Static v. dynamic typing


#1

Ruby is SLOW! Even compared to Python.

i’ve never had a problem with this and am routinely process files
ranging
from
40mb to 3gb - one only has to use c extensions.

That’s true. You can always fall back on an extension. The thing is, I
actually prefer Python for C++ proto-typing. It maps so well to C++,
STL,
and Boost.

why do you see dynamic typing as a contributor to slowdown?

I’m no expert. There were interesting remarks on this point from the
author
of Boo. This is from the Boo FAQ:
Performance: since it is statically typed, can I expect a performance
equal
or close to c# or vb.net?

Yes.

At the Digital Mars D language website (D is the way C++ should be)
there is
a comment that a compiler writer should have had more input into the
design
of C++. D is blazingly fast, but significantly easier to read and to
code
than C++. It also includes integrated testing and design-by-contract.

So ‘agile’ does not have to mean ‘slow’.

Boo and Ocaml are examples of fast, agile languages, and if more companies
would adopt them, I’d switch in a heartbeat.

ocaml is very good - but it’s a massive paradigm shift for most people’s
thought train… most real world problems are imperitive.

Ruby is quite a paradigm shift itself! Continuations are mind-blowing.
I love the whole concept.

What I’m saying is that the author of Boo, by realizing the difference
between dynamic-typing and type-inferencing, achieved the best of both
worlds.


#2

On Fri, 3 Mar 2006 08:53:06 +0900, Christopher Dunn wrote:

why do you see dynamic typing as a contributor to slowdown?

Lisp is dynamic typed, but there are Lisp compilers with C-like
performances.