You are not logged in.
Hello,
A long time ago I have created my own symbols which I used to replace normal characters when writing. My intention was to write down information that no one else could read, in order to keep it private.
I need advice on how I could create a custom set of characters on my system, so that afterwards I could get the system to use my symbols rather than normal Latin or Cyrillic characters.
Getting the system to use the characters would not be a problem as I could probably write a script to do it, but I don't know how sets of characters work on a system (I believe that character sets are a very different thing from fonts). I have done some research but it only got me even more confused.
Thank you very much, any help is appreciated.
Offline
This could be achieved in many ways; but honestly, the easiest way of doing this would be to just encrypt whatever files you wish to keep unreadable.
All the best,
-HG
Offline
If you "encode" your secrets in this way, you will have created a substitution code.....these are the most basic of codes and generally take about 5 mins to crack, using nothing more than a pen & paper. We used to make them up & break them as children.
Not trying to rain on your parade here, but if you really want to keep a secret, there are lots of free encryption application available.
Last edited by flipper T (2013-07-15 21:19:37)
If I'm curt with you it's because time is a factor. I think fast, I talk fast and I need you guys to act fast if you wanna get out of this. So, pretty please... with sugar on top. Clean the [censored] car. -The Wolf
Offline
If you just substituted Latin or Cyrillic characters for your own you hid the information only from noobs, but a professional should be able to decode it pretty easily, especially you're going to use your own character set to write a lot of stuff thus providing a large corpus.
$ rot13
png
catAs you can see, simply substituting one character for another isn't a very good way of hiding information (read 'man rot13' if you don't know what rot13 is).
Offline
Not a GNU/Linux discussion, moving to Apps & DEs...
Offline
If you just substituted Latin or Cyrillic characters for your own you hid the information only from noobs, but anyone with even a basic knowledge of substitution ciphers and character encoding should be able to decode it pretty easily, especially you're going to use your own character set to write a lot of stuff thus providing a large corpus.
ftfw
I also recommend using proper encryption if you want to hide data. A substitution cipher might prevent someone who casually walks by your screen from understanding what is on it, but it might just make the person curious enough to take a picture of the screen and crack the code when he has a few minutes free.
My Arch Linux Stuff • Forum Etiquette • Community Ethos - Arch is not for everyone
Offline
I think this is far worse than what others are getting at. I'm not sure which of the following two possibilities you are getting at, but they'd both have severe problems:
1) Do you want to work in your 'code'? If so you might make a "rot 13 font" for a simple example, where each character is displayed as the 13 character rotation. Then when a text file was opened on your system, or if they looked over your shoulder as you typed, they would not be able to make heads or tales of what they saw (unless they knew your code). This sounds reasonable enough, except that all that would be needed to "decode" the information would be to use a normal font. This would be on par with taking your text document (in that other OS) and changing the font to "Wingdings". Sure, it looks unreadable ... until one switches the font back to arial, or whatever.
2) Perhaps you mean to write what would be legible to you, so when you want to type "and" you'd actually be saving the rot-13'ed character codes. This is what could be replaced by using something like rot13 after writing the document, but either way seems fine ... seems. The problem here, is if you really do "rot-13" your whole system, so that ever key press is interpreted as 13 characters off, then you'll have a real problem typing any commands! Youd have to remember to type `zf` instead of `ls` or `dmoxmy` instead of pacman (I just made those up, they're not actually rot13s, unless by chance). Are you going to learn your "code" well enough to translate on the fly so your shell can make sense of the characters you type?
EDIT: upon rereading my own point 2 it may not be clear. If every character you type is displayed normally, but interpreted (and thus saved in your file you are creating) as a coded character, and if you apply this setting systemwide, then when you type on a command prompt, your typing of "pacman" will actually be read by the shell as something very different. To get the shell to 'see' "pacman" you'd have to feed it the inversely-coded version of that word so that after encoding, it corresponds to the original.
Further, while it has been implied in all the previous posts, security by obscurity is not much security at all. It may be fun, but as you are using arch linux, there are much more sound security measures at your fingertips - you may want to learn about those instead.
EDIT2: I missed in your OP, that you acknowledge that character sets are different from fonts. I assumed this was all about fonts (sorry). I still think much of this still applies though.
Last edited by Trilby (2013-07-15 21:49:41)
"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman
Offline
Thank you all for the replies.
I missed my point, I want to replace normal characters with the symbols so I can become fluent. I write down notes (on paper) lots of times, but I don't read them as often as I would want to.
The symbols involve more than substitution of characters, some of them represent sounds or combined syllables, it is not as good as actual encryption, obviously, but it's fast(er) for me to read.
I use TrueCrypt containers to hold private information.
My plan was the following:
1. Create characters
2. Create / extend font to supports new characters
3. Use self-made program to take system-wide language files in Latin characters and replace it with new characters.
This way, most of system uses my characters and I will have to read messages and notifications that use my characters, and I will become fluent.
Note: Terminal would not use new characters (maybe only some output from application), it would slow down my work.
Maybe I did not search on Internet long enough, I know Windows has "eudcedit" program to create characters in "charmap". I was hoping for Linux to have similar program. I don't know how characters are handled in general, otherwise I would have tried writing program myself.
Offline
It sounds like you are re-inventing the wheel...
If I'm curt with you it's because time is a factor. I think fast, I talk fast and I need you guys to act fast if you wanna get out of this. So, pretty please... with sugar on top. Clean the [censored] car. -The Wolf
Offline
It sounds like you are re-inventing the wheel...
http://en.wikipedia.org/wiki/Shorthand
If you "encode" your secrets in this way, you will have created a substitution code.....these are the most basic of codes and generally take about 5 mins to crack, using nothing more than a pen & paper.
It sounds like you are contradicting yourself. I know shorthand, and so do others.
My problem relates to using my symbols on my computers, not what symbols to pick.
Offline
No, you clarified your question & I clarified my answer.
Anyhows, I wish you all the best in achieving your aims & apologize for not being able to provide a solution ![]()
If I'm curt with you it's because time is a factor. I think fast, I talk fast and I need you guys to act fast if you wanna get out of this. So, pretty please... with sugar on top. Clean the [censored] car. -The Wolf
Offline
I understand what flipper T is getting at, he's not contradicted himself - IMO - merely adding to his initial statement, it's a basic sub-cipher with a shorthand element. You could make your own font set, which would handle the sub-cipher part.
You may be better off chatting with developers behind stenography tools like Plover
Last edited by opt1mus (2013-07-19 17:09:55)
Offline
Thank you all for the support provided.
I will see what I can do about it.
The developers behind Plover seem the best approach so far, thank you opt1mus.
Offline