Skip to content

Interim Draft 1 First feedback 23 1 19

Louis Kueh edited this page Jan 23, 2019 · 2 revisions

TODO

  • remove why functional language, vs code
  • motivation - add valid reason to learn f#
  • context: dif programming languages, ides, tools that help the programmer. How does this project compare to them?

Write aim

  • Better than existing tooling for similar
  • Generalise how similar it is to other languages such as haskell
  • Better tooling for people learning F# (beginners) -> related work (tools have...) -> how to fufill (implementation)
  • type of tooling

Project Specification

  • for each rules from the analyer - what we can take, what we can't take (e.g. already in use)
  • things we are going to do
  • After writing aim, write objecties, which are more specific than the general aim
  • Background F# Compiler -> tech in F# compiler = classes of things

What you can do

  • types of help that we can give - more general view of the help
  • help understanding compiler levels
  • help with stylistic issues
  • not compilers errors/not
  • MORE STRUCTURE!

It's more than explaining compiler errors. You have got to use context to make suggestions. Obtain context via what the user wrote. E.g. did you mean the functions to be this. Give specific corrections to the code

  • heuristics - name, white space sensing class of errors
  • bracket errors

Identify different types of work -> different classes -> different types (high level)

Planning

Min 5 items max 15

  • depedencies
  • tasks, risk, fallbacks, extensions
  • risks -> too long or no longer possible

Evaluating

  • simulate error
  • test suite, beware of false positive
  • possibly use two sets of test subjects, see if one understands the test with no analyzer, see if another understands faster with the analyzer
  • self evaluate by introducing code on which to run the analyzer, then evaluate if it's correct

Implementation

  • Saying what progress you've made
  • no need to go into details

Source reliability

  • Rate it by yourself
  • Source is reliable, author is reliable, or reliability does not matter
  • Section on how you evaluate each source systematically. For example, use colour or stars to indicate.

Overall

  • Motivation restructure
  • add planning
  • evaluation - link to the plan
  • Different categories for analysers that are higher level.

Clone this wiki locally