From: Martin Espinoza <mespinoz@tivoli.com>
To: ml@qoole.com
Reply-To: ml@qoole.com
Date: Fri, 02 Jan 1998 17:15:10 +0000
Subject: Re: QOOLE IDEA

This is a multi-part message in MIME format.
--------------5848AA48FE7FFDE7913CAF66
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

bciuppa wrote:
> Second. I care what it's called because I'm not stupid.

Oh yes. I'm stupid. If we're going to degenerate into name-calling,
please put some thought behind your attacks.

> I don't try to use terminology when it's incorrect.

"I try not to use terminology when it is incorrect."

> In a binary search you would have the ability to determine which half to throw away
> because your set would be sorted in some fashion to make that distinction. Your set is
> not.

That's not true, actually.

Let's say I have a level (I'll have to represent two-dimensionally), I
divide it once...
( the x being player start and the X being where I have a solid block)

+--------+       +--------+
|   x    |       |   x    | Okay, I find out that the leak is in the
solid block.
|        | ===>  |________| 
|        |       |XXXXXXXX|
|        |       |XXXXXXXX|
+--------+       +--------+ So now I make the following change:

+--------+       +--------+
|   x    |       |   x    | And now, if I still have no leak, it's in
the
|________| ===>  |________| area CURRENTLY blocked off. If I have a
leak,
|XXXXXXXX|       |XXXX    | I know where THAT is, too.
|XXXXXXXX|       |XXXX    |
+--------+       +--------+ 

This solution becomes more complex with each additional dimension; That
is,
if I am working in 3d space, I have to worry about height, etc. That
doesn't hurt
this much, as I work with the block covering the entire height of the
level until
I know where it is in X+Z space. THEN I worry about the Y.

> In fact if your leak should go across the boundaries of your halves you will
> never be able to throw away the correct half. And if you have multiple leaks that go
> across halves you will never be able to find the leak let alone discard the correct
> half.

It doesn't have to be a half. You can nudge the solid section one way or
another.

> So in close. Your technique just happens to be working for you. In most cases doing
> what you propose will not allow you to find the leak with any accuracy. No matter how
> fast your machine is.

It's let me find every leak I've created so far.

> I am now done with this conversation.
> You bring no facts to the argument. Only -- "It worked for me". And a -- "I don't know
> why it worked either."

Uh, as above, I know EXACTLY how and why it works.

> It's true what they say.
> A little knowledge is a dangerous thing.

A foolish man looks into a mirror and sees only an image.
A wise man looks into a mirror and sees himself.

---------------------- mespinoz@tivoli.com -------------------------
        You are what you do when it counts. --The Masao
--------------5848AA48FE7FFDE7913CAF66
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Martin Espinoza
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Martin Espinoza
n:              Espinoza;Martin
org:            Tivoli Systems
email;internet: mespinoz@tivoli.com
title:          Associate Engineer
x-mozilla-cpt:  ;0
x-mozilla-html: TRUE
version:        2.1
end:            vcard


--------------5848AA48FE7FFDE7913CAF66--