Specifying the scope of the problem before solving

    On the very first week, the new idea about incremental view updating was accepted. However, after working with specific example that Prof. Kato gave, there were some troubles with tracking the information from the view back to the source of the forward transformation.

    One week later, Prof. Hu sent us a paper named "Origin tracking", and we should consider if we can apply the technique to solve our problem. Before the meeting today, the main problem is that how create rule from graph input so that we can trace. After the discussion and explanation from Profs. , there are 3 important things that I recognize:

    First, one is that creating rule from graph input isn't the main problem because in the Term Rewriting System, the comparison between notes can be animate at any points. Moreover, with the simple rule that I suggested in the slide, it is actually the rule as it is.

    Second, the previous example of Prof. Kato was so complicated (with the 'or' operator - apply for father|son and mother|daughter). So, I need to create a more simple one (just very simple).

    Third, the final things that is very important to me is that the scope of the problem. It's very large example so that I can't test my expressions (extend-where, delete-where and replace-where). As a result, I need to create a simple example so that I can test my expressions. After that, I can create more complex and complicated compositions.

    That's all I have to remember and keep in my mind (keep it as simple as possible to test, and then develop to complicated one)