My pre-1.0 wishlist

[Hal Murray]

esr at thyrsus.com said:
> The need for that kind of manual intervention seems like a crash landing to
> me. It suggesrs that there's no way to package up this hypothetical test
> suite so that users can run it after building the software. 

I wasn't trying to solve that problem.

I started by assuming we had a stable test bed.  That obviously doesn't hold 
if Joe Random pulls the sources, builds a system, and now wants to test it.

We might be able to come up with a description of what various graphs should 
look like.  I suspect the big bad internet is too wild to make that practical 
at the yes-no level.  We might be able to collect a bunch of examples so 
people could do pattern matching.  I'm thinking of a paragraph or two for 
each example to explain what's going on and what it means.


-- 
These are my opinions.  I hate spam.