2013-03-15- Risten2 Meeting

Tomi has worked on the article display, but there are some issues both with the app (some part crashes) and with the data (what to present and how, and it is not in one common format).

Then there is the issue with searching and missing hits for dictionaries with their first hit after the 1000th total hit (these dictionaries are not listed as possible sources, only after the search becomes so narrow that their first hit appears above the 1000th position).

Tomi wants to rewrite the app using another JS framework: DurandalJS, it provides a framework for optimizing the code and adjusting the application to different platforms.

Sjur suggested that we in the browser search each dictionary independently, to get hits back fast, and instead update the hit list as more hits are returned. Then each dictionary's hit list can be restricted to only the 50 first hits or so. Should give a much more responsive application, at least from a user perspective.

Decision:

• rewrite to search each dictionary independently
• rewrite to use DurandalJS
• we need to look at standardising the xml structure next week
  • the default structure is like smenob and fkvnob
  • that structure needs to be enhanced to handle terminology data

Ideas for the interface:

• dictionary buttons are greyed out as the dictionaries are searched
• the greyed button should contain a spinning wheel showing the ongoing search
• if there are hits, they get full colours again
• if there are no hits, they disappear (with some fancy css3 effect: ))

Risten2 beta DEADLINE: April 10.