The copy everything worry

The copy everything worry

Post by Brandon J. » Sat, 23 Oct 2004 02:46:12



I suppose that is another way of viewing it. "Hosed either way." For less
difficult problems, I can see cases where FP may be a win, if the state is
restrictable without too much *** and discipline.

--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA

"We live in a world of very bright people building
crappy software with total shit for tools and process."
- Ed McKenzie
 
 
 

The copy everything worry

Post by Philippa C » Sat, 23 Oct 2004 03:09:59


That's pretty much the case from where I'm sitting. With lazy evaluation
you have to beware the infamous space leak, but that's things like storing
(1+(1+(1+(1+1)))) instead of 5 - and some data structures take advantage
of this, the behaviour of a map structure in Haskell can be...
interesting...

--
XXXX@XXXXX.COM

 
 
 

The copy everything worry

Post by Jesse Jone » Mon, 25 Oct 2004 02:33:14

In article < XXXX@XXXXX.COM >, Brandon J. Van Every



Pure FP doesn't have to be like that though. For example, in Clean you
can mutate state if you have a unique reference to it (so that your
side effects aren't visible elsewhere).

This seems like a very natural model to me that addresses performance
concerns like updating large data structures, but it's expression in
Clean is rather complex.

-- Jesse