On Mar 15, 2005, at 3:12 PM, Yoann Padioleau wrote:

>> To which, I'd assume the majority response would be, "So?"
>
> So some of his arguments are right. You make object "So?" but
> we could continue a long moment that way.

Perhaps, perhaps not.  His point seems to be that programming in a 
"functional style"[1] is inherently slower than an imperative style 
because a list or a map have different performance characteristics than 
do arrays.  To which the only response is along the lines of "True.  
They are different data structures, and they behave differently -- 
sometimes worse, sometimes better."  But the point is that needlessly 
restricting yourself to such a style seems like such an odd thing to do 
that I have a hard time caring about the truth of the assertion.  The 
truth of a statement is orthogonal to its silliness.

> Well not trying to define stuff is better ?

That's not really the intent of the quote (or at least I don't think it 
is).  I read it as saying that those who insist that e.g. "functional 
programming" *cannot* include the notion of mutable data structures, or 
that it *cannot* be OO if it doesn't offer encapsulation or classes, 
aren't really bringing anything useful to the table.  You can argue 
'till you're blue in the face whether or not mutable arrays or strings 
have any place in a "functional" language, but when you're done, have 
you really accomplished anything?

> "dollars to donuts" ?
> I am an american newbie so I have no idea of what it means :)

Sorry.  It's shorthand for "I'll wager my X dollars against your X 
donuts that I am correct," and is a way of expressing confidence in 
your position.  It used to mean a lot more when you could get a dozen 
donuts for a dollar...

[1] Where functional style is restricted to, among other things, no 
mutable data structures.

William D. Neumann

"You've got Rita Marlowe in the palm of your hand."
"Palm of my hand?  You haven't seen Rita Marlowe..."

		-- Will Success Spoil Rock Hunter?