A85: Re: The possible

To: <assembly-85@lists.ticalc.org>
Subject: A85: Re: The possible
From: "Dines Justesen" <dines@resnet.gatech.edu>
Date: Fri, 6 Nov 1998 00:43:43 -0500
Delivered-To: assembly-85-outgoing@towerguard.unix.edu.sollentuna.se
Reply-To: assembly-85@lists.ticalc.org

>1. Damn near everything conceivable is possible, with the right
>equipment and knowledge.

Almost anything you can think of is possible, but NOT everything ! The
program you suggested is impossible to make.

>2.The problem with negative responses to ideas is because of people like
>Dines Justesen who brag about how long they've programmed the ti-8x, and
>have an ego they can't have bruised just can't, or aren't willing to,
>admit that other people who are fresh coul think of doing something they
>haven't done, or think of a way to things different.

First of all you did NOT think of anything different. The same idea has been
mantion before on A85, LZ/LZA and A82 ! Of course someone can think of new
ways to do stuff, I have never said otherwise.

>3. I was wrong. The idea I had was wrong, it'd been a while since i did
>anything like this so i was rusty, but i'm not now. My idea was wrong,
>but i hate asshole who call me stupid. So i'm going to write the
>compression program just to prove to Dines Justesen that he's not as
>smart as he thinks. any help would be appreciated and given credit for,
>just email me, not the list, and we'll make Dines Justesen and everyone
>like him swallow their words.


I did not say you where stupid, I pointed out that what you suggested could
not be done, which you same to agree on now. I also wrote that someone else
had tried it before, and they found out that it was not a good idea. I never
said that it was impossible to write a compression program for the ti8x, as
matter of fact I am sure that one can be written. The question is will it
give the average ti8x user more available memory ? My guess is that using a
simple form on arithmetic coding you could probably save 10 % of the memory
taken up by programs. The problem is that a program like that would probably
take up ~4k. So if a user has 10ks of program he might save 1 or 2k, but if
the compression program is bigger than 2k it wont free up any memory. To
make someone use a program like that you would probably need to free a
couple of k's, which means you would have to compress all programs to half
their original size.

Since I have never tried to make a compression program I can of course not
be sure that these numbers are correct, but as far as I can see it will be
hard to save a lot of mem using a program like this.

Another problem is that you would have to have some free mem on the calc to
use a program like this. Not being able to use all your mem for programs
means that a compression program would have to compress programs even more
to save space.


>And lastly, hey Dines Justesen, have you ever tried to zip a zipped
>file? I can tell you now it doesn't work, so niether does your little
>formula, so you're already wrong! That argument it over!


The formula I used was the chain rule for entropy, which was discovered 50
years ago. If you pick up any book on information theory or data compression
you will see that it used there too. The formular does work.

If you take a zip-file and zip-it you will normally not get any compresion
(or atleast very litlle), which is exactly what the formula says ! The whole
point of the formula was to prove that ziping a zipfile wont work, so I do
not understand what your problem with the formular is.

For one reason or another you seem to have missed the whole point of my
email ! I replied to your email to answer your questions, NOT to tell you
that you are stupid. I stated that the program you wanted to make was
impossible. The reason for this was to make you aware that the routine you
had thought of would not work, which you same to have found out too. Besides
that I included the information on what happen on the list earlier, because
i thought it migt interest you to know that other people had tried it
before.

After receiving this email i re-read all the emails I have send to the list
about this subject, and I am still not able to see why you are so angry at
me. I have not found any place in the email which I can see gives you any
reason to become so angry, but what ever I said I am sorry.

If you do decide to make a compression program for the ti8x I have couple of
suggestion on which algorithm might be goos to use. If you are interested
email me or the list.

Dines