: tbd 11:43:47 scribe: fsasaki 11:44:00 topic: roll call 11:44:08 checking attendees ... 11:44:12 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:00:30 Jirka has joined #mlw-lt 12:00:42 dF has joined #mlw-lt 12:01:02 Felix, is this the channel for the upcoming call? 12:01:16 hi dF, yes, it is 12:01:27 so far it is only Jirka and me 12:02:19 good, do we change IRC when we go for the second agenda point? 12:02:46 no, we stay here. I will make that clear in the minutes 12:02:59 Yves_ has joined #mlw-lt 12:03:07 Des has joined #mlw-lt 12:03:29 present+ yves, des, df, jirka, phil 12:03:49 present+ Yves 12:04:56 present+ ankit 12:05:03 Ankit has joined #mlw-lt 12:05:16 present+ Ankit 12:05:26 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:06:10 chriLi has joined #mlw-lt 12:06:18 present+ chriLi 12:06:36 agenda is here http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2013Jul/0040.html 12:06:54 topic: MT Confidence clarification (MLW-LT WG topic) 12:07:34 yves: there was the question wherer MT Confidence should be restricted to self generated score or not 12:07:39 .. felix made change after the discussion 12:07:50 result would be here http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#mt-confidence-score-generation-tools 12:08:03 .. woudl you have any comments? Time to look at it? That is the current text 12:08:22 phil: didn't have time to look at the text, but at the thread 12:08:36 ... I had in my mind something generated by the MT engine 12:09:06 .. but I am happy enought to broaden this, as some people said in the thread 12:09:23 .. as long as we are careful about the definition 12:09:33 yves: any other comments? 12:10:14 Yves_, comment from me on IRC: the change I made was only removing the "self reported" part and adding above note 12:10:42 Yves_: MT Confidence still means the same, but it is not linked only to the MT engine 12:10:51 Yves reading the note in the link above 12:11:36 Yves: the note makes the distinction between the two types of generators for confidence information 12:11:48 yves: so it seems everybody is ok with the change 12:12:01 yves, where was no issue, but I have added an item in the change log 12:12:04 topic: XLIFF mapping 12:12:11 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:12:20 yves: several entries in XLIFF mapping 12:12:24 http://lists.w3.org/Archives/Public/public-i18n-its-ig/2013Jul/0057.html 12:13:00 yves: mail is about update of LQI and ... 12:13:26 dF: I had dropped of, what did you hear last? You didn't hear what I said about MT Confidence 12:13:45 yves: no - but go ahead 12:13:57 pnietoca has joined #mlw-lt 12:14:09 topic: mt confidence again 12:14:12 present+ pnietoca 12:14:25 present+ chriLi 12:14:27 dF: tools should rather use lq rating than mt confidence 12:14:45 phil: one thing that came up in relation to broadening mt confidence 12:14:54 .. we don't have a standoff version of mt confidence 12:15:03 .. not sure whether that would be an issue 12:15:23 dF: this category doesn't need standoff because it does not need mashing 12:15:51 .. if you broaden the definition you open the use case to provide conflicting confidence scores 12:16:12 .. we never intended to have standoff because we said this is self confidence 12:16:25 .. one should use rating because it has standoff 12:16:45 yves: it doesn't, only lqi and standoff provenacne have rating 12:16:50 .. what you said is the same for rating 12:17:03 .. you can use different engines but there is only one way to mark it up 12:17:21 dF: don't you feel that third party conformance is for rating? 12:17:28 yves: I looked at the example fo quest 12:17:41 .. the value they are getting is the same like the mt confidence self reported 12:17:54 .. to me lq rating is much more like a composite index based on many things 12:18:02 .. that is much more linked to conformance and human voting 12:18:14 dF: we have one issue or the other 12:18:32 .. usage of multiple scores either on confidence or rating 12:18:40 yves: don't think that this is a huge issue 12:18:51 .. don't think that these data categories will be used a lot 12:19:21 phil: isn't this tied up with MQM / QTLaunchpad? 12:19:26 yves: LQI is, not sure about rating 12:19:36 Zakim has joined #mlw-lt 12:19:49 phil: will QTLP have a rating? 12:19:53 yves: don't think so 12:20:07 .. the quetion is: if you have a value like quest where should it go 12:20:26 phil: is it harricy either mt confidence or lqi? 12:20:31 .. until we have lots of use cases 12:21:00 .. was not sure if we need a new quality type 12:21:20 yves: looking at declan's feedback he was ok with result of quest on MT confidence 12:21:43 dF: quest or other MT metrics are closer to MT confidence 12:22:18 .. if you are an MT enginge owner it is breaking of transaction if you allow for overriding self reported confidence with a third party tool 12:22:29 yves: this is processing related 12:22:30 q+ 12:22:50 yves_, can I say something? 12:23:11 yves_: people will not read the spec and will just put an MT confidence score 12:23:31 phil: don't feel strongly enough to stand in the way of broadening this 12:24:32 felix: to dF, would current draft be ok? 12:24:45 .. and check with MT providers if it's ok 12:24:51 .. we need more feedback 12:25:04 .. asking if we can move forward with current version 12:25:13 .. we could reverse if needed 12:25:24 .. there is no testing impact 12:25:52 df: it's not substantive 12:25:59 .. so change would be ok 12:26:17 Pedro has joined #mlw-lt 12:26:25 second quaestion: what's the current version 12:26:25 present+ Pedro 12:26:35 .. self-reported? or not? 12:26:45 felix: not self-reported 12:27:06 .. action would be whether we revert or not 12:27:45 felix: would current version be ok? 12:27:51 df: either is fine 12:28:02 felix: then let's use the current one 12:28:18 .. and get an action for the feedback 12:28:50 action: davidF to get feedback on the MT Confidence broadening during PR - due 29 August 12:28:50 Error finding 'davidF'. You can review and register nicknames at . 12:29:15 http://lists.w3.org/Archives/Public/public-i18n-its-ig/2013Jul/0057.html 12:29:18 action: david to get feedback on the MT Confidence broadening during PR - due 29 August 12:29:18 'david' is an ambiguous username. Please try a different identifier, such as family name or username (e.g., dlewis6, dfilip). 12:29:28 action: dfilip to get feedback on the MT Confidence broadening during PR - due 29 August 12:29:29 Created ACTION-559 - get feedback on the MT Confidence broadening during PR [on David Filip - due 2013-08-29]. 12:29:36 topic: XLIFF mapping 12:29:44 http://www.w3.org/International/its/wiki/XLIFF_1.2_Mapping#Text_Analysis 12:29:45 yves: first on text analysis 12:29:54 dF has joined #mlw-lt 12:29:55 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:30:21 yves introducing on text analysis mapping 12:30:32 yves: any comments / objections against the mapping? 12:30:36 no comments 12:30:40 yves: now lq rating 12:30:45 http://www.w3.org/International/its/wiki/XLIFF_1.2_Mapping#Localization_Quality_Rating 12:31:29 yves: no resolution on that so far - also question on phil 12:32:16 yves: do we want to allow granularity for lq rating 12:32:28 .. or have it only on the xliff target level 12:32:40 .. the ITS 2.0 definition does not say "you should or should not go on the word level" 12:32:50 phil: discussed that last week 12:33:01 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:33:28 des: what would be the case against this? 12:33:45 dF: in source content, no a sub sentence level 12:33:58 .. lq rating can appear. So we shouldn't prevent that in XLIFF 12:34:08 .. we can discourage it, but it shoud be allowed as a mapping 12:34:33 yves: why would we even discourage it? 12:34:44 dF: it would compete with terminology information 12:35:01 yves: disagree - in term confidence is avail., it is very different from a vote 12:35:24 des: you could use it to mark up terminology 12:35:35 dF: you would not mark terminology but terminology errors 12:35:47 .. that is different - you could still have terminology markup 12:36:29 yves: confidence in terminology does not say "it is a good translation" 12:36:48 dF: so we drop the language about local level 12:36:53 yves: agree - will make the change 12:37:04 .. there is not a lot to change anyway 12:37:09 yves: now MT confidence again 12:37:14 http://www.w3.org/International/its/wiki/XLIFF_1.2_Mapping#MT_Confidence 12:37:18 .. now the XLIFF mapping 12:37:49 yves: we had a discussion about origion XLIFF attribute 12:37:58 .. we discussed that in bled 12:38:06 .. we decided not to overload the entry 12:38:22 .. this linked to the question we had before 12:38:32 .. when multiple engines have a score, what to do 12:38:49 .. there is no resolution, having standoff here would be a major change 12:39:10 christian: this would be something we would put on our lists of enhancement requests for ITS2? 12:39:13 yves: exactly 12:39:19 .. how to put that on the list 12:40:18 felix: would put that on the ITS IG wiki for the time being 12:40:30 will have a pointer to the location soon 12:40:55 http://www.w3.org/International/its/wiki/XLIFF_1.2_Mapping#Locale_Filter 12:41:03 yves: now locale filter mapping 12:42:00 hi yves, all, I would propose to put the issues and proposed features here http://www.w3.org/International/its/wiki/IssuesAndProposedFeatures . Currently it is focused on ITS 1.0, but we can update it with proposals for ITS 2.0 12:42:21 yves describing the locale filter mapping 12:43:11 yves: nobody has made a comment no this one so far 12:44:23 dF: wonder - there might be more for XLIFF 12:44:32 .. we should think of valid transformations 12:44:42 .. what should be guidance to get from one to the other 12:45:10 .. ITS information is not longer available 12:45:32 .. the extractor with extended information - what are you doing 12:45:34 yves: you merge the target 12:45:41 david: it no longer contains the extended information 12:45:48 yves: why would you need the extended information? 12:45:59 david: you expect to merge back into multiple languages 12:47:00 .. if the extractor does not define a target, the LSP still will need to define a target 12:47:17 yves: if they understand ITS - otherwise you cannot process the file 12:47:57 pedro joining the meeting 12:47:57 present+ pedro 12:48:18 discussion about xliff - ITS extraction and merging 12:48:32 yves: think that has not to do with the mapping, it is an XLIFF problem 12:49:49 leroy has joined #mlw-lt 12:51:26 action: davifF to draft guidance for second processing of Locale Filter when target is not defined in first processor in wiki 12:51:26 Error finding 'davifF'. You can review and register nicknames at . 12:51:39 tadej has joined #mlw-lt 12:51:42 action: davidF to draft guidance for second processing of Locale Filter when target is not defined in first processor in wiki 12:51:42 Error finding 'davidF'. You can review and register nicknames at . 12:51:49 dfilip to draft guidance for second processing of Locale Filter when target is not defined in first processor in wiki 12:52:43 yves: one question on termInfoRef is still open 12:52:50 .. we both came to an agreement on that 12:53:07 http://lists.w3.org/Archives/Public/public-i18n-its-ig/2013Jul/0022.html 12:53:51 felix: I agreeded with the proposal to say that there is no hard wired definition what is available then one resolves termInfoRef 12:53:55 yves: correct. 12:54:02 .. if you have disagreement comment on that asap 12:54:11 .. otherwise no other mapping issues currently 12:54:15 topic: aob 12:54:50 present+ leroy 12:54:52 thank you bye! 12:54:56 dF: next week we may be able to use this gotomeeting again 12:55:00 yves: thanks all, bye 12:55:05 adjourned 12:55:45 I have made the request to generate http://www.w3.org/2013/07/24-mlw-lt-minutes.html fsasaki 12:55:59 Thanks for scribing Felix. Bye. 13:10:38 Jirka has left #mlw-lt 14:24:06 Zakim has left #mlw-lt