Week 1

  • May 31, 2012
  • 1 Comment

The first several days have consisted of introductions (to the project, to the other members of the project, to the various software that will be used for the project, etc.) and the setting up of the comuters we will be working on. So far, it has been an enjoyable experience meeting the other group members. They really seem like a fun, bright bunch of people, and I expect I'll enjoy working with them this summer.

As a part of the undergraduate research program, we will complete a 2 week long project designed to introduce us to the work we will be doing for the rest of the summer.

The work I will be completing over the course of the summer concerns the rules editing itnerfae of the Looking Glass IDE. Rules, in the context of the Looking Glass, are domain expert-generated scripts meant to detect patterns of code wherein a non-domain expert is likely trying to do something, but does not know the best way of going about doing that something. As an example, a user may attempt to have a character walk diagonally by having the character first move up then move left. In theory, the rules feature will detect this and notify the user that there is a better way of going about having the character walk diagonally (in this case, by putting the move up and mvoe left instructions in a doTogether ordering box).

The domain expert writes and test rules by going to Window > Internal Testibg > Edit code tests... There, the domain expert writes a Jython script that will hopefully introduce a new rule to the Looking Glass IDE.

Presently, the code that deals with the rules testing is tightly coupled with Croquet. This is non-optimal, so it is up to me to figure out how best to decouple the rules testing from Croquet.

Another challenge is that the Jython scripts are currently being tested locally. A more ideal option is to send the Jython script to the server, and have the server run the script and retunr a response (be it positive or negative). I will also have to figure out how best to go about this.

So, this is essentially the work I will be doign for the next two weeks. This work is designed to introduce me to the work I will be doing for the rest of the summer, which will deal with the entire rule testing interface of the IDE.

 -Patrick

Comments

  • kyle

    kyle said:

    <p>The work you've done so far to break-out the test runner is great. I can't wait to read next week's blog... and maybe the ideas you have for making the test output more generic.</p> <p>&nbsp;</p>

    Posted on Jun 01, 2012

Log In or Sign Up to leave a comment.