Slashdot Mirror
Morse Code Enters The 21st Century
N8TWJ writes "The International Telecommunications Union has decided to bring
Morse Code into the 21st century by adding the loved (or hated?) - at (@) symbol. Lets hope the spammers don't start sniffing 20 meters for da-dit-da-dit-dah-dit..." According to the article, Paul Rinaldo, chief technical officer for the American Radio Relay League, says: "It's a pretty big deal... there certainly hasn't been any change [in Morse Code characters] since before World War II."
Obviously, morse code operators have had ways of getting around using 'at' signs for almost two hundred years, but it's a cool addition, and is as big a deal as they make out.
:-) If any Morse operators here have some good examples, do post!
I find morse code interesting because so many old timers complain about how the Internet makes 'ppl tlk lk this' and say 'nce 2 c u l8r', but that sort of abbreviation has been used in Morse Code for decades! The typical banter you see on IRC or in SMS cellphone text messages is pretty tame compared to what those operators could pump out on the wires
--_- -- -_- --- _ A -__- ___ --- _ A _- ___ A -__ -_ -- _ A _-- -_ __ _-
) . Let's hope Slashdot's filter lets me post this.
(Check this link to translate: http://www.translatum.gr/converter/morse-code.htm
Small potatoes make the steak look bigger.
According to the article:
:)
The new sign, which will be known as a "commat," consists of the signals for "A" (dot-dash) and "C" (dash-dot-dash-dot), with no space between them.
This means that the @ is: dit-da-da-dit-da-dit so I don't mind the spammers listning for da-dit-da-dit-da-dit
Jeroen
Secure messaging: http://quickmsg.vreeken.net/
Wouldn't spelling as a-t it out be shorter/faster?
at = dit dah, dah
@ = dit dah dah dit dah dit
Or does an inter-symbol pause really take as along as 3 symbols?
While it is cool and all that, forget sending your email address as morse code when your ship is in trouble because nobody is listening!.
Nuff said...
Morse code is an early variable-length data compression stanard (similar to Huffman codes or Shannon-Fano codes). By representing common letters with shorter codes ("E = "," and "T" = "-") and rare letters with longer codes ("Z" = "--.." an "Q" = "--.-"), Morse code manages to encode the 26 letters of the alphabet in 4 bits maximum and much less than 4 bits per letter on average.
Although Morse did use letter frequencies in constructing his code, it is not a truely optimized code, from what I can tell. Numbers are encoded with a cumbersome 5 bits per digit. Also, the transmission time of messages might be further reduced with minor rearrangements of the code to use more dots (short transmission time) in more frequent letters and more dashes (long transmission time) in the less frequent letters.
Two wrongs don't make a right, but three lefts do.
Morse codes can be quite useful, even today.
There is a Linux kernel patch floating around (IIRC it was in -ac, don't know about mainstream) that adds the output of kernel panics via keyboard LEDs. Nice for early oopses.
The main character in Neal Stephenson's Cryptonomicon uses his keyboards LEDs to read a set of documents about cryptography. He uses the LEDs to spell out the documents in Morse code and writes some code using his space key as a morse tapper.
;-)
However, looking at the Morse chart he would have a problem writing code and reading mathematical notations with the limitations of the Morse alphabet.
I'd never thought about that before! You're owned Stephenson
A little planning goes a long way...
Perhaps the most famous Morse communication is the international distress signal S-O-S. It consists of three dots, three dashes, and three more dots. Surely these days its the international Nokia sound of an incoming S-M-S three small beeps, two long beeps, three small beeps.
There are a number of morse code symbols that don't correspond to graphical letters, digits or interpunctuation; you may say that they are similar to ASCII control characters in denoting things such as "end of message".
Problem is, when you listen to and take down morse code by hand, you need an easy way to indicate those control symbols too. When I had some morse training in the 1970's (voluntary after-school classes), we used "+" to indicate "end of message" (.-.-.) and (surprise) "@" for "end of contact" (...-.-)...
I don't know whether "@" was in common use for that purpose or if it was just my teacher's idea. To me, @ thus meant "end of communication, time to do something else" long before I learned about spam!
I have to ask: with so many thousands of glyphs in Chinese, what did they do to utilize telegraph lines when that was the only means for fast long distance communication?
I'm assuming they didn't create a different sequence for each one....
"Provided by the management for your protection."
Some ham operators wouldn't mind more changes to spice up the language. While Morse code has a period, a question mark, and even a semicolon, it offers no simple way to articulate excitement. "I was hoping they'd add a character for the exclamation point," said Yocanovich....
Which leads me to wonder: how many emoticons can you express in Morse Code? Or do they have their own equivalent already?
> c a t a t a t n t . o r g
c ha nge. Especially in Morse. You were able to read that mashed sentence with only a little extra trouble. Try that in Morse code and it's a total parsing disaster due to the binary symbols and the variable-length sequences that denote letters and symbols.
;-)
>
> where is the 'at'==@ in that letter sequence?
Simple, I and my other fellow HF CW friends have been doing it for years. Per your example:
"c a t a t a n t . o r g"
Itsnotasifyoucruncheverythingtogetherduringanex
Newbies sending Morse can be EXTREMELY difficult to understand because of poor spacing.
Proper morse code has three levels of pauses:
0) Between dits/dahs
1) Longer ones between words.
2) Really long ones while us Old Timers try to remember just what the hell we were talking about.
...let Morse Code die, please! What is next? Six-bit Bardot?
There's the Q-codes that have already been mentioned, and prosigns, but for real message word count reduction look no farther than the ARRL numbered radiogram messages.
Were I on vacation in Florida when a hurrican hit, I could send the message back home that "We are all safe, don't worry. There was only slight damage were we are, and we will be coming home soon." as: ARL ONE ARL FOUR ARL TWO
There is some header information needed by the message passing system (NTS), but the base message goes from 21 words to 6.
K
I recently wrote a PHP program that hides morse code messages into other sentences. Guess I'll have to update it now. Check it out here: CodeBreak - Hidden Morse Code
Second: it's inefficient. It's a binary code (dit and da) and yet it's redundant. The Shannon entropy for English is lower than the calculated average message length for Morse. They should be using a Huffman code.
Suggestion: Create a Huffman code from well-known English letter frequencies. When constructing this code, always put the more-common subtree on the left side of the new Huffman node. Then, when translating to das and dits, always use dit for a left-going branch and da for a right-going branch (dit = 0, da = 1). This accounts for the fact that the "da" symbol takes longer to transmit than "dit".
The result won't be Morse code, but we could name it something else... "Huffman-Morsoid," or say, "Horse code."
What's that, a new prosign "TAR"? Perhaps you meant the "Commercial A", or "at sign", rendered as "AC" - "didadadidadit"?
Yeah, anyway, it seemed silly when it was announced back in early January, and still does. "didah dah" ("at") is only a problem with an address like "kitkat@attbi.net", and is"didadadidadit"?
shorter than "didadadidadit". And is that an "AC", as they say, or is it a "WR", or a "PN"? They all sound the same.
Back in the '60s or so, Irv Hoff (who later was instrumental in developing ham packet radio) was a radioteletype ham. 110-bps ASCII was around and in heavy use for landline teletype and those new-fangled minicomputers. But hams were still limited to 60-bps Baudot (5-bit) teletype and obsolete machines. (This was apparently because the FCC didn't have the budget for buying new moniitoring equipment, so it wouldn't legalize the new coding scheme and speed.)
Irv wanted to go faster. Morse wasn't speed-limited and FSK modulation was allowed. So he built a couple copies of a device to convert ASCII to and from morse and send it either at 110 baud or (using tape) some truly hysterical speed.
Of course the FCC heard the high-speed signals that they couldn't translate and came after him. And he said "But they're just morse! Tape them and slow them down and you'll hear it."
So the FCC did. And shortly threw in the towel and legalized 110-baud ASCII.
Which was the whole point of the exercise. B-)
Now I think Irv may have came up with some arbitrary codings for the ASCII symbols that weren't part of Morse during that exercise. If so, and if they were easy for humans to use, it might have made sense to standardize those.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
God help me if I had used "larn," as I almost did to sound even quainter. I would have had to include a link to dictionary.com to show that it was a proper word. In future perhaps I'll include a full etymological study of my posts.
Or I could just ignore the spelling/grammar Nazis. I actually learn from them on occasion, but I think it's best not to let them know that. They certainly don't need any encouragement.
The usage isn't even particularly archaic and is supported by most dictionaries, although not in the examples of usage. It's a bit too "hick" for city folk tastes.
Which is exactly why I used it.
KFG