These are all good ideas in theory. I'm glad we can have such great ideas floating around. I personally can't wait to have 50K of memory on my TI-85 so I can start putting huge programs on it. Except that you need to figure in an average of 1.7:1 compression if you want any decent speed on a z80, then the fact that 2-3K is taken up by the shell, most of which I assume can't be compressed since you can't compress the algorithm doing the decompression. You're at 25K of usable memory, 1.7*25 is still darn close to 50K, then you have to figure in the fact that the programs will have to be decompressed into memory before they're run (unless you can write a dynamic decompression utility), so if your programs are all 4K, you lose 7.8K of compressed potential. 42.5 - 6.8 = 35.7K. TI-BASIC programs, REALs, etc, can't be compressed, so you need uncompressed memory for those. I collect 2-3K of that stuff without even trying. Now we're down to 30K compressed. And no program can exceed 4K. If it does, we lose even more RAM. Seems to me that with compression, we might only have an extra 10K, not another 28K magically appearing. Is it worth it?

I think it is a great idea.

Another use would be for engineers workingon ideas.

They could encrypt them.

I think that an encryption program would be great! I have tried to put RC-6 encryption onto my calc, but I don't have the assembly skills to do it, and, an even biger problem, (At least for the 83+) is that the z80 is a 8 bit processor. RC-6 is designed for 32 bit processors. If someone could wright this prog, it would be great. I don't care about compression, just the fact that it would take you average comp months to figure it out. It would be nice though if it was public key.

For String Assembly programs:

Compress your data(with some standardized utility) and have the de-compression utility decompress and load the correct characters at one time, re-compress/trash this de-compressed portion of the string. Then move on to the next character.
you would feasably never have to uncompress the entire string leaving only the current command 'active' in memory.

While compressed, the first characters should be the compression ratio (or not if everyone sticks to a standard)

Problem, You may have everything to run the game, but you also need to download the de-compression utility!

Sincerley,
AGENT-TI
ß^)

I know a decent amount about this type of thing. I think it would be nice.
Have you ever heard of PGP? It's like this, except it would take the FBI, like, 1*10^100 years to crack.

Ryan

Alright, I like the crypto idea!
But why use such a complcated algorithm for just sending messages in a class room?
If the purpose of this program is to send messages to your fellow classmates, why not use the old german favorit, eignma!

I know, eveyone is saying how you can get a program that can break a message done in eignma in 8-min (IF you have both the cyphertext and the plain text, and IF you know that the eignma box is using standared rotors), but what are you really tring to encrypt here?

The algorithm uses only the 26 letters of the alphabet, so you don't have to worry about useing hex coding to send people messages, you can just use paper!

And the key for it is just 6 letters and 3 numbers! (i.e ANE-EWY-341 (numbers 1-5)), so you can say, "Turn to page 153, first 3 letters of the first paragraph, and the last 3 leters of the last paragraph" as your key.

Heck, you can set eveyone up to turn to a curtan page evey day for the key for that day!

If you notice, I am trying to make this crypt thing sound "fun". Like it or not, lots of these crypt algoritms are rather complcated and copyrighted. Not to mention the export laws of the united states, or the fact that implmenting a public key algorithm requires lots of work (random number generator, how to do key exchange, etc.)

Egnima, its easy and you can write it in BASIC:)

(Just wondering, why use RSA? Isn't Diffman-Hofman public now? Why not use DES? (or better yet 3DES) as that is public as well, might have to bump the key down to 40-bit so it can get passed the export laws but hey, I don't think your teacher will go THAT far with trying to decrypt it:))

Alright, I like the crypto idea!
But why use such a complcated algorithm for just sending messages in a class room?
If the purpose of this program is to send messages to your fellow classmates, why not use the old german favorit, eignma!

I know, eveyone is saying how you can get a program that can break a message done in eignma in 8-min (IF you have both the cyphertext and the plain text, and IF you know that the eignma box is using standared rotors), but what are you really tring to encrypt here?

The algorithm uses only the 26 letters of the alphabet, so you don't have to worry about useing hex coding to send people messages, you can just use paper!

And the key for it is just 6 letters and 3 numbers! (i.e ANE-EWY-341 (numbers 1-5)), so you can say, "Turn to page 153, first 3 letters of the first paragraph, and the last 3 leters of the last paragraph" as your key.

Heck, you can set eveyone up to turn to a curtan page evey day for the key for that day!

If you notice, I am trying to make this crypt thing sound "fun". Like it or not, lots of these crypt algoritms are rather complcated and copyrighted. Not to mention the export laws of the united states, or the fact that implmenting a public key algorithm requires lots of work (random number generator, how to do key exchange, etc.)

Egnima, its easy and you can write it in BASIC:)

(Just wondering, why use RSA? Isn't Diffman-Hofman public now? Why not use DES? (or better yet 3DES) as that is public as well, might have to bump the key down to 40-bit so it can get passed the export laws but hey, I don't think your teacher will go THAT far with trying to decrypt it:))

Alright, I like the crypto idea!
But why use such a complcated algorithm for just sending messages in a class room?
If the purpose of this program is to send messages to your fellow classmates, why not use the old german favorit, eignma!

I know, eveyone is saying how you can get a program that can break a message done in eignma in 8-min (IF you have both the cyphertext and the plain text, and IF you know that the eignma box is using standared rotors), but what are you really tring to encrypt here?

The algorithm uses only the 26 letters of the alphabet, so you don't have to worry about useing hex coding to send people messages, you can just use paper!

And the key for it is just 6 letters and 3 numbers! (i.e ANE-EWY-341 (numbers 1-5)), so you can say, "Turn to page 153, first 3 letters of the first paragraph, and the last 3 leters of the last paragraph" as your key.

Heck, you can set eveyone up to turn to a curtan page evey day for the key for that day!

If you notice, I am trying to make this crypt thing sound "fun". Like it or not, lots of these crypt algoritms are rather complcated and copyrighted. Not to mention the export laws of the united states, or the fact that implmenting a public key algorithm requires lots of work (random number generator, how to do key exchange, etc.)

Egnima, its easy and you can write it in BASIC:)

(Just wondering, why use RSA? Isn't Diffman-Hofman public now? Why not use DES? (or better yet 3DES) as that is public as well, might have to bump the key down to 40-bit so it can get passed the export laws but hey, I don't think your teacher will go THAT far with trying to decrypt it:))

Calculator cryptography would also be a good Idea for IR link chats. That way nobody else can intercept what you say.

I'm finishing up a C++ program that I think might interest you. The purpose of the program is to convert ascii text into a Ti-Basic file that could be linked and run on an 82/83 (I only have an 82)
Anyhow, the purpose of the program is so that you can read text files (maybe a chapter of an etext) on your calc in a nice format, and display unsupported ascii characters (@ # $ % & ; " \ | _ ` and ~) using drawing (Pxl-On, Circle, etc) functions :) . You can scroll up and down in the file, which is displayed on the graphing screen, and can be 37 screens long (After that another file will be made, starting where the other left off). The program does word wrapping, and allows the user to pick whether line returns will be discarded or not.
Now, how this is related to your article: Before I read your article I was talking with a friend about the program I'm making. He thought that if an assembly program could be made, not only would it be more efficient (obviously), but it could also *possibly* use compression. He thought that
the WinZip codec could be used to encrypt the file on the PC and then linked to the calc. The asm prog on the calc would have to read from this "zipped" file, but only decompress one page at a time. Although I don't know assembly, I thought this was a great idea, but probably hard to implement, although plausible. However, I'm sticking to my "C++, Ti-Basic generating" program: it will probably allow around 90-100 "screens" of text. Not bad, eh? Tell me what you think. (Also, if you're interested in this program..)

Honestly, is passing secret messages in class that much fun that it would be worth spending large amounts of time writing a program for it?
-About the GPC
Compressing and decompressing anything that you want to run seems like an extreme pain. Personally, I think it would be better to come up with a program (probably asm) that would make programs more efficient, especially longer ones.
No, I am not trying to whine about the lack thereof, nor am I saying that I do not have enough time. I just think it would be nice to have one.

TextTo82 has been released. Download it under "Windows Utilities"

A note, I have programmed a cryptography program for the TI-89 which uses the Hill cipher. It's not 384 bit encryption or anything, but you can make it fairly complex and for practical purposes unbreakable, since no high schooler or college person would care enough to do it. Any bit of encryption would be (most likely) enough to deter someone from reading the message (provided it is not a simple substitution based cipher). My program can handle messages up to 48 characters, although it can be expanded if needed. Email me if you'd like to see it.
Steve

I wrote an encryption program once, but I..uhh..lost the code in a fire (ok, so it was an ASM crash...) Anyway, the way it decrypted was I had the letters A-Z stored in Str1. There was an encrypted message, that the user would type in. That would be converted into it's appropriate number (A=1, B=2...) After that, a math function would be run on it, and one would get a different number. The new number was converted to the approptiate letter. It worked extremely well, though I never got to see it put to practical use...

Programs like that are easy to crack.

I wrote a program like that. It converted the pwd into A1B2C3 format, the text the same, and xored them. But it was VERY slow. Anyone here actually use PGP?

Ryan

I wrote an encryption program once, but I..uhh..lost the code in a fire (ok, so it was an ASM crash...) Anyway, the way it decrypted was I had the letters A-Z stored in Str1. There was an encrypted message, that the user would type in. That would be converted into it's appropriate number (A=1, B=2...) After that, a math function would be run on it, and one would get a different number. The new number was converted to the approptiate letter. It worked extremely well, though I never got to see it put to practical use...

I wrote an encryption program once, but I..uhh..lost the code in a fire (ok, so it was an ASM crash...) Anyway, the way it decrypted was I had the letters A-Z stored in Str1. There was an encrypted message, that the user would type in. That would be converted into it's appropriate number (A=1, B=2...) After that, a math function would be run on it, and one would get a different number. The new number was converted to the approptiate letter. It worked extremely well, though I never got to see it put to practical use...

I wrote an encryption program once, but I..uhh..lost the code in a fire (ok, so it was an ASM crash...) Anyway, the way it decrypted was I had the letters A-Z stored in Str1. There was an encrypted message, that the user would type in. That would be converted into it's appropriate number (A=1, B=2...) After that, a math function would be run on it, and one would get a different number. The new number was converted to the approptiate letter. It worked extremely well, though I never got to see it put to practical use...

I wrote an encryption program once, but I..uhh..lost the code in a fire (ok, so it was an ASM crash...) Anyway, the way it decrypted was I had the letters A-Z stored in Str1. There was an encrypted message, that the user would type in. That would be converted into it's appropriate number (A=1, B=2...) After that, a math function would be run on it, and one would get a different number. The new number was converted to the approptiate letter. It worked extremely well, though I never got to see it put to practical use...

Hey I have a decent idea, why not convert string text
to a string of binary then put it in a picture
(ie pixel on pixel off)? Then you could implement a
wide array of picture encryption methods.The data would be very secure (in relation to "note passing"),
but the only problem would be dec/enc proccess time.
Another possiblity could be to Xor the binary string,
or you could use some sort of matrix encryption with the plain text and do row and collumn swapping.
Reading the matrix would be quick and efficient in regaurds to picture encryption, but matrices do take up a lot of space..=(
if you were to just do simple string ecryption then i would recomend the Nexor method.
It uses the sine wave to encrypt text using 2 different keys, it is fast and simple but strong enough that someone could not brute force it through the rand function.

any questions?
pyxle@sdf.lonestar.org