Origin of 'Daemon' in Computing | Hacker News Reader

Origin of 'Daemon' in Computing

takeourword.com

238 pointswizerno103 comments

The *nix world is full of dark-but-fun terminology. Daemons run the system. New files get 666 (before the umask takes away unnecessary permissions). Parents kill their children before killing themselves. And sometimes you have to kill zombies.

From: https://devrant.com/rants/1101391/my-daily-unix-command-list...

unzip; strip; touch; finger; mount; fsck; more; yes; unmount; sleep.

yum

Except you shouldn’t fsck while mounted.

Works fine in FreeBSD/ufs :p

> unzip; strip; touch; finger; mount; fsck; more; yes; unmount; sleep.

T E C H N O L O G I C

…

T E C H N O L O G I C

This is a weird thread to happen upon when my other monitor has variables named sexp.

Tail/head

It’s almost like these commands were all made by nerd teenage boys.

or rather nerd teenagers finding a way to joke on terms that didn't relate to sex until they thought of it and started spreading.

Device is not ready

Almost feels like a right of passage when you inevitably google something like "kill self" (in reference to killing the current process) and get a popup telling you about suicide resources.

Or indeed a rite of passage

Or "kill orphaned children" and be put a list somewhere

"Kill orphaned child process"

Sometimes you have to kill orphans too!

Unics (from Multics).

Zombies or orphans, depending on which side you want to play as today. Plenty of killing to be done on both.

Zombies can't be killed for they are already dead; they can only be reaped, by waiting on them. (This is why init inherits orphans, so it may reap them when they eventually die.)

From: https://www.unixprogram.com/churchofbsd/index.html

One day I was at a restaurant explaining process control to one of my disciples. I was mentioning how we have to kill the children (child processes) if they become unresponsive. Or we can even set an alarm for the children to kill themselves. That the parent need to wait (wait3) and acknowledge that the child has died or else it will become a zombie.

The look of horror the woman sitting across had was unforgettable. I tried to explain it was a computer software thing but it was too late, she fled terrified, probably to call the police or something. I didn't really want to stick around too long to find out.

> We also assume that this is the meaning behind the daemon.co.uk, host to many United Kingdom web sites

Not sure if it was the origin of the company name, but the domain was demon.co.uk not daemon. E.g. I had pretence.demon.co.uk with them for a few years.

https://en.m.wikipedia.org/wiki/Demon_Internet

Yeah, never heard of a 'daemon.co.uk' in the 90s, but likewise had a Demon account...

I find the “a la mode” vs “au jus” discussion right under the daemon one very interesting!

I wasn’t familiar with both of these expressions but I looked it up and “a la mode” is an American culinary expression, meaning “served with ice cream”. And “au jus” is also an American culinary expression, meaning “gravy” or “broth”. Now, even though they are both derived from a French expression that is a prepositional phrase with à (meaning with), it does not matter any more when they were borrowed to English.

“A la mode” became a new adverbial expression meaning just that: “served with ice cream”. You can have pie a la mode = pie served with ice cream, but obviously not *pie with a la mode = pie with served with ice cream.

And “au jus” became a noun expression meaning “broth” or “gravy”. And you must say sandwich with au jus = sandwich with gravy and can’t say *sandwich au jus = sandwich gravy.

What is extremely interesting here is that it bothers the prescriptivist who wants language to be a certain way he feels it is supposed to be, also the author on that webpage.

Yeah, I was hanging out with someone recently who kept using "au jus" like "sauce", i.e. "you could make that with an au jus" , "ooh yeah that would be so good with an au jus on the side!" or similar ...

I immediately ordered my daemon to cook me some pilipili au jus de cuniculus.

Also, I think I will risk opening my eyes now.

You can say sandwich au jus but it refers to the sandwich with gravy not the gravy

"Warning: This paragraph is about science so, if this topic causes you undue alarm, please close your eyes until you've finished reading it."

Amazing.

Ha -- I read the title and said to myself, "gotta be Maxwell, right?" The jolt of pleasure I get from being right about things like this is unreasonable.

Yeah, author tries to throw us a curve ball at first with that koo-koo backronym.

I remember learning about Maxwell's Daemon through the Apple II game Dr. Maxwell's Molecule Magic, in which you take the role of the daemon. You must toggle the barrier on and off in order to trap enough gas molecules at high enough pressure to launch a rocket ship. Once you think you have enough, you can then launch the rocket to see how well you did. If you were successful, the rocket would blast off the screen and an image would show of an astronaut on the moon saying "Hi, Mom!" (Speech was provided via PWM through the Apple II speaker.)

Eleven-year-old me was easy to entertain. Especially if rockets, robots, or science was involved.

Earlier threads:

- [2023] https://news.ycombinator.com/item?id=35283067 (24 comments)

- [2022] https://news.ycombinator.com/item?id=31069163 (127 comments)

- [2018] https://news.ycombinator.com/item?id=16299583 (46 comments)

- [2011] https://news.ycombinator.com/item?id=2691752 (45 comments)

Unrelated to the word "daemon", but related to the article, I was a bit surprised by this assertion:

> Eventually, though, the theory of quantum mechanics showed why it wouldn't work.

I was familiar with the information theory arguments (the same presented in Wikipedia[1]). Is that why they mean here by "quantum mechanics" or is there another counterargument to Maxwell's daemon?

1: https://en.m.wikipedia.org/wiki/Maxwell's_demon#Criticism_an...

I'm guessing that the daemon's ability to allow only fast molecules through the gate depends on knowing their position and velocity simultaneously?

But the daemon doesn't need to know them all that precisely.

I think you'd have to be pretty precise to know if it's heading towards a hole that's only just large enough for it.

Why would you make the hole so small?

The main requirement for the hole is that it's small enough that (with high enough probability) only at most one or a handful of molecules will make its way through.

And that size is completely independent of the size of your molecules, and only depends on how many there are per unit volume. There's a lot of 'empty space' between molecules in a gas.

Good question. I was going off the linked article which states:

> In the middle of the divider was a tiny gate, just large enough to admit one molecule of gas.

Still, that's quite a small hole relatively speaking. So you'd have to be fairly precise about both position and velocity. Potentially more than is allowed by Plank's constant. I dunno though, this isn't one of the counterarguments in the Wikipedia page, so probably you're right.

Air molecules are large and massive, so their de Broglie wavelength is actually much smaller than their physical size (and that's much smaller than the hole needs to be to let in one at a time at ambient pressure and temperature), and you don't need to know their speeds all that well, eg if you just want to 'pump' all the air from one chamber into another.

So all in all, a classic description would work reasonably well. (Remember that quantum uncertainty is related more to de Boglie wavelength than physical size.)

I'm not sure I understand - why is the wavelength size relevant? In my understanding, the standard deviations of position and momentum are at least some constant multiple of Plank's constant.

Sorry, I meant to say that for particles (or collections of particles) with a short de Broglie wavelength classical effects dominate quantum effects.

The typical scale at which quantum effects dominate for a system is roughly equivalent to the de Broglie wavelength of the relevant parts.

Does that make sense?

It seems to come from measuring the particles at all. One result is that the demon has to store information about the particles, and erasing that information to free up memory increases the entropy of the gas/demon system.

https://en.m.wikipedia.org/wiki/Maxwell%27s_demon#Criticism_...

Yes, exactly, and this is what allows quantification.

http://www.av8n.com/physics/thermo/entropy-more.html#sec-pha...

With 'entropy' being an obsolete term for (lack of) information, and

> “classical thermodynamics” is a contradiction in terms.

On the "information" part, I'm not sure that's correct, have you seen the papers that dispute the Bayesian versions of thermodynamics?

I think practically though, even before you hit anything "quantum", the requirement that you physically interact with the system is what dooms you.

No, got a link ?

Searching for them did bring me to an interesting discussion :

https://www.lesswrong.com/posts/YSFRazdoWXKHgNakz/link-the-b... (2015)

and then to :

https://bayes.wustl.edu/etj/articles/second.law.pdf (1998)

Which confirms my suspicions, but also sheds lights on how old the confusion is !

There are a bunch of assumptions that are easy to make (because they almost always are true), but very hard to get rid of when they aren't :

- that entropy is objective/ontological rather than subjective/epistemic

- that entropy is equivalent to disorder

- that temperature can always be defined

- that entropy is extensive

(- I think there was at least another one, but I had to do something else in-between and I don't remember now)

- oh yeah, maybe it was that there's a difference between a distribution and a macrostate ? (not sure about it myself)

Now, I don't know what the Bayesian framework can bring to the table here (not being sufficiently familiar with it down to the nuts and bolts of calculations), but if it can prevent us (and future students) from making these mistakes over and over and over again, it would be real progress.

https://arxiv.org/abs/1508.02421v1

https://arxiv.org/abs/1508.02421v3 (updated, 2017)

They claim to fix the criticisms, see the section "The Bayesian arrow of time."

Who knows how well they did though.

As far as I can tell they're still making impossible assumptions, because certain Bayesian problems can't be calculated under a certain amount of energy, and some can't be calculated at all while embedded in spacetime (excepting time travel, and sometimes even then).

I think it's necessary to increase (on expectation) the entropy in a closed system when measuring, unless you take measuring to be magic and not a physical process.

Thanks, I need to look into those whenever I can find the time, however it also sounds like they are trying to fit General Relativity in there ?

I can't say for sure if this is doomed in the first place because GR infamously is not compatible with quantum mechanics (and trying to match them by force is only going to produce confusion and nonsense), or potentially revolutionary in actually managing to reconcile them...

It probably (if the calculations are right) is unable to actually do much of anything useful (because it's too complex to avoid being extremely correlated with the rest of the universe ("embedded")), and even if it could it wouldn't be better than a standard ASI in most real-world situations.

That's assuming you aren't trying to claw back more energy than you lose, I'm pretty sure that's not possible to reliably do without crazy hypothetical physics.

Always thought it was because it denotes a process that stays alive and so is like a little living demon in your computer.

Of course it's not the origin of the usage, but I always found Lovecraft's quote from Lactantius to be pretty adequate:

"Demons have the ability to cause people to see things that do not exist as if they did exist. -- Lactantius"

http://blogicaster.blogspot.com/2011/10/lovecrafts-lactantiu...

I knew the origin of daemon from high school Greek philosophy courses.

At the time, I thought "when an I ever going to use this stuff in real life?" Then I got into computers.

I know, right? I used to read about the role of eunuchs in ancient royal courts, and I thought to myself, who does that anymore?

I recall reading a mail exchange posted on some usenet group about, IIRC, some boss or similar higher-up being shocked to discover the UNIX systems they ran were full of demons, and as a devout christian wanted the sysadmin to put an end to that immediately. The sysadmin was repeatedly trying to explain that a daemon was not the same as a demon, without success.

Tried to find the post again, but no dice :(

A real BOFH would've just complied with the boss' request and made sure all the daemons were killed and stayed dead.

Wonderfully written. “Warning: This paragraph is about science so, if this topic causes you undue alarm, please close your eyes until you've finished reading it.”

"Disk And Executive MONitor" does sound kinda cool, though.

I’d guess someone backed into that acronym when asked “why do you call these things daemons?”

Does Anyone Else Make Outrageous Names?

Do you pronounce it as demon or like Matt Damon?

I believe it officially is pronounced the same as demon. But colloquially I hear a lot of (and sometimes find myself using) “damon”.

I’m in the Matt Damon camp. Always pronounced it that way, never really gave it much thought. Just seems “right” to me.

And now we can perhaps hash out whether the `lib` in `/usr/lib` is pronounced with a long or short 'i'. I hope I'm not the only one who pronounced it the first way with no real thought, and never noticed until I heard someone else say it with a long 'i' that that was obviously the logical pronunciation.

Surely the logical pronunciation is the way you'd pronounce it in library, so a long 'ai' rather than any kind of 'i'? Though I personally always use the short 'i'. I was going to justify that by saying it's the same as /usr/bin, but that's also short for binaries, so should also be an 'ai'.

I've been pronouncing it with the short 'i' for 30 years, but mainly, possibly only, in my head.

In 1998 I started a new job, and my boss pronounced "URL" as "earl". That threw me for a loop, had to fight my way through our first conversation before I figured out what he was saying.

I pronounce API as "appy", which sometimes draws quizzical looks (people think I'm using next-level cutesy slang for "application"). But I never could do the "earl" thing. Or "sequel".

> Surely the logical pronunciation is the way you'd pronounce it in library, so a long 'ai' rather than any kind of 'i'?

Yep, that's what I meant to say with:

> … never noticed until I heard someone else say it with a long 'i' that that was obviously the logical pronunciation.

But maybe the sentence structure was too tortured for it to be clear what I was saying.

> Though I personally always use the short 'i'. I was going to justify that by saying it's the same as /usr/bin, but that's also short for binaries, so should also be an 'ai'.

Oh, shoot, even after I noticed the logical pronunciation of "lib" (long 'i') it never occurred to me that the same applied to "bin". I guess I just can't say any paths out loud any more.

Perhaps the difference for you is that "bin" is already an English word with an official pronunciation.

Personally, I also use short-i for "lib," because I tend to pronounce shortenings of text as if they were words themselves.

Do you pronounce the vowel in /var as in "bar" or as in "bare"?

Also, for those who try to pronounce everything rather than spell them out, where does it end?

I now have a newly discovered, morbid interest in how such folks say path elements like "selinux", "httpd", and "pgsql"...

> Do you pronounce the vowel in /var as in "bar" or as in "bare"?

Ha, they're everywhere!

> that's what I meant to say

Ah, that makes sense. I thought you meant long 'i' as in extending the duration of the 'i' sound, like in 'beep' vs 'bip'.

I have always pronounced lib like the word liberal.

I was mind-blown the first time I heard someone pronounce etc as "et-see".

et-see rolls off the tongue so much better than ee-tee-see that it makes perfect sense now.

I've pronounced it as `etcet` as short for `et cetera`. But that's probably my Australian habit of shortening everything.

Hard agree. It's an archaic word, it seems a shame not to revel in that fact whenever you use it! I've been using "automata" a lot recently, and that's another one that is just more fun the unusual way. Also it helps that it's clear either way -- no one will be confused upon hearing "daymon" even if they're not used to it, unlike "etsee", "user-slash-libe", "user-slash-bine". Or, god forbid, "oo-zir slash"...

It should more properly be written as dæmon. The æ ("ash") character is usually pronounced more like "ee", as in encyclopædia. I've never heard anyone say "encycloPAYdia" :-)

because it's spelled encyclopedia

US English spells it as encyclopedia, British English spells it as encyclopaedia.

Fascinating! This is why I stick with nice, clean structural linguistics, this applied stuff gets sticky. I just confirmed on Youtube that the (some?) British people do indeed pronounce "Aesthetic" as "ah-stet-ic" not "ee-stet-ic", and upon diving a bit, it seems that the rule is "don't ask for a rule, you fool! It's 'e' now except for when it isn't." Thanks for the interesting tidbit!

  The letter æ was used in Old English to represent the vowel that's pronounced in Modern English ash, fan, happy, and last: /æ/. Mostly we now spell that vowel with the letter a, because of the Great Vowel Shift.
  When æ appears in writing Modern English, it's meant to be a typographic variant of ae, and is pronounced the same as that sequence of vowel letters would be. So Encyclopaedia or Encyclopædia, no difference.

https://english.stackexchange.com/questions/70927/how-is-%C3...

Highly recommend the protracted arguments in the comments, that's a wonderfully pedantic StackExchange. Big shoutout to someone in 2012 defining "NLP" as an unusual word -- how the world has changed! It's only a matter of time before they open an AP/IB course in NLP...

"Aesthetic" gets even stickier! In the UK I tend to more commonly hear it pronounced as "es-thetic".

The Great Vowel Shift indeed makes written English much more confusing than it perhaps should be. English is already a messy hodge-podge of a language, then our writing system started to get standardised (or standardized, if you're American!) right as pronunciation started to change, leading to the written version of words suddenly no longer being anything like the pronunciation.

Hmm, I'm a Californian and I pronounce daemon as demon, understanding the first vowel as the same vowel as for Aesop. Indistinguishable from the vowel in "beam" and "niece".

But I pronounce the first vowel in aesthetic differently. For me, it's somewhat in between the vowels in "bed" and "bad" but closer to the former.

…TIL how to pronounce “Aesop”! Thanks for saving me eventual embarrassment - now I know why other people don’t mix up Aesop Rock and ASAP Rocky!

As a fellow Californian, I’d say we have authority anyway - I was taught in school that Ohio has the least specialized dialect, but that’s based on newscasters and such. The 21st century is the Californian century!

…that is, assuming Brussels’ English is out of the running, I suppose ;)

The letter "æ" as used in Old English does indeed correspond to /æ/, but we don't use that letter (or even digraph) for this purpose anymore. In all the words where it is still occasionally used, it corresponds not to Old English "æ", but to Latin "ae", which is [ae̯].

Oh I get it so it's like ä!

Yes. Daemon is just the archaic spelling of demon. The ae is a vowel sound that didn't survive to modern times. The word was NEVER pronounced "damon." To my knowledge.

The original pronunciation of ae/æ in words originating from Latin or Greek is basically like "I". As usual, English molded it into something else in many cases, which is why we write "demon" these days. But if you insist of "daemon", then it really ought to be pronounced like the original Greek δαίμων.

I pedantically use day-mon (working/servant spirit) to distinguish from dee-mon (evil spirit) but I suspect I'm very much in the minority.

https://en.wikipedia.org/wiki/Daimon vs https://en.wikipedia.org/wiki/Demon

I've never found any significance to associate the unix term with Demons and consider it a mis-association.

Don't let MAGA hear you... they'll start banning and burning linux boxes.

Dae-mon, ah-ah-aah.

They rely too much on Professor Corbato's account:

Corbato confidently gives the reason but A) doesn't claim to have coined the term, and may not know what the coiner's thinking was; B) at the time may have had a different understanding than some other members of the group - it's not the sort of thing that people have a meeting about; and C) is writing about something that happened decades ago.

Corbato then cites Take Our Word's prior description - people who weren't present when the word was coined, and who openly say they have no idea: "This is so reminiscent of Maxwell's daemon watching his molecules that we can only assume that whoever dubbed these "system processes" had Maxwell's daemon in mind. Unfortunately, we have found no hard evidence to support this."

Then Take Our Word cites Corbato, creating a loop. The only evidence in that loop is Corbato's flawed (being human), prompted (by reading Take Our Word) memory of possibly limited knowledge from decades ago.

They do have more evidence, which I overlooked (apologies to the OP):

"Professor Jerome H. Saltzer, who also worked on Project MAC, confirms the Maxwell's demon explanation. He is currently working on pinpointing the origin of the erroneous acronym etymology for daemon in this sense."

Saw this on a conspiracy theory subreddit recently, thought it was absolutely hilarious:

"Boy I love trapping demons in microscopic silicon megastructures to do my bidding, I sure hope nothing goes wrong"

I told my eight year old the “we taught sand how to think” joke, and he loved it, and also indulged me in a brief explanation.

It really is absolutely wild that it all works when you go to the absolute fundamentals and start working forward.

The funny thing is that before that we taught light bulbs how to think(thermionic logic), and before that we taught magnets how to think(relay logic) and before that we taught levers how to think(gear logic)

It’s magic all the way down.

Didn't some Wharton professor post something similar too recently?

I always wondered about this. Now I know. Thank you :)

Maybe we should have used `eudaemon` (happy spirit) to avoid alarming the common folk.

from PIE dai-mon- "divider, provider" (of fortunes or destinies), from root da- "to divide."

to divide power, compute.

Huh, I assumed: it does work whilst remaining unseen

Interestingly, Safari Reader View doesn't work on this page. Firefox's does.

Alternatively, here's a readable mirror: https://ei.cs.vt.edu/%7Ehistory/Daemon.html

And another: https://old.reddit.com/r/linux/comments/7w7914/the_origin_of...

This was the internet we loved, I miss it.

> Safari Reader View doesn't work on this page.

What do you mean? I just tried it and it was fine.

For me (macOS 15.0.1), it only extracts the biggest quote (Your explanation of the origin of the word daemon ... United Kingdom web sites), the rest of the text is lost.

Works on iOS. Page is so horrendous that I immediately reached for it.

“Text of the site below because looking at that site hurt. A lot.” For fucking real!

Reply on news.ycombinator.com