MONK Interface call
Dec 10, 2008

Present: Catherine, Stéfan, Stan, Duane, Carl, Andrew
Regrets: Amit

We discussed the Word Cloud tool. It does two things: it reads the workset and displays its lemma frequencies. It alternatively reads the two menus and produces Dunnings output.

Catherine suggests that we have a separate toolset for workset creation and editing. Stefan suggests that at least we have a separate step in all toolsets. In either case, the goal is to provide more screen real estate.

Carl, Catherine, and Amit have spoken about the process of having text incorporated into the help system. This will replace our earlier strategy of writing into the javascript code. So each tool or toolset will have an HTML file. We will need to migrate the few pieces we currently have at https://apps.lis.uiuc.edu/wiki/display/MONK/Wiki+version+of+help+text

Can one of us write the more technical descriptions that Martin has requested? We would like these specifically geared to each of the tools. Maybe Duane can tackle them, if this doesn't take him off another critical path. For each tool, we need to explain about the analytics involved--what do people need to know to use it properly, to roughly explain what they've done, and to understand how confident they could be about the results.

We talked briefly about the workbench tool arrangement for classification. Our goal with that toolset is to be able to support easy iteration, but the tiny palettes are a bit of a problem.

Andrew has done the sample of a documented tool (the SEASR manager) using the jsdoc toolkit. There is a link in the ticket for the documentation compiler. His next step is to expand the component creation guide and that link will go there too. There is also a Getting Started guide about getting your development environment set up.

Andrew has also stripped out the unused code to make it smaller for developers to download.

Next week we continue our focus on documentation.