124 comments

[ 6.2 ms ] story [ 222 ms ] thread
The article hints at, but doesn't really discuss, the concept of information entropy [1]. Each word in that message has a very high entropy because it conveys a lot of information. I read an XKCD What If article [2] a while back that gives a really cool and intuitive introduction to the concept. I don't know all that much about computer science so it was a great way to get learning more.

[1] https://en.wikipedia.org/wiki/Entropy_(information_theory)

[2] https://what-if.xkcd.com/34/

> As part of Chan’s research, the National Oceanic and Atmospheric Administration provided old weather maps that helped him determine the precise date of the weather observations in the coded note: May 27, 1888.

When the article introduced the weather report idea, I was hoping that this note was spy communications made to look like weather reports.

(Maybe the coded note wound up in a pocket because Elizabeth Bennet was joking with her sisters, pretending to be a spy.)

(comment deleted)
The note seems to have more than two lines? Have they decoded the other ones as well.
Anyone have a link to a document with the codes?
The code referred to is a "commercial code". There's a scan of a page from a 1910 code on Wikipedia.

Amazing that a phrase like "Confined yesterday, Twins, both dead, Mother not expected to live" was given a single code word ("Annosus").

https://en.wikipedia.org/wiki/Commercial_code_(communication...

"Unicode — The Universal Telegraphic Code Book" https://archive.org/details/unicodeuniversa00unkngoog/mode/2...

From Unicode, it's interesting but not surprising that so many of the codes are for significant personal and business problems, like the birth of children, missed travel connections etc. "Diota" "Amputation is considered unnecessary". "Annexus" "Confined to-day, Twins, one alive, a girl, Mother not expected to live".

Towards the back of the book is shown a very early 'DNS'; abbreviated addresses for businesses. "Supplies, London" meant "Junior Army and Navy Stores Limited", a bit like a generic supplies.co.uk. "Jowoto, London" meant "Johnson, Walker & Tolhurst".

You have to remember that telegrams were expensive, you'd not just send one for anything; it would be for major events (like a birth, or death) and so encoding the variations of that would be worth it.

In fact, it's just bitmapping.

BTW "Confined" has the meaning of being in labour, child birth.
Or almost. They're both euphemistic really - confined to bed; soon to labour.
Man, as a parent this chills me. Before modern medicine the statistics of deaths in childbirth were scary.
Death was something a lot more people spent a lot more time with than we do now. It was 23 years between the death of my last grandparent and the death of my father...and another 16 year til the death of my Father in law.
My uncle used to tell stories about death when he was growing up (30s or 40s). A line he would use always sticks either me, “back then, when you got sick, you died.”

He would go on to say just how common it was to know a child or adult that died. It was just a normal part of existence.

It used to be that if someone said "Jeff got in a car accident," it usually meant that Jeff was dead or disabled for life.

Today, it simply means that Jeff is going to be 45 minutes late for work.

Since I lived in the "surely dead" era, I still react that way if someone mentions someone getting in a wreck. People must assume I'm upset over Jeff's surely totaled car (something that didn't always happen in the olden days).

Yeah, the short code for it reminds me of the line from Broken Arrow (1996): “I don't know what's scarier, losing nuclear weapons, or that it happens so often there's actually a term for it.”

https://m.imdb.com/title/tt0115759/quotes/

It also puts a very different perspective on 1) why you always read in history books "so and so's third wife" and 2) the obsession with things like "child-bearing hips". Starting a family was downright hazardous; modern sensibilities are comparatively unburdened by those problems.
> the obsession with things like "child-bearing hips"

I am unaware of this being a past obsession, do you have a source?

I thought it’s a modern meme. Oxford lists earliest use as 1910: https://www.oed.com/dictionary/childbearing-hips_n

Might have been so obvious it need not have even been mentioned. Just like today, you wouldn’t say “I’m really looking for a healthy woman with an unmutilated face”.
As a non-parent, this is why I’m a non-parent.

Even though mother and child mortality is minimized, there are 999 life-altering things that can go wrong.

Take the simplest ones: what % of the population has the kind of mental illness that’ll destroy your relationship with them? E.g. borderline personality, sociopaths, etc.

Easily 5% or more.

That alone is a 1 in 20 chance of your kid ruining a significant portion of your life, not to mention you bringing a kid into the world with an inevitably ruined life.

And then there’s physical disability, not to mention financial burden.

It’s not worth it.

I’m very glad other people roll those dice: I’m happy to live in a populated world. But I don’t touch the dice myself.

I think 1 in 20 is a bit too high. 1 in 100 maybe to have a significant impact on your life. There are lots of tests now that so you have the chance to abort.

16% of the world population has a significant disability according to the WHO. Most of them are not born with it: age related (dementia, blindness,..), serious work accidents, lasting effects of diseases,...

Loads of them can get treated or won't alter life as much in the western world.

If your family history has no cases of mental illnesses, chances are very low to get them.

For a serious learning disability (not able to attend normal school), it's probably 1 in 20. Most of these kids will still live a fairly normal life in the western world.

Also, chances of disabilities also increase with the age of the parents. Best to have your kids between 20-35 to keep the odds down. Above 40 is a lot more dangerous than under 30.

> In 2021, there were an estimated 14.1 million adults aged 18 or older in the United States with [Serious Mental Illness]. This number represented 5.5% of all U.S. adults.

https://www.nimh.nih.gov/health/statistics/mental-illness

That's more than I expected for sure. But I do wonder how 'concentrated' these are. I'd think (mental) illnesses are really tied to genetics.

And social-economic factors play a big role too in food, stabile stress-free upbringing and healthy activities.

(comment deleted)
i also feel my children ruining my "pleasant calm" life all day and night every day, but when you see them discovering the world, learn how to speak, making their own language by their unconstrained little logical mind, marvelling at the world's tiny wonders which you can't remember ever deemed other than ordinary and given, first connecting the dots themself, acting exactly the way as you did when you were in their age yet completely uniqly, or independently developing the same lovely habits as your long dead grandma... is priceless and worth all the fatigues of sleepless nights for me.
You rolled the dice and won. You had a 17/20 chance of winning and you did.

Sleepless nights and fatigue? That’s the least of the associated worries.

People who rolled the same dice and ended up in the 3/20 group usually don’t comment.

Sometimes they do:

Some subset with a damaged life shrugs and says it was all worth it.

it looks like gamblers sustain the world's human population :)
There is a modern saying that goes like "People who lost their parents are called 'orphans'. People who lost their spouse are called 'widows'. But losing a child is so horrific that people who lost a kid has not even a word".

The reason that there's no word for people who lost a kid is because until yesterday every family lost a child. The word for someone who lost a child was "parent".

This is something I think about. If this happens today, people consider their lives ruined, go into deep depression, get divorced and often never recover. This can't have been the case before in time. How heavily does culture effect the experience and consequences of misforture? It seems to me like we may have some form of entitlement, like spoiled children, that has made us weaker. If that's the case, can we change our culture? I think so and I think we should consider that if it's a possibility.
> "we may have some form of entitlement, like spoiled children, that has made us weaker"

An alternative way to look at it is that the people of the past were subjected to massive unprocessed trauma at every age, and that made them weaker than us.

Who’s weak and who’s strong in this scenario: a bloody body is presented following a tragic car accident. The body is maimed, broken, but ultimately salvageable.

One person sees the body and shrieks in horror and goes into a fit of panic.

Another (an ER doctor) is cool as a cucumber from years of desensitization, calmly treats and saves the body.

I’d have a hard time describing the ER doctor as “weaker”. They’re unquestionable more capable and useful. And I propose stronger not weaker.

Strength is subjective, especially if your definition is “capability”. That ER doctor may be so desensitized that they are incapable of providing emotional support, and the one who ran may be a therapist who can provide the individual in the car crash emotional support. So in the end both are strong, simply in different ways.
(comment deleted)
There are examples to the contrary, people who despite the tragedy went on to do other things in life.

President Joe Biden has outlived two of his children. In 1972, his daughter and wife died in a car accident.

I don't talk about this very often for obvious reasons, but I've lost a child.

Grief over losing a child is not entitlement, or weakness, or being spoiled. It's grief, as raw and human as it's always been.

It used to be normal for children to die in the mid-1800's-England. Something like 25% of children born didn't make it to their first birthday.

Today, that number is more like 0.34%.

I suspect the difference is that in the 1800's it happened a lot. You weren't the only one. You had people who were empathetic. Nearly any person you'd come in contact with likely had either experienced the same trauma, or loved people who had. That doesn't mean people didn't grieve over the loss. It just meant that living with the grief of a lost child was normal.

In "modern" times? If it happens to you, you're an outlier. Not only are you grieving the loss of your child, you're alone. No one you know has also lost a child. You're grieving the loss of potential, the legal garbage, insurance, funeral costs, you're doing it either with your partner, also grieving, or alone (or maybe worse, with a partner, but still "alone"). The amount of times you'll hear "I can't imagine what that must be like" help reinforce that feeling of being alone, because it's simply true. It's unimaginable. It was for me, and even having experienced it, still is.

Whether it's the 1800s or 2000's, humans are resilient. While losing a child often does lead to depression and divorce, we can heal from these things. We have an advantage over 1800s parents in that we have huge support networks that are no further away than our pockets that blow away whatever community support might have existed in the 1800s. Not to mention medicine and advances in our understanding of human psychology.

Grief isn't weakness. Grief is simply human.

Your post has nothing to do with mine. Grief is normal but letting grief permanently ruin one's life isn't. It's new and is a weakness.
> letting grief permanently ruin one's life

And this is why we will disagree. What, by your definition, does "ruin one's life" actually mean? I bet it differs from mine. An easy bet, because we've led different lives with different experiences. What you think of as "ruined" may be my version of "it took 15 years to finally get over something tramautic"

What I see in my own grief, and the many, and I mean many parents who I interact with as part of my own support, is that each of us takes it day by day. Some people have a lot more days of waking up wishing they were dead than other people. Some people channel the grief into starting foundations. None of their behavior is strong, or weak, or entitled, good, bad, right, wrong, or whatever subjective sort of adjective you want to categorize it with. Nor are their lives "ruined", even if they're 14 years into waking up every day wanting to die. Day by day. It's people who have lost someone, someone so innocent, someone they're supposed to protect - there are no words to describe the pain. So while you're trying to decide if someone's grief is "weakness" or "entitlement", they're experiencing it.

Grief is evidence of your love for someone, and in my mind, that sort of love takes a certain sort of strength. To love, knowing it cannot last forever, takes strength. That is as old as humankind itself, and not new, nor a weakness.

You get different attitudes. In Russia it seems to be "Putin urges Russian women to have ‘eight or more’ children amid soaring deaths in his Ukraine war". Churn out some spare ones. https://www.independent.co.uk/news/world/europe/vladimir-put...
The irony is that up until not long ago it is used to be quite common in a single family to have ten kids, but now a few children (less than five) is the norm.
Is the code for the weather report at all similar to the one for "help I am a prisoner in a vintage dress factory"?
That certainly led me to a very pleasant rabbit hole! Different codes invented for different trades or professions, also...
I really hope those messages had a CRC check so as not to start wars unnecessarily.
Used for telegraphy, too. I recommend reading "The Victorian Internet" by Tom Standage.

https://www.unicode.org/L2/Historical/Unicode-Telegraphic-Ph... is an example but from what I've read there were massive volumes that included all sorts of specific stuff.

Companies coming up with their own codes to save on telegraphy costs were cool, code books with random letters smooshed together. Obviously telegraph operators had trouble with this, reducing their pace so telegraphy companies banned usage of non-words in telegrams (more on that in the book I mentioned).

Reminds me of when I first encountered the extraordinarily compressed 12 bit "words" in a ARINC717 data stream, where they're conserving bits that are written as-is, serially. Like three decades of computing passed it by. If something has four distinct values, it gets two bits. Then, right on top of that: is it a syncro response curve? Output the binary for the curve function. On and on, for thousands of "words", all jammed together. I used to imagine being a future researcher, trying to decode these big blocks of undifferentiated binary, the amount of legwork I'd have to accomplish to get just a few layers deep.
ASN.1 PER - used widely in telco, notably GSM standards, is also extremely dense. Given what they charged for data, it's not surprising they'd want an encoding that's as efficient as possible for the overhead.
Ah, yes, telegraphic codes. I once found a large code book for one of those in the stacks of a Stanford library. Every "word" had four syllables, and mapped to a longer phrase.

In the book was a loose piece of paper with a note that the telegraph company was changing their billing rules and that only known words would count as one word for billing purposes. Anything else would be charged at a higher rate for random letters and numbers.

Here's a typical telegraphic code, "The Anglo-American Code to Cheapen Telegraphy and Furnish a Complete Cypher".[1]

[1] https://babel.hathitrust.org/cgi/pt?id=nnc1.cu55719287

It’s easy to judge in retrospect, but shouldn’t the slashes have been a clue? Once I saw them my first thought was this was a checklist and someone marked each line off as it was sent.
The CNN article makes it seem like it was a mystery for a long time, but if you look at the original blog, it's clear in the comments that from the very beginning readers of the blog suggested it might be a telegraph code, but they didn't know exactly from which book.
Some mysteries are from lack of suggestions, some from lack of listening.
CNN blocks ad-blockers (seemingly):

> Browser Blocked

> We apologize, but your web browser is configured in such a way that it is preventing this site from implementing required components that protect your privacy and allow you to view and change your privacy settings. This functionality is required for privacy legislation in your region.

> We recommend you use a different browser or disable the “EasyList Cookie” filter from your “Content Filtering” settings (found under “Settings” -> “Shields” in the Brave Browser).

UPDATE: The blocking is triggered by the "I don't care about cookies" browser extension (Firefox); disabling the extension on the site gets me past the block. Utterly stupid move on CNN's part.

Using uBlock Origin [on Firefox Win10] and didn't have any problems.
Exact same for me... didn't notice anything strange. I also erase cookies when I close the browser, might also perhaps have an impact depending on previous visits to CNN.
I block ads, don’t really mind when a site doesn’t feel like serving me as a result. I’m intentionally not accepting their ads-for-content trade and asking what they’ll send for nothing. If that’s nothing, fair enough.

I am surprised nowadays, given how bad the trade has gotten, that people are interested in discussing their content.

Right, I got the same message when accessing the page... so I just clicked back and came to the comments. I don't care enough to read the article to disable my security mechanisms, there's plenty of alternative content :)
> I don't care about cookies

That extension was bought by Avast.

There's a community fork called "I still don't care about cookies" available for Firefox and Chrome.

Why use Omit instead of t56? Or better yet, write the values in a pre-defined order and just do 56 0.08 32 no 12.

Why did they feel the need to use actual words?

Can imagine this approach being less error-prone for laymen, in the same way that passphrases (word word word word word) can be friendlier than passwords (word12345678).
Presumably has built in error-correction. You don't want to go from t56 "Cholera, family dead, stock market crash" to t66 "Buy farmland, rains predicted good" in one typo.
They meant t56 as in "temperature is 56", not as in every random phrase gets assigned a jumble of numbers and letters. Though I suppose error-correction could still apply.
Actual words were easier for the telegraph operator to send and receive so they were charged less than a scramble of letters or nonsense. Words were charged individually, but could be up to 10 letters long. It makes sense to avoid the shortest words to get some error correction, "— ." = "T E" could be misheard as "—." = "N".

The international regulations limited the cheaper rate messages to only certain languages, of which one was Latin, so the codebooks used that to avoid confusion with modern languages.

(See my other comment and read the introduction in the linked book.)

(comment deleted)
Oh so words were charged indiviually. That's strange but ok. So that means if you could hypothetically serialize/compress the whole data set into one word you'd be most cost effective.
But as said before, it looks like words were limited to 10 letters. So you could encode/compress a bit, but not a long message into a single 243-letters word...
> But as said before, it looks like words were limited to 10 letters.

On average or individually? A cap of 10 letters doesn't really make sense if you're specifically allowing Latin to be charged by the word.

You get charged some multiple of sum([ceil(length(word) / 10) for word in words]), plus a flat base fee. At least that's how it was the last time I sent a telegram.
Good point to link to what3words.

Have a look, it’s a geographic encoding system.

Because the telegram company charges more for non words.
Was hoping for secret treasure, but that was still very interesting. Thank you for the post.
Anyone tried it to get access to Satoshi wallet?
But why stored in a secret pocket?
Pockets are for the help, this is a nice dress so a secret pocket is probably the only type of pocket they'd want.
Still, why would an elegant woman carry a weather report encoded in telegraph jargon...?
Maybe she didn't know it was a weather report encoded in telegraph jargon and instead thought it was some kind of crypto puzzle that she kept handy to work on in idle moments.
It could have been part of her job:

> Rivers Cofield also notes that the dress – although beautiful and fancy to our modern eyes – was not exactly what someone would wear to a ball. It was more like the ‘business casual’ of the day, and to her that does indicate it might have been worn to work.

> Chan has also documented that a number of women worked as clerical staff at the Washington, D.C. offices for the Army Signal Service in the 1880s.

https://www.noaa.gov/heritage/stories/cryptogram-in-silk-dre...

As dools notes, female clothing doesn't usually have pockets. So if you want a pocket at all, it will probably be a secret pocket.
My understanding is that that was not true of clothing of that era. That being said, the description of the pocket is unclear and it may have been a normal pocket, which would have been built into the folds of the dress so as not to be overtly visible.

ETA: If memory serves, pockets disappeared from women's clothing as skirts became more form-fitting in the 1910s and 1920s.

> If memory serves, pockets disappeared from women's clothing as skirts became more form-fitting in the 1910s and 1920s.

Definitely not true considering this quote from 1899:

> No pocketless people has ever been great since pockets were invented, and the female sex cannot rival us while it is pocketless.

(as quoted from https://www.vox.com/2016/9/19/12865560/politics-of-pockets-s... )

According to the article, female clothing has just never had pockets:

> Once upon a time, everyone carried bags. In the Medieval era, both men and women tied their bags to the waist or wore them suspended from belts; these bags looked very much like Renfaire fanny packs. As the rural world grew more urban and criminals more sophisticated, people cunningly hid their external pockets under layers of clothing to hinder cutpurses; men’s jackets and women’s petticoats were outfitted with little slits that allowed to you access your tied-on pockets through your clothing.

> Only in the late seventeenth century did pockets make their move to become part of men’s clothing, permanently sewn into coats, waistcoats, and trousers; women’s pockets, however, failed to make the same migration. Lacking built-in pockets, women continued to hide their tied-on pockets, which were large, often pendulous bags.

> The French Revolution changed everything. While the mid-eighteenth century lavished in rococo, wide skirts that screamed decadence and wealth in their yards and yards of fabric, the end of the eighteenth century whispered restraint. Skirts pulled in close to the body, the natural waist crept ever upward, and the silhouette thinned to a slender column. This neoclassical look had no room for pouchy pockets, yet women still needed to carry their stuff. The reticule, a small, highly decorated purse, was born — and like a pernicious poltergeist, it has never really gone away.

To hear the article tell it, demanding pockets in their clothing was an early form of women's liberation around the period you describe, but one that failed.

It’s encoded words for weather forecast.

Save you a click.

The story is pretty interesting, and definitely worth the click.
Why would someone encode words for a weather forecast?
As the article explains, it's to save on telegraph fees, which were charged per word.
Seems like OP didn't save me a click, then. Dang.
It was an early form of compression.
Telegraph operators preferred operating on real words, so you had to encode your data into English words to save on transmission costs.
These garments aren't washed with water?
Sending 10 words costed a dollar,which according to an inflation calendar is around $27 today.
I am sure telegraph spam was low during that era.
My wife once bought a Burberry coat for 100 bucks off Poshmark and it came with a free gram of rather nice cocaine!
The cocaine is the most sure indicator that it was genuine!
I wonder how the legal nuances of unknowingly mailing vintage cocaine to someone would work. Would the seller still be held accountable for this?
This was pre fentanyl contamination of the supply. You gotta have your head screwed on even more tightly these days...
(comment deleted)
It's always a bit surprising to me how little monitary value very old items can have.
Seems like the code books would be useful for phone texting. (Note that they're actual words, not acronyms like lmao.)
Meanwhile…someday our great great grandchildren will find a MacBook from “the early 20’s” and when they open it they will see a file opened to VSCode with cryptic symbols on it.

“Iran?” they’ll wonder, “perhaps it was code for the ongoing conflict with Iran.”

But in fact it will just be someone who literally died trying to get Webpack to work on a website for pirated Tarantino films.