Re: another take on hackers and painters

[John Clements <address@hidden>]

On Wednesday, May 21, 2003, at 10:04  AM, Dan Sugalski wrote:

> At 5:00 PM -0400 5/20/03, John Clements wrote:
>> On Tuesday, May 20, 2003, at 04:15  PM, mcguire@cs.utexas.edu wrote:
>>
>>>> Clerical programmers [...] will not know what to do with the freedom
>>>> given by things such as dynamic typing.
>>>
>>> Ok, so I'm none too bright.  What exact freedoms do you get with 
>>> dynamic
>>> typing?  It seems like a type problem is going to result in a broken
>>> program either way, so you still have to think about types and making
>>> sure the right collection of stuff gets put in the right place.  The
>>> only real advantages I can see are:
>>
>> ...
>>
>> Depends what you mean by "a type problem."  If I'm adding a number 
>> and a string, that's definitely a "type problem,"
>
> Is it? What's wrong with 1 + "3" producing 4?

Personal taste:  I would much rather get the (static or runtime) check, 
along with the ability to perform a dynamic coercion (string->number) 
when I really want to take a string to a number.

Obviously, there's a broad spectrum here.  I get the sense that 
languages like Perl have a DWIM attitude, where any operation that can 
be completed safely should be, whereas statically typed languages ML & 
Haskell take the point of view that ensuring that every program 
location corresponds to a single type is a crucial part of writing a 
program.

As a Schemer (and of course I speak for no one but myself), I believe 
in the utility of static type systems, but I also really like 
heterogeneous lists, union types, object systems, non-let-bound 
polymorphism, eval, and a host of other things that are difficult in 
existing statically typed languages.  Whenever I can get a check that 
will rule out bad programs without preventing me from writing the 
programs I want, though, I'm all over it.  I would place this coercion 
firmly in that camp.

john clements