On Mon, 24 Nov 2003, Adrian Ettlinger wrote:
> I see Shlomi beat me to the punch with an excellent discussion of solver
> design strategy. Thanks, Shlomi. (As I said, his understanding of this
> stuff is much more sophisticated than mine).
Heh heh, thanks. Just note that while I can label myself a professional
programmer, I do not aim to call myself an expert on Games Artificial
Intelligence, much less computer science in general. I am familiar with
most of the basic concepts of CS, and most of the new trends, but I have
not studied this thing deeply. I am capable of designing my own algorithms
and understanding some theoeretical papers, but otherwise did not
familiarize myself with a lot of the theory.
Freecell Solver is mainly the product of a lot of thought, and some
experimenting and trial-by-error, with very little research. It is very
plausible that I re-invented the wheel times and again there. But, of
course, I many times duplicated or was inspired by concepts I knew from
elsewhere.
I do know that Artificial Intelligence involves a lot of heuristics and
very few if any algorithms that are proven to work all the time. In fact,
games that can be solved with such algorithms usually don't belong in AI.
Thus, us hobbyists may not be substantially less knowledgable than the
fore-front researchers in this field. Still, anyone can benefit from
learning techniques that previous researchers and implementors have
done.
Another misconception about AI is that it requires LISP or that LISP is
useful solely in AI. In fact, AI can be implemented in every
Turing-complete language (albeit naturally, in some languages it would be
easier than others), and LISP is a very nice and powerful language that is
useful for many other things besides AI. The reasons we did not see LISP
as fully deployed for real world applications as much as we saw such
languages as C, C++, Java, Perl, Python, Tcl, etc. are given here:
http://www.paulgraham.com/popular.html
Naturally, a lot of AI programs are written in LISP. But not all. And
Freecell solvers are one domain in which writing it in LISP would make
much less sense than writing it in C or C++.
Regards,
Shlomi Fish
> ------------------Adrian
>
>
>
> To unsubscribe from this group, send an email to:
> fc-solve-discuss-unsubscribe_at_yahoogroups.com
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
----------------------------------------------------------------------
Shlomi Fish shlomif_at_vipe.technion.ac.il
Home Page:
http://t2.technion.ac.il/~shlomif/
An apple a day will keep a doctor away. Two apples a day will keep two
doctors away.
Falk Fish
Received on Mon Nov 24 2003 - 05:05:26 IST