Library of Cases
See Database Interop Hackathon/Use Cases for hackathon use cases.
The library of canonical uses would serve many uses
- examples for classroom instruction
- test data sets for developing software
- target data for comparative evaluation of software
- instances of format use
However, there is one use that may not be obvious to people. The world is full of talented programmers looking for things to do. If we want to harness this talent, or to draw in people whose first love is not evolutionary analysis, we need to make it clear what we do on a day-to-day basis. This means not only providing generalized descriptions of "use cases" (see the extensive use case document we generated for the NESCent phyloinformatics hackathon) but also providing supporting material to study specific instances, including sample input data sets, output results, and precise descriptions of software used.
NESCent's phyloinformatics projects for the Google Summer of Code have attracted interest from many talented young programmers. There is an enormous potential here to get valuable work done on open-source projects!
Analysis of current needs
(what do we have now? what is missing? what challenges are coming up?)
Starting points include the NESCent use case document from the phyloinformatics hackathon.
Another approach to assessing and anticipating user needs is to ask what kinds of programs are in use (Felsenstein's list) and to ask what kinds of ad hoc scripts are begin written currently, e.g., http://www.personal.uni-jena.de/~b6biol2/ProgramsMain.html.
Goals for the working group
(specific goals for this topic)
Strategy for achieving goal
(be sure to include specific deliverables or milestones)
Start with a lightweight implementation such as a wiki with attached files.
After getting an initial set of cases, do some user testing (ask users to execute examples; ask computer-savvy students to write a script for a given case)
Re-evaluate implementation (continue wiki? impose some kind of use-case formula? )