Changes between Initial Version and Version 1 of Ticket #81


Ignore:
Timestamp:
May 4, 2006 10:20:22 AM (8 years ago)
Author:
malcolm.wallace
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #81 – Description

    initial v1  
    11The implementation of type inference in yhc (also used in nhc98) is occasionally incorrect, always difficult to understand and modify, and entirely undocumented.  As a result, there is no hope of extending it to deal with multi-parameter type classes, functional dependencies, associated types, GADTs, or any of the other exciting and widely-used type-system extensions.  The solution - rip it out and replace it completely.  There are a few key requirements:  it must be maintainable, it must be correct, it must give reasonable error messages.  Speed is not a goal. 
     2 
     3= Interested Mentors = 
     4 * Malcolm Wallace 
     5 
     6= Interested Students = 
     7 * ?