• WP1 (Centres Setup)	1 PM
  WP2 (National Registry)	1 PM
  WP8 (LAP)	30.5+23 PM
  WP9 (Tool Adaptation)	0 PM
  WP10 (Metadata Adaptation)	2 PM
  WP11 (Management)	2.5 PM

Compared to the proposed milestones from the project proposal, the project effectively started 18 months later than originally expected. The CLARINO grant was finally approved in the spring of 2012; the job search for the main LAP position completed in August that year, and LAP work effectively started in October 2012.

• 8.1	LAP requirement and interface specification	2012	4	2013	3
  8.2	LAP first testable prototype	2013	1	2014	1
  8.3	LAP inclusion of analysis tools	2014	1	2014	3
  8.4	LAP first production version	2014	3	2015	1
  8.5	LAP documentation, workshop, evaluation	2015	2	2016	3
  8.6	LAP Tuning and complete version	2016	4	2017	3

  9.1	Tool adaptation, testable versions	2013	1	2014	3
  9.2	Tool testing and final versions	2014	2	2015	2

Due to repeated leaves of absence of core staff, LAP personnell allocation has lagged behind projections by close to nine person months at the end of 2014. Accordingly, the implementation of LAP milestones 8.2 and onwards have fallen behind schedule by a little more than half a year. With its progress report for the second tertial of 2014, the LAP team proposed the following schedule revisions:

• 8.2	LAP first testable prototype	2013	1	2014	3
  8.3	LAP inclusion of analysis tools	2014	3	2015	1
  8.4	LAP first production version	2015	1	2015	3
  8.5	LAP documentation, workshop, evaluation	2015	4	2017	1
  8.6	LAP Tuning and complete version	2017	2	2018	1

It remains to be determined where things stand regarding WP9, and whether and how the schedule should be adjusted there.

This work package comprises the following sub-tasks: (a) user requirements and technology survey (including the wider CLARIN and META-NET contexts); (b) architecture specification; (c) middleware design and implementation; (d) user interface design and implementation; (e) integration of language analysis components; (f) support and training to component providers and users (including documentation); (g) portal maintenance and operations; and (h) liaising with other CLARIN(O) participants.

The analysis portal presupposes authentication and authorization services in WP3, access to the data in the repositories at the content providers, and access to compatible tools (WP9).

[...]

LAP will assist in license clarification, interface definitions for conversion from and to inter- change formats, ‘wrapping’ of tools for HPC use, general ‘hardening’ to improve scalability and interoperability, and on-line documentation. Technology developed (or currently used) in Norway will be complemented with standard language analysis tools available from inter- national players, with a bias towards general-purpose, open-source systems. [...]

Existing grammars and taggers will be adapted so as to produce a CLARIN compatible format and become more easily integrated in pipelines within the LAP, as well as in other conformant platforms. This work will be carried out by tool providers in cooperation with IFI (UiO). Rule-based taggers for Norwegian Bokmål and Nynorsk (comprising a tokenizer, morphological analyser, guesser, and rule-based disambiguator), as well as statistical taggers for orthographically transcribed spoken Norwegian, Swedish, Danish, Icelandic, and Faroese will be adapted. The tagger adaptation will be carried out by the Text Laboratory (UiO), while some work on the Oslo–Bergen Tagger will be done in cooperation with Uni Research. Furthermore, the constraint grammars for Sami will be adapted and optimized by UiT. For the three Sami languages, there are efficient morphological transducers and, for one of them, grammatical and dependency parsers available.