Opened at 10:13.
Present: Børre, David, Jovsset, Maja Lisa, Sjur, Thomas, Tomi, Trond
Absent: Ciprian
Agenda accepted as is.
Released Thursday last week. There's a bug now for the CG part, Trond and Saara are working on it. Otherwise the release was a big success:)
Also good coverage in the Sámi news, see this link.
The PR process for the Oahpa! release:
- first press unit at the University
- they make a press release / a story in our internal media
- the external journalists read it
- they come for making reports
Things to learn from the press coverage of the events we have had so far - use the professionals! That is, the PR people at UiT or SD. Don't write the press release ourself.
Meeting memos can be found at http://giellatekno.uit.no/ped/index.html#Meeting+memos
TODO
- Register oahpa.no ( Trond )
Børre has installed giza++, mosesdecoder, srilm, mkcl etc on the XServe, which will facilitate statistically based machine translation. There are experiments with the two ways of doing MT:
- rule-based machine translation
- based on apertium: sme2smj
- based on cg?
- statistical machine translation - ideas/plans for the future:
- sme2smj (for comparison),
- sme2nob (gist translation),
- nob2sme (OpenOffice.org localisation etc.)
To accomodate future enhancements in different directions (in rough order of importance):
- test bench for all parts of our language technology efforts
- test bench enhanced, but not yet complet
- set up the Leopard Server features for collaborative support:
- permanent chat rooms
- stored (and indexed) chat transcripts of the chat rooms
- iCal server / group calendars
- wiki
- wiki? on G5 (is part of Leopard Server) or other web-based documentation
- improve Forrest stability and i18n support ( the divvun crashes)
-
Sjur has been working on better i18n and pdf rendering
-
Børre has some ideas for getting back to serving static html files
- reorganise the documentation:
- differ between target groups
- get better grouping
- decide what to write in forrest and what in wiki (cf. Apertiumand http://xixona.dlsi.ua.es/apertium/) for a similar split)
- update/add missing parts
- migrate lexc lexicons to XML, splitting the task
- Name lexica (the Name project)
- Dictionaries (already in XML, task is to integrate them)
- At least migrate the lexc open POSes (Komi as a pilot case)
- change the look of the documentation web
- sfst?
- replacement for xfst -> see omorf
, omorf wikiand omorfi
- replacement for hunspell/open-source proofing tools
- investigate the NSIS installer, potentially replacing the InstallShield package from Polderland
- corpus content moved to Max Planck repositories? Norsk språkbank?
- update infrastructure to allow content-restricted spellers for special target groups
SVN issues:
- http access not yet available (https only at the moment, but it works as anonymous checkout if the user is not specified)
- read access to the whole repo is working, BUT:
-
gt/*/polderland should be protected
-
plan should be protected
- certain users need to be on the UiT VPN to be able to commit (bug #705)
- UTF-8 problem in svn commit mails: the commit log text is garbled
TODO:
- infrastructure remake: ( Børre, Ciprian, Saara, Sjur, Tomi, Trond )
- more modularised make / build infra (prepare for smn, sms, sjd, others)
- merge gt, kt and st into one
- modularised preprocess and spellrelax
- make a test-all target that runs all tests we have ( Ciprian, Sjur, Trond )
- delayed until we have restructured the make/build process
- define and document testing routines ( Ciprian, Sjur, Trond )
- delayed until we have restructured the make/build process
- follow-up migration to svn:
- close the open SVN repository at requested paths ( Ciprian )
- completely closed at the moment, until we have solved the path-based control
- prepare and discuss with external users: Jack Ruether ( Trond )
- test iCal Server (on G5) ( Børre )
- remove TOC from the giellatekno home page by using dispatcher ( Trond )
North Sámi
(nothing new, see proofing bugs below)
Lule Sámi
(nothing new, see proofing bugs below)
South Sámi
Possible linguists to be invited to the seminar:
- Håkan Rydving - Trond
- Ove Lorentz - loan word standardisation in Snåsa Oct. 2006
- (decision made: 26. july 2006) - contacted, will come Wed+Thurs
- Michael Rießler - Trond
- Eino Koponen - Sjur
- Torbjörn Söder - David
- Lars-Gunnar Larsson - David
We should arrive such that we are ready to start Monday morning.
TODO:
- step two for adjectives ( David, Maja, Trond, Sjur )
- we'll try to find a time this week as well ( David, Maja, Sjur, Trond )
- finish reformulating the proper noun grammar like the verbs ( Sjur, Trond )
- Missing list (focus on closed class words and addition on missing nouns, adjectives and verbs) ( David, Jovsset )
-
David made a new missing list from an OCR-ed book
- workshop on South Sámi orthography and grammar ( David )
- date: last week of February (23.-27.)
- Place: Trondheim
- meeting rooms ( Sjur )
- ordered at the Univ. - not confirmed
- we'll cover travel and hotel expenses
- work on the agenda ( David, Sjur )
- meeting Tuesday 9.00 to make it final- DONE
- prepare a presentation on derivational grammar ( Thomas )
- clean the smanob data ( Ciprian, Trond )
- Done ( Ciprian ), result in smenob/src/, but the list is short.
Instead of building our own webforms and back-end update scripts, use XForms with a premade connection to our xml db. Orbeon XForms is such a tool (open source).
From the meeting with the terminology and IT teams last week:
- no major rework on the present search interface now
- no work on the editing section; instead:
- add existing lists of sanctioned terminology as separate term entities
- add a dictionary if we can make one with sufficient quality
This means the following tasks:
- find already approved lists, in paper or electronic form ( term team )
- convert paper lists to electronic lists ( term team )
- convert lists to standard XML ( Sjur, Tomi )
- add prepared lists to risten.no ( Sjur, Tomi )
TODO:
- fix i18n bug in risten.no/G5 (so they will work without the proper locale request) ( Sjur )
- fix bugs in lexc2xml; add comments to the log element ( Saara )
- finish first version of the editing ( Sjur )
- test editing of the xml files. If ok, then: ( Sjur, Thomas, Trond )
- make terms-smX.xml <=== automatically from propernoun-sme-lex.xml (add nob as well) (the morphological section should be kept intact, in e.g. propernoun-sme-morph.txt) ( Sjur, Saara )
- convert propernoun-($lang)-lex.txt to a derived file from common xml files ( Sjur, Tomi, Saara )
- implement data synchronisation between risten.noand the cvs repo, and possibly other servers (ie the G5 as an alternative server to the public risten.no - it might be faster and better suited than the official one; also local installations could be treated the same way)
- start to use the xml file as source file
- clean terms-sme.xml such that all names have the correct tag for their use (e.g. @type=secondary) ( Thomas, linguists )
- merge placenames which are errouneously in different entries: e.g. Helsinki, Helsingfors, Helsset ( linguists )
- publish the name lexicon on risten.no ( Sjur )
- add missing parallel names for placenames ( linguists )
- add informative links between first names like Niillas and Nils ( linguists )
Dictionaries
Our first electronic dictionary was released together with Oahpa! last week. There are certain problems on the Windows side for StarDict. Mainly, we need to improve our user documentation, but it is too short at the moment. And the StarDict download for Windows need to be a zip file, not tar.
The dictionary works well on Linux as well, Børre tested it with QStarDict. But the documentation is lacking (missing?) for Linux.
There are other dictionaries available for StarDict on http://xdxf.revdanica.com/down2/index.php?down_format=StarDict
TODO:
- Set up the apache server for the risten.no beta on the linux box + Xserve. ( Børre )
- set up Tomcat and eXist (latest svn) on XServe ( Børre )
- set up risten.no on eXist/XServe (as a beta version site) ( Sjur )
- set up required infra for smenob on risten.no/XServe ( Sjur )
- Continue the dictionary infrastructure discussion ( Ciprian, Sjur, Trond )
- end user documentation (how to download and install) ( Ciprian, Trond )
- make a public release
- Contact Davvi Girji about cooperation on electronic dictionaries ( Sjur )
- update and split dictionary user documentation before Thursday ( Børre, Sjur, Trond )
Hunspell
Making compounding work correctly in Hunspell is too much work, the formalism is too cumbersome and different from what we have. We are looking at alternatives to Hunspell, and will return to this issue later.
The version we have at the moment (beta6) will be our last one for a while.
OpenXSpell
Working everywhere (almost) in MacOS X applications, provides Sámi spell checking system-wide through a hunspell back-end. OpenXSPell
As a middle-man between Enchant and MacOS X speller API, this is still interesting. As a direct Hunspell front-end it is not, cf above.
Testing
Spelling Error Markup
TODO:
- Set up ways of adding meta-information (source info, used in testing or not, added to lexicon or not) ( Saara )
- test new and nested error markup ( Sjur )
- nesting still needs to be tested, depends on new ccat feature
Speller testing
TODO:
- test the error type selection feature in ccat ( Sjur )
Testing open-source Norwegian spellers
Sjur has invited the open-source group to test their spell-checker using our test bench. The response has been positive, we'll see what happens.
We should go to their developer meetings, and present our work and how to work with language technology.
Speller bugs
New lexicons compiled by Tomi .
List of bugs returned from Polderland:
Open issues based on test results:
sme
Version: Davvisámi, version 1.1, 2008-12-17
- 399 - REGERSSION: missing numerals (plural forms)
- 425 - X not recognized; single letters were left out - OPEN
- 435 - roman numbers - inflection of single letter numbers rejected, as well as some complex numbers (but is ok in smj ) - still OPEN
- we should pregenerate all numbers once and for all, and store them in a separate lexicon file
- 581 - consonant doubling - seems FIXED
- 595 - prefix+name without hyphen ( ovdaLot instead of ovda-Lot ) - still OPEN
- 603 - suomabealdi accepted - still OPEN
- 606 - compound-tags LEXICON VUOHTA - OPEN
- 613 - short gen. as second compound part - still OPEN
- 619 - numerals and pronouns to NAMÁK and SASJ fails - vihttasoarttat remaining - still OPEN
- 629 - a taking part in compounding without hyphen - still OPEN
- only open case has word A-finálaid compounded
- 642 - noun/adj/proper + hyphen + ain - still OPEN
- 647 - numerals+NOUN - still OPEN , open case has uppercase letters
- 648 - unmotivated suggestions with numeral+noun - still OPEN
- 728 - vowel shortening GenCmp+Left-tagged - still OPEN
- 779 - caseforms of pronoun okatahat
smj
Version: Julevsáme, version 1.1, 2008-12-17
- 435 - roman number - single letter numbers now recognised
- we should pregenerate all numbers once and for all, and store them in a separate lexicon file
- please note that inflection of single letter numerals is fine in smj , as opposed to sme
- 594 - REGRESSION: lågenanguoktáj not recognized
- 595 - prefix+name as split comp without hyphen - still OPEN
- 596 - C-giellan is not accepted - OPEN
- 647 - numerals+NOUN - still OPEN , open case has uppercase letters
- 648 - unmotivated suggestions with numeral+noun - still OPEN
- 650 - noun prefix+name compound without hyphen - still OPEN
- 652 - UPPERCASE-typos only get acronym-suggestions - still OPEN
- 692 - numeral-variants - all but one fixed ( gáktsalågenantjuotakta ), but still OPEN
- 711 - REGRESSION: numeral compounds and cases ( 10-biejvvásattja ) - FIXED
TODO:
- document how compounding is controlled in the PLX conversion ( Tomi )
Hyphenator bugs
Open issues based on test results :
sme
Lexicon version: Davvisámi, version 1.1, 2008-12-17
- 468 - Márkomeanu - still OPEN
- 547 - hyphen in front of vowel: Lotnolasealáhusas - still OPEN
- 548 - mid syllable hyphenation: Háliidivččen - still OPEN
- 549 - division without hyph: Váccedettiin - still OPEN
smj
Lexicon version: Julevsáme, version 1.1, 2008-12-17
- 547 - hyphen in front of vowel: Jienastimnjuolgadusá and Orgánajs - still OPEN
- 670 - Hard hyphen replaced with soft hyphen: 10-biejvvásattja (the word is not rec.; Bug #711) - still OPEN
TODO:
- fix PL hyphenator errors ( Tomi )
Installer changes
TODO:
- make a Mac installer for the InDesign Divvun tools ( Tomi )
- replace InstallShield with an open-source alternative ( Tomi )
- meeting Tuesday 12.30 to go through the present installers ( Sjur, Tomi )
User documentation
TODO:
- InDesign documentation ( Sjur )
- Norwegian translation received from Davvi Girji
1.2 release
Content:
- several smj bug fixes
- lexicalisations
- InDesign Mac & Win
- new OOo beta
- improved installers, at least for Mac, preferably also for Windows
Winter holidays
| Who |
When |
| Sjur |
16-20.2. |
| David |
19-20.2. |
| Børre |
nothing |
| Maja |
nothing |
| Thomas |
nothing |
| Trond |
not holiday, but going to Greenland 16.2.-10.3. |
| Tomi |
nothing |
| Jovsset |
nothing |
School in Troms have second week in mars
Sámi week in Tromsø
The Oahpa! release was awesome, see above. Also the dictionary release. The library presentation had too few attendees:/
The meeting with the minority language watch-dog was ok. Trond reported on several points, especially the lack of support for the Sámi letters in governmental registries.
Text to speech
The TTS meeting with Antti was held (without Maja and Thomas). Topics:
- what kind of feedback is needed from us? Quite a lot of linguistics
TODO:
- meeting with Antti Suni during the Sámi week ( Maja, Ove Lorentz, Thomas, Trond )
TNC, rikstermbanken and Sámi terms
- Mekanihkartearbmat
- Pedagogalaš ja psykologalaš tearbmat (?)
It should be ok, but they need to send a formal request, to samediggi@samediggi.no. From there it will be forwarded to the relevant person.
Corpus contracts + open source
Postponed until the svn repository is fully functional (it is too open now).
Other issues
- Maja would like to have a bigger monitor -> coming soon (it is ordered)
- nothing new yet - Børre will call and check
The next meeting is 2.3.2009, 9.30 Norwegian time.
The meeting was closed at 12:38.