Difference between revisions of "Apertium Turkic/TODO"

From Apertium
Jump to navigation Jump to search
(Created page with " == Website == Get [http://turkic.apertium.com/ http://turkic.apertium.com/] up and running. === software infrastructure === * <s>Get apertium-apy working stably</s> * [http:...")
 
Line 23: Line 23:
   
 
== Things that need to be figured out ==
 
== Things that need to be figured out ==
  +
* [http://www.google-melange.com/gci/task/view/google/gci2013/5872152972623872 How can we count lexc stems effectively?] - JNW's bash script can be generalised (and rewritten in python), and it'll come close
  +
  +
=== Issues introduced by new build process ===
  +
* How can we do single-category testvoc now?
  +
* How can we make vanilla transducers (without MT-specific "wrong" POSes)
  +
* How can we count trimmed stems?

Revision as of 09:02, 3 January 2014

Website

Get http://turkic.apertium.com/ up and running.

software infrastructure

optional: spell checker stuff

what to include

make the following pairs available to the site:

  • pairs: kaz-tat, tur-kir,
  • transducers: kaz, tat, kir, tur, bak, chv, kum, nog, kaa, uzb?, tuk?

prettifying

Things that need to be figured out

Issues introduced by new build process

  • How can we do single-category testvoc now?
  • How can we make vanilla transducers (without MT-specific "wrong" POSes)
  • How can we count trimmed stems?