Re: dynamic vs. static typing

[Matthias Felleisen <address@hidden>]

On Sunday, November 23, 2003, at 11:48 AM, Dave Hudson wrote:

> Is there any reason why such an approach would preclude the use of type inferencing say?

Not too much. Decidability mostly. Hans Boehm, mid 1980s for one system, and some UPenn
people for others and so on. (Yes, same Hans who did the "hostile" (eh "conservative") gc. 

HM type inference for ML is an incredibly sweet spot in the type language landscape. You can omit 
some or all of the type defs for non-recursive algebraic data types and still get everything back. If you
modify a few things here and there, you don't know what you have. 

That's why I am suspicious of CL (nobody mentioned that here, they had this way before Curl) or Curl
or friends like that. I am interested in pragmatics but I want to make sure that they are well founded. That's 
of course my problem. Some here (may) take whatever they think makes life easier, as have millions of 
programmers done before and those very same people will argue to great success (blue screens of death, 
core dumps, millions of dollars blowing up in the sky, other planets, a few lives here or there, ...).  

-- Matthias