Taking notes from the article:

  1. An ideal live programming system should be able to harness the live program to offer improved abilities for code creation and manipulation
  2. We can annotate the source code view with runtime information, such as the most recent result for a certain expression. We can also use the running program as a sort of magnifying glass, which shows us which sections of the code are relevant for the current state. This is a fundamental feature of Smalltalk-based environments 1, 2, where the user can click on a graphical object, and navigate menus to see that object’s definition. In an ideal live coding environment, code and runtime are intertwined, and the user can seamlessly jump between the two
  3. We choose a dataflow-based programming model, where a program is represented as a directed graph of terms. Each term has a list of inputs, and a function that specifies how to compute the output value. A function may be defined as a nested graph of more terms, or it may be a simple atomic operation. A function may also have an external effect, as long as its result value is purely computed from its inputs

from http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&arnumber=6617339






blog comments powered by Disqus