Hacker News new | comments | show | ask | jobs | submit login
Another Reason to Learn Morse Code: Kidnapping (hackaday.com)
134 points by wglb 3 months ago | hide | past | web | favorite | 82 comments

The most moving Morse code story I’ve heard is this one from Somalia: A man writes a letter complaining about conditions in a hospital, and the government imprisons him in solitary confinement, with no one to talk with and nothing to read. After eight months, his next door neighbor whispers across: “Learn ABC through the walls....” and he learns an ad-hoc Morse Code through knocks and taps. Then, after two years, the neighbor begins “reading” to him, via the code, the 800 page novel Anna Karenina.


The article doesn't seem to be exactly about using Morse code when kidnapped, unlike the title, but about instances of Morse and other code used in various circumstances today and in the past.

Having said that, to reflect the title itself I don't think Morse code has a very good bang-for-buck ratio for kidnapping.

Absolutely most people never get kidnapped, it's a very rare occasion statistically. Learning Morse code is not enough: you will have to use it repeatedly to keep remembering it, and I don't think you need Morse code very much in general life so you'll need to exercise your skills intentionally just in case a very unlikely event would happen. Further, to benefit from Morse code while being kept hostage you will need some medium through which to transfer knocks or beeps and you need another party who is both listening and can read Morse code.

It might make sense if your lifestyle calls for higher risks to be kidnapped, but even then I think there are wiser ways to spend your minutes trying to avoid being kidnapped in the first place.

I think learning just the SOS distress signal (...---...) has a decent bang-for-buck ratio. And because of its frequent use in films there's usually not even much learning involved.

If you want, just remember that groups of three of anything are a distress signal. It's why smoke detectors beep in groups of threes for example. Groups of three whistles indicate distress. Three flashes of light. Etc.

Although I don't know how true this is, I've tried to keep to this in other interfaces, avoiding three-blink patters on lights when someone is not in trouble.

Just don't panic when the cruise ship you're on signals that it is operating in reverse by sounding the horn three times.

There is cause for concern when hearing seven short and one long blast, though. (General alarm; something has happened. Better head for the muster station to find out what...)

From the recent El Faro sinking post here on HN:

There's a General Alarm signal (above), but no Abandon Ship signal. The captain gives those manually.

(I may be blatantly wrong!)

To the best of my knowledge (not a sailor, but my work has netted me several hundred days on vessels offshore), 'Abandon Ship' is always an explicit, verbal order from the captain (or whoever follows him in authority if he is incapacitated)

The General Alarm typically calls for the crew to proceed (at haste!) to the muster station to be informed of the nature of the emergency and be put to work handling the situation.

Luckily, I've only ever had to abandon ship during a couple of drills.

FYI, general distress flag code is two black flags, one square, one ball. OTOH, flailing your arms in the air is also recognized as general distress.

Just be alert. The world needs more lerts.

And the emergency button in your car is a triangle.

And the sign on the inside of your trunk / boot is a triangle.

In lifeguard training I've gotten, 3 whistles is what you do to start emergency procedures.

Everyone my age learned SOS from a commercial jingle and/or playing Atari.

Older people learned SMS from the Nokia sms tone :)


Woah I had no idea!

And now i am old...

What jingle and what Atari game?

I believe he's talking about the SOS cleaning scrub commercial from 1994.


For those that don't want to click, it's a commercial for a brand of dish cleaner named S.O.S. There are dirty pots and pans that clang out the S O S message in Morse code, with subtitles spelling out the letters as they come out. A screen appears at the end asking "are your dirty pots trying to tell you something?"

It should be noted, that the popular distress signal is "...---..." and not SOS as it is sent as single symbol. Also it is not the only one special morse symbol used for distress (eg. ten dots is also an distress signal with slightly different meaning)

I think its more about wardrobe choices.

Wearing close that easily shred so you can hang small pieces of your shirt on nearby bushes as a trail.

Or unique pieces of jewelry you can drop along the way for the Ranger to find.

It also doesn't hurt to have at least 2 days of laundry in the basket so the blood hounds have something to scent from.

Is there a shop that specializes in clothes for people propense to getting kidnapped?

The gun and knife store. :P

Learning Morse is quite easy, it’s nothing compared to learning a new programming language or a new human language, so I don’t really see the waste of time in this case.

If you want to learn Morse it is a lot easier if you don't try to learn it in alphabetical order. Instead you should learn it in order of the lengths of the signals, which corresponds roughly to the frequency order of the letters in English, i.e. learn the one-beep-long letters first (E, T), then the two-beep-long letters (A, N, I. M), then the three-beep-long letters, etc.

E . T -

I .. A .- N -. M --

S ... U ..- R .-. D -..

H .... B -... L .-.. F ..-. V ...-

W .-- K -.- G --. O ---

Z --.. C -.-. X -..- P .--.

J .--- Y -.-- Q --.-

The key factor is to learn the sound of each letter at speed, rather than counting dots and dashes. If you're to use Morse with any degree of fluency, you need to instinctively recognise the sound "dah-dit-dah-dit" as C, rather than counting "dash, dot, dash, dot" or mentally scanning through T, N, K and then C. It's helpful to think of Morse as a kind of music rather than a coding system.

Most successful learners use the Koch method, which transmits characters at 20wpm equivalent speed but with longer gaps between each character. You start by learning to distinguish two similar letters (A and N or K and R), then adding letters one by one. Learning at slower speeds or using a visual letter chart is actively harmful - it encourages you to count dots and dashes rather than listening to the sound of each letter as a complete rhythmic pattern. The order in which you learn the letters is largely unimportant.


This method disconnects it from your normal language processing. Start by learning simple words so you associate it with your language. You'll be at 20wpm no time.

So much this. Morse code is not a language of symbolic translation. You need to hear entire words.

I was taught that you should learn at speed (~20wpm) with 2 characters and build from there.

Looking in to it, I see there's two ways - Farnsworth & Koch. The downside of the former is that you get stuck with slow speed and have a harder time to get faster. The downside of the latter is that the initial learning process seems more pointless because there's no actual words.

I found this site useful for practice: http://aa9pw.com/morsecode

A newbie question - What is the reason E and T have one beep only?

From wikipedia on Morse code:

> In his earliest code, Morse had planned to transmit only numerals, and to use a codebook to look up each word according to the number which had been sent. However, the code was soon expanded by Alfred Vail in 1840 to include letters and special characters, so it could be used more generally. Vail estimated the frequency of use of letters in the English language by counting the movable type he found in the type-cases of a local newspaper in Morristown. //

M [1] is always well down the frequency in general lists. I wonder if Vail's corpus was biased, whether he discounted local names.

Interesting to note that Morse planned a code using a dictionary; should we be calling it Vail code?

[1] http://letterfrequency.org/

Efficiency of transmission. Having shorter codes for more frequent letters reduces the average length of the encoding.

After listening to a Camille album from time to time, I subconsciously internalized the backing vocals of the song "Show Me The Waves". https://youtu.be/S0PMZg8lZ-M

Knowing that it's Morse code for the title of the song, I can reconstruct the individual codes for the letters AEHMOSTVW.

If someone were to write a hit song containing the whole Morse alphabet in a way that gets stuck in your head easily like in this Camille song, we might all learn it without trying.

If you're familiar with "YYZ" by Rush you know two more letters!

I've listened to this song probably hundreds of times over the years, and I never realized that was why it's called YYZ. I'm definitely not forgetting those letters now!

It's widely claimed, but most folks can't hear it because the rhythm is totally wrong. Listen to high low freq shifts instead.

And Kraftwerk’s «Radioaktivität / Radio-activity» - the title is tapped out in (not very good!) morse code.

I think morse code could be helpful for someone with locked-in syndrome, like Jean Bauby. In 'dictating' his memoir 'The Diving Bell and the Butterfly', it took him an average of two minutes to write each word by blinking at the time when the next letter he wanted was presented to him.


I've told my wife that if I'm ever in a situation where I'm hospitalized and seemingly unable to communicate, to call one of my ham radio friends over, because I'm going to be trying morse code (via blinking, twitching a finger, etc.).

Land mine, has taken my sight …

Did he have full access to his mental faculties otherwise? Why not spend the first 15 minutes asking to be taught a more efficient narrow-bandwidth means of communication, like morse code?

Interestingly, it seems Stephen Hawking used a similar input method, so maybe there's something particularly efficient about it that we don't realize http://www.hawking.org.uk/the-computer.html

Probably because of the last line:

> However although they work well for other people, I still find my cheek operated switch easier and less fatiguing to use.

When comparing how many physical actions are necessary per letter in morse code, vs this [0] example of him using his speech synthesizer, it's a lot more obvious.

[0] https://youtu.be/OTmPw4iy0hk?t=79

Meh, with an eventual VR/AR headset with eye tracking these patients will be able to look at letters they want to type and then blink to type them out that way.

Before you casually dismiss it as a worthless idea, consider that some victims have fewer capabilities than Bauby was left with.

If a person has lost all capability such that they cannot even blink or move their eyes, I’m not sure what else they could do to communicate. Breathing?

I don’t think the idea is worthless, just not worth as much as what others may think.

I'd genuinely want to be euthanised in this situation. But that's a whole other debate.

Nobody asked for that opinion here, and it's not relevant to the conversation.

I believe this already exists - an eye tracking camera that lets you indicate which letter you want next, with some predictive text capability. Can't find the exact video now, though.

The market leading system is made by Tobii Dynavox.


Totally dating myself here.

In the Hardy Boys book “Hostages of Hate”, Frank Hardy’s girlfriend, Callie Shaw, is kidnapped and is able to send a secret message to Frank by blinking her eyes on a video recording.

Funny what sticks in your mind from a young age.

There is a Hardy Boys book in which one of the boys makes a phone call to family with some nefarious person present. He holds an innocent conversation while sending a message in Morse code by tapping the phone with a pencil.

Based on an actual event from the Vietnam war iirc.

In the 21st century this would be:

kidnapped by terrorists and forced to read a statement.

IIRC this was used in Homeland, the TV show.

Had a question about this for while: Are movies where people communicate using knocks really accurate?

I would assume to effectively use Morse code, you’ll need a way to convey short silence, long silence, short beep, and long beep. Knocking misses the long beep.

I don't know if they're accurate, but there is Tap code for this exact reason.[0]

Not that I have known this before, but people talk about it in TFAs comments as 5x5.

[0] https://en.wikipedia.org/wiki/Tap_code

This has been a major annoyance for me my entire life. Of course, you cannot possibly produce a long knock, any more than shooting someone slowly is a vengeful way of killing.

What I guess you can do is vary the acoustics of your knocks (think knuckles vs palm) but this will require the recieving party to do much more processing when decoding.

I think that's a little bit pedantic. Complaining loudly about how the knocks technically aren't, themselves, longer or shorter is a digression from the reality that any sane human can detect and recognize consistent patterns in a series of knocks on a door, or beats on a drum.

  shave and-a hair cut... two! bits! [0]

  match in-the gas tank... boom! boom! [1]

  –··–· ··
Everybody recognizes that knock pattern, whether they know the trivia behind its origins or not. The singular instrument leaves the interpretation to the pattern of the beats alone.

Meanwhile, a rimshot, involving two drum hits and a cymbal crash offers two channels of interpretation.

  ba-dum tiss! [2]
The rhythm and the sound could each carry separate messages, which means the carrier signal would have to be agreed upon, before sending the message. Is the cymbal crash the best carrier for the dashes, or is the rest note in between the drum hits and the cymbal crash actually the pause that signals the dash?

[0] https://en.wikipedia.org/wiki/Shave_and_a_Haircut

[1] https://www.youtube.com/watch?v=i5KBVCoKYt0

[2] https://en.wikipedia.org/wiki/Rimshot

Knock whenever switching from beep-off to beep-on. The end of letters will be difficult to disambiguate but essentially unambiguous, because short pause + long pause (dot at end of letter) is distinct from medium pause + long pause (dash at end of letter), and they're both distinct from short pause (dot inside letter) and medium pause (dash inside letter).

American POWs in Vietnam reportedly used a tap code based on a 5x5 matrix of letters. Each letter is encoded as the indices within that matrix, so "tap, tap" is A, "tap tap, tap" is B and so on, through to "tap tap tap tap tap, tap tap tap tap tap" for Z. The letters C and K share the index [3, 1].


You can do longer pauses between knocks for dashes, and then even longer ones for the actual pauses. KnockKnockKnock vs Knock. Knock. Knock.

But there also are alternative codes for knocks, e.g. doing two groups of 1-5 knocks per letter (=25 letters encoded)

Two quick knocks can mean a dash. Also, isthereaproblemwithnospaceswhentryingtosaveyourlife?

> Also, isthereaproblemwithnospaceswhentryingtosaveyourlife?

I think that's a bit of an oversimplification since there's no ambiguity in text about where one letter ends and the other begins. The issue isn't spacing between words, it's spacing between _letters_

GP comment stated that four symbols (dot, dash, short pause, long pause) were required for Morse code. This is wrong; Morse code involves three symbols, dot, dash, and pause.

I assume your parent comment interpreted the spurious requirement for "long pause" as an assumption that for communication to take place, you'd have to have a robust distinction between letter spacing and word spacing.

Instrument pilots use Morse code to identify ground-based navigation aids such as Instrument Landing System (ILS) or VHF Omnidirectional Range (VOR)[0].

The VOR does not account for the aircraft heading. It only relays the aircraft direction from the station and has the same indications regardless of which way the nose is pointing. Tune the VOR receiver to the appropriate frequency of the selected VOR ground station, turn up the audio volume, and identify the station’s signal audibly. Then, rotate the OBS to center the CDI needle and read the course under or over the index.

Say a pilot is flying by reference to Avenal VORTAC[1]. She would tune a nav radio to 117.1 and listen for Morse code tones AVE, the station’s identifier and helpfully printed on the chart for quick reference. Nice avionics units will perform this translation for the pilot.

[0]: FAA Instrument Flying Handbook, https://www.faa.gov/regulations_policies/handbooks_manuals/a...

[1]: https://upload.wikimedia.org/wikipedia/commons/f/fc/VOR_on_s...

U.S. Senator Jeremiah Denton, Jr., was a POW for ~8 years during Vietnam. He did something similar.

> Denton was forced by his captors to participate in a 1966 televised propaganda interview which was broadcast in the United States. While answering questions and feigning trouble with the blinding television lights, Denton blinked his eyes in Morse code, spelling the word "TORTURE"—and confirming for the first time to U.S. Naval Intelligence that American POWs were in fact being tortured. [0]

You can see the video [1] on Youtube.

[0]: https://en.m.wikipedia.org/wiki/Jeremiah_Denton

[1]: https://m.youtube.com/watch?v=rufnWLVQcKg

It's in the article that you are commenting on.

Did you even look at the article? Denton’s story is mentioned in it and the video is embedded.

Nope. Lesson learned.

I learned Morse code (only listening and transcribing, not producing) at the German Army. It was an interesting, but also stressful experience, since we weren‘t practicing on natural language, but meaningless gibberish.

Alas, I have lost the ability. Everything is gone.

I've tried several morse keyboards for android, but all of them seem to be more fun than useful. if only I could get a cross between the hackers keyboard and morse, I could stop finger fucking these tiny screens and actually type what I want

Can you actually input morse faster than a keyboard?


The world record in sending letters (random 5 letter groups) is 283 for one minute: http://www.hst2017.org/en/records

This is with a special key that's made for high speed telegraphy. Anything on a mobile device would most likely be a lot slower.

There are some sample sounds of similar speeds, which you may find interesting.

I don't know. I feel that speed is not important to me, it's the accuracy (gained with practise, of course) that I want. I make a lot of typos, it takes me much longer to edit something on a touch interface as my thumbs just don't seem to be very accurate. I guess you could say I'd rather slowly input the right thing the first time than to edit it quickly the second time

There are so many things to learn these days but learning morse code to fight against kidnapping sounds about as esoteric/specific/paranoid as learning shorthand so as to be able to transcribe important conversations quickly without having to rely on audio equipment. Survivalists probably have more practical tips at hand before needing to learn ham radio and morse. Heck, learning to run fast and long is probably more useful for most people.

The "long knock" problem can be resolved in multiple ways. A dot can be a single knock, and a dash can be two knocks very close together, for example.

I learned Morse many decades ago and still remember much of it, but if I was in that situation now I'd probably think of FM-encoded ASCII first... and while it might not be as "efficent" as Morse, I think ASCII is much easier to remember and only requires two states to transmit.

Any idea where to learn it? Just in case.

Here are some resources from the ham organization ARRL on learning Morse code:


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact