What is the origin of “daemon” with regards to computing? (2011) | Hacker News Reader

What is the origin of “daemon” with regards to computing? (2011)

english.stackexchange.com

218 pointsdjoldman127 comments

According to the jargon file, ITS called daemons dragons and SAIL (whatever this is ?) called them phantoms. Add to that zombie processes on Unix and user rights angels from FreeBSD's capsicum and the fact that processes come from executable files in ELF format and have debug infos in DWARF and it is quite the fantasy menagerie.

I suggest Kobold for whatever kind of process we need to name next.

> SAIL (whatever this is ?)

Science and Artificial Intelligence Laboratory

Wrong: it's the Stanford Artificial Intelligence Laboratory (founded 1962 by John McCarthy -- who co-authored the document that first defined AI as a field, invented LISP, time sharing, and garbage collection, heavily influenced the design of ALGOL, and so on: one of the pioneers of computer science):

https://ai.stanford.edu/

https://en.wikipedia.org/wiki/John_McCarthy_(computer_scient...

LISP even has its own song.

https://www.gnu.org/fun/jokes/eternal-flame.en.html

Heh. I have a recording of that by Kathy Mar which is somewhat distinguished as a filk song by being sung by someone who actually sings well.

Neat

Filk- like folk, but for IT people.

Nope, the term came out of SF fandom decades ago.

Read a few articles and even novels from the 80s and poeple constantly refer very good programmers as "wizards". Also the sub culture of DND sticks hard. I guess most likely the reason was computer programming back in 80s was mostly low level considering one had to use assembly/C/Pascal for serious PC programming.

“Do you want to play a game?”

Text based games on main frames.

The influence could be the parallel time lines of the rise of swords and sorcery and computing in the 1970s and 80s.

And what makes a good hostname anyway? Obscure (not to be confused) unique, memorable?

- Adventure

- Dungeon/Zork

- Rogue/Hack/Nethack

I wonder why no one develops a graphics library for mainframe and makes hobby games out of it.

And of course: the Scary Devil Monastery.

"This is not a rabble of mindless interfaces. These are uruk-hai. Their scans are deep, and their firewalls broad."

We also have GNOME and ORCs [0], and there’s GobLin [1].

[0] https://lwn.net/Articles/728339/

[1] https://distro.misiones.gob.ar/goblin/bienvenida/en-index.ht...

Stanford Artificial Intelligence Lab

Older than Stanford Computer Science department

SAIL was the Stanford AI Lab. Marvin Minsky and John McCarthy founded the MIT AI lab; John was recruited by Stanford and started a lab there. It was in the lovely DC Powers building in the foothills -- despite what you would think "DC Powers" was actually someone's name!

It also had a PDP-10 and a TV system but of course it's own implementation of both the TV hardware and the PDP-10 O/S (WAITS)

I can't find a picture of the building unfortunately -- it was really cool.

Are these some pictures You were looking for?

http://infolab.stanford.edu/pub/voy/museum/pictures/AIlab/

Search the directory for DCPowers.

I wasn't looking for a specific photo, but yes, those are a good find

Interesting, only in the anecdotal sense, that in today's world of incessant "AI" hype, a top commenter on HN would not have encountered the SAIL acronym.

Maybe Stanford is not as important as it thinks.

I was at Stanford for nearly 10 years. Not formally in CS or AI, granted, but SAIL doesn't ring a bell for me. Maybe it's just not that well known.

> processes come from executable files in ELF format and have debug infos in DWARF and it is quite the fantasy menagerie

We don’t call new processes spawn for nothing!

I always assumed it was Maxwell's demon [1]. The little hypothetical creature that exists to manage the state of a system to maintain order is a fitting description.

[1] https://en.m.wikipedia.org/wiki/Maxwell's_demon

Both of the popular stackoverflow answers seem to agree, at least to some degree. Which makes a lot of sense to me -- Maxwell's daemon lives right in the physics/information theory zone, and close enough to signal theory stuff that, while it isn't required reading, any number of engineering students might bump into it.

The story of the etymology is one thing, but the common understanding of what "daemon" is, outside of computing, speaks to the inherent volatility and evolution of meanings in the English language. If you check Webster's dictionary[1] you'll see that (1) it's merely an alternative spelling of "demon" and (2) it's primary definition is that of 'an evil spirit.' It's not until the 5th option that a reference to computing is mentioned: https://www.merriam-webster.com/dictionary/daemon

[1] Noah Webster created his first dictionary as a resource to use alongside the King James Bible because he had observed, in his lifetime, the drift of the meanings of words to the point that what the words meant when the KJV was written were being understood differently. His goal was for Bible scholars to have a fixed point of reference to what the meaning of words should be when reading and interpreting the Bible. We can see this in the US Constitution where the common understanding of the word "regulate" in the late 1700s means to "to make regular" -- which in modern language would be rendered as un-regulated. Reading the Interstate Commerce Clause with the understanding of the words as they were used at the time gives almost an opposite understanding to what was meant at the time.

> Reading the Interstate Commerce Clause with the understanding of the words as they were used at the time gives almost an opposite understanding to what was meant at the time.

I think this has a typo?

There's no compilre on the "reply" button and I'm a programmer... so typoes are possible.

Which word(s)?

> Reading the Interstate Commerce Clause with the understanding of the words as they were used at the time gives almost an opposite understanding to what was meant at the time.

Reading it with today’s meaning might make it opposite if what was meant at the time. Or reading it with the original meaning might make it opposite of what it would mean today.

But reading with the older meaning shouldn’t make it the opposite if what was meant at the (older) time.

Nothing wrong with typos, hope I didn't come off that way. jagged-chisel got the gist of my idea. Was just curious if it was intentional or not

> (1) it's merely an alternative spelling of "demon"

Combine this with the BSD Daemon illustration [0] and you've got my personal idea of where daemon came from.

Which was shattered just last week when I attended a conference presentation by a jungian analyst who definitely did not use daemon in the sense of demon.

That was a weird and frustrating moment.

0. https://en.m.wikipedia.org/wiki/BSD_Daemon

How did the jungian analyst use daemon?

Apocryphal encounter when Texas rednecks [sic] learned about the BSD mascot:

https://unixsage.com/humor/daemonsintexas

> Native: And what kind of football team has the devil as a mascot?

> Me: Oh, it’s not a team. It’s an operating – uh, a kind of computer.

Branagan could have short-circuited all of this if she had only known (remembered?) Arizona. [0]

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

"To make regular" is to make things the same/constant/even. Are you saying that in the late 1700s "to make regular" meant making things uneven?

My understanding is that "regulated" as used in the founding documents was basically synonymous with "ordered" or "organized". It's the opposite of "dysfunctional" or "chaotic", but it doesn't imply any particular authority structure.

Here's a dictionary[1] from 1708 defining "To REGULATE" as:

> to govern, direct, or guide; to frame or square; to determine or decide.

There are also entries for "REGULATION" (the Act of regulating) and "REGULATOR" (a Person that regulates or directs).

Here's another dictionary[2] from 1768 defining "To REGULATE" as:

> To adjust by rule or method. To direct.

A plain reading of the US Constitution[3] paying close attention to the use of "regulate" plainly contradicts this interpretation of the word.

[1]https://books.google.com/books?id=t01gAAAAcAAJ&printsec=fron...

[2]https://books.google.com/books?id=bXsCAAAAQAAJ&pg=PP9#v=onep...

[3]https://constitutioncenter.org/interactive-constitution/full...

Reading from your second link, Samuel Johnson's Dictionary of the English Language:

> ORDER. s [ordo, Lat.] 1. Method; regular disposition. 2. Established process. 3. Proper state. 4. Regularity; settled mode. 5. Mandate; precept; command. 6. Rule; regulation. 7. Regular government. [8-12: other senses not obviously relevant here] 13. Measures; care. [14: a sense specific to architecture]

(Italics are my emphasis; I've highlighted every word related to "regulated". "s" after the headword seems to indicate a noun.)

> To ORDER, v. a. [from the noun/] 1. To regulate; to adjust; to manage; to conduct. 2. To manage; to procure. 3. To methodize; to depose fitly. 4. To direct, to command. [5: one other irrelevant sense]

Here we see that "regulate", "adjust", "manage", and "conduct" are all synonymous. We see that the concept of "order" has special senses relating to organization and to command, as is also true in English today. The only mention of the verb "regulate", which I discussed in my first comment, clearly indicates that it is used in the sense of organizing. There is a special application of this or a related concept to government, "regular" government, noted in sense 7.

"To REGULATE" is given two senses:

> To REGULATE. v. a. [regula, Lat.] 1. To adjust by rule or method. 2. To direct.

And REGULATION is:

> REGULATION. s. [from regulate.] 1. The act of regulating. 2. Method; the effect of regulation.

This is interesting in that it indicates that (in sense 2) "regulation" is a state of being, like "order", resulting from being regulated. The modern sense of the word, referring to a formal rule, is not listed in this dictionary. However, that appears to be an oversight by Johnson: here are the relevant senses of REGULAR:

> REGULAR. a. [regularis, Lat.] 1. Agreeable to rule; consistent with the mode prescribed. 2. Governed by strict regulations.

I don't see an interpretation of "regulations" as used here other than "formal rules". This is what was meant by "regular government" or "order" - a government itself subject to strict regulations. Sense 1 is so unclearly worded that I've included it in case it's relevant, but I believe it's just the sense in which "played" is the regular past tense of "play", while "went" is the irregular past tense of "go". This "regular" means "conforming to a rule".

And:

> REGULARITY. s. [regularité, Fr.] 1. Agreeableness to rule. 2. Method; certain order.

What does "rule" mean?

> RULE. s. [regula, Lat.] 1. Government; empire; sway; supreme command. [2. a ruler] 3. Canon; precept by which the thoughts or actions are directed. 4. Regularity; propriety of behaviour.

> To RULE. v. a. [from the noun.] 1. To govern; to controll; to manage with power and authority 2. To manage. 3. To settle as by rule.

One word has come up in almost everything we've looked up:

> METHOD. s. [methode, Fr. μέθοδ@.] The placing of several things, or performing several operations in the most convenient order.

This isn't especially clear, but there are related entries:

> METHODICAL. a. [methodique, Fr. from method.] Ranged or proceeding in due or just order.

> To METHODISE v. a. [from method.] To regulate; to dispose in order.

This states explicitly that to regulate something is to bring order to it. The sense survives in modern English in a related word, regularize; that distinction was not drawn in the eighteenth century.

I'll continue in a reply to my own comment, because I'm about to shift topics.

My reading of all of that is that "regulated" would have absolutely been understood as meaning "subject to regulation by proper authority" at the time of the Constitution's drafting. The Framers would have considered the states to be the proper authority, and thus the Second Amendment prohibits Congress from interfering in the states' ability to organize militias.

To recap: I made this claim:

>> "regulated" as used in the founding documents was basically synonymous with "ordered" or "organized". It's the opposite of "dysfunctional" or "chaotic", but it doesn't imply any particular authority structure.

With reference to the first part of my comment (above), I'm going to treat Samuel Johnson's dictionary as having conclusively established that this claim is correct in a general sense; "regulated" bears exactly this meaning some of the time. The dictionary, however, isn't really able to speak to the question of "what does the word mean as it's used in the founding documents?".

You have made this counterclaim:

> A plain reading of the US Constitution[3] paying close attention to the use of "regulate" plainly contradicts this interpretation of the word.

I say that this is false. There are three uses of the word "regulate" in the linked text of the Constitution:

> The Congress shall have Power [...] To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes;

> The Congress shall have Power [...] To coin Money, regulate the Value thereof, of and foreign Coin, and fix the Standard of Weights and Measures;

> A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.

There is only one use of "regulated". It is open to interpretation whether it is simply a participial form of "regulate" or whether it has developed a special sense.

Let's look at some other founding documents:

https://founders.archives.gov/documents/Washington/03-08-02-... is a letter from George Washington, in which he complains about the "irregular and disjointed State of the Militia", and states that the solution should be "a well regulated Militia law". In this letter, we can note the following aspects of usage:

1. A "well regulated Militia" is the opposite of an "irregular and disjointed Militia".

2. The effect of a "well regulated Militia law" is to "reduce [the Militia] to some order". "Order" is of course, as we learned from Samuel Johnson, synonymous with "regulation", so this is an unsurprising usage. But George Washington's use of synonyms lets us be fairly precise about what he means - his meaning lies in the space indicated by all of his synonymous words.

3. The specific method by which the militia should become "well regulated" is not to impose any additional rules on the militia, but to provide it with "good Officers" who would set a better example than the current officers, and "protect, instead of distressing the Inhabitants [of the province]".

https://quod.lib.umich.edu/e/evans/N25292.0001.001/1:2?rgn=d... is a speech from 1798, after the Bill of Rights had been written, in which George Clinton nevertheless helps us understand the contemporary meaning of the second amendment by referring to it with a synonym rather than a direct quotation:

> THE means of national defence should rest in the body of the people. A well organized militia is the only safe bulwark of a free people, competent on all occasions to repel invasion and suppress insurrection. Standing armies are not only expensive but dangerous to the liberties of the state. In republics every citizen should be a soldier.

It seems likely that George Clinton understood "a well regulated militia" to mean "a well organized militia", since he used those exact words to refer to the concept.

So no, I can't conclude that a close reading of the Constitution leads to the inescapable conclusion that "regulated" as used in the Constitution cannot be interpreted as meaning "ordered" or "organized"; the opposite is true.

In the Washington letter, he clearly is referring to "regulating" the militia by modifying the law under which it is organized. The primary concern he and Livingston express about the existing New Jersey militia law is that it allows "[a] Man capable of bearing Arms [to] buy off their Service by a trifling Sum" in lieu of service, as well as "suffering an appeal to lay before a Court of Judicature". I don't think this endorsement of mandatory conscription of armed men without option of appeal to the court tends toward supporting a libertarian view of the Second Amendment, nor does it support the case that "regulated" does not imply government regulation.

You can also refer to "organizing" or "reforming" the militia, or anything else, by modifying the law under which it is organized. But that fact doesn't even suggest that "organize" or "reform" refer to the act of issuing laws. And it is clear that being "regulated" [Washington does not refer to "regulating" at all, only to the state of being regulated] also doesn't refer to that act, as Washington uses it. It refers to the opposite of being "irregular".

The argument "I don't like George Washington's ideas about militia reform -> 'regulated' means governed by a body of formal laws" isn't something I would be proud to advance.

If what you're trying to argue here is that the Second Amendment does not _obligate_ states to pass militia laws, I would have to agree, although that doesn't seem to me to be a point of controversy. Otherwise, this is beginning to feel like Calvinball.

It is clear from your own sources that the Framers would have considered state legislation to be an appropriate vehicle to attain the goal of a "well-regulated militia". The original question here is whether the Constitution's usage of "regulate" or "regulated" differs in any meaningful way from the modern understanding of those words, and you have convincingly demonstrated that it does not.

> It is clear from your own sources that the Framers would have considered state legislation to be an appropriate vehicle to attain the goal of a "well-regulated militia".

Of course? But what conclusion are you trying to draw from that statement?

The second amendment states that the right of the people to keep and bear arms shall not be infringed. George Washington proposes a "well regulated militia law" which obligates everyone who can bear arms to serve when called upon. In what way does that interfere with the right to bear arms?

> The original question here is whether the Constitution's usage of "regulate" or "regulated" differs in any meaningful way from the modern understanding of those words, and you have convincingly demonstrated that it does not.

But here, you're not actually reading my comment(s). We can see that the Constitution's usage of "regulated" is considered synonymous with "organized" and opposite to "irregular" and "disjointed". No such usage would make sense given the modern understanding of the words. But that is how the words were used at the time.

> We can see that the Constitution's usage of "regulated" is considered synonymous with "organized" and opposite to "irregular" and "disjointed". No such usage would make sense given the modern understanding of the words. But that is how the words were used at the time.

I don't understand this at all.

Modern usage of "regulated" can mean either "organized" or "under law". Indeed, in many cases it is just a difference of degree. Perhaps the more common usage is "regular" (as in "regular season"), but it's not some massive difference in meaning where they mean the opposite.

Here, I'll read off hits for "regulated" in COCA (Corpus of Colloquial American [English], https://www.english-corpora.org/coca/ ). There are more than 7000; I'll just take them top down until I lose energy.

> German business owners are more than eager to escape their high-wage heavily regulated union environment

> It is also regulated and controlled by a supranational regulatory body (the WTO)

> and Finland where cell phones, cable TV, other services and products are better regulated and at far lower costs

> "yet they still aren't regulated as hazardous waste. If we really want 'clean coal', we need

> Making the mark-up as opaque as possible. When you are forced or regulated into revealing the mark-up

> fees that get taxed in NY and (hello?) regulated under NY and US laws.

> an aspect of human behavior that is regulated by society through the operation of justice

> aspects of human behavior that are totally regulated by society

> but one of the largest sources, at cook, will be regulated by a [traffic] light

> Greenstein said the exchanges as envisioned in the federal law are too regulated, likely raising costs and limiting choice.

> The fact is, "fair markets" (free markets are not fair, and fair markets are not free) must be carefully and dutifully regulated. And over most of the US's past, it had extensive and good regulation.

> You look at the financial industry as the least regulated. I disagree. The problem is that money is the most regulated of all things. We have legal tender laws and taxation

> what we call "free markets" are actually highly protectionist, regimented, regulated, and designed to undermine competition

> Natural gas supplies were regulated almost out of existence until the government

> But, owning personal arms IS being regulated and pretty much has been from the inception of the Second Amendment

> contraception was also illegal in Ireland until 1980 (!) and then were strongly regulated (yes, your doctor had to prescribe you a condom)

...

> Marijuana should be regulated and taxed by the states

> City Council outlawed corporate signage on downtown buildings. We REGULATED, in other words

OK, I'm losing energy. One page of results has 101 entries. I'll code the first page:

Referring to government (or supragovernmental [the WTO] or divine) law: 80

Referring to mechanical control: 11 ("needs a regulated voltage to run")

Referring to some other type of control: 4 ("Sex [is] an aspect of human behavior that is regulated by society")

Referring to a formal rule that is not imposed by any body with enforcement authority [e.g. an agreement on date nights between a husband and wife]: 2

Typo: 1 ("what legislation is necessary to regulated GMOS")

Unclear: 2

That last category, "unclear", does contain an example that might be construed as meaning "orderly":

> I do not subscribe to the notion that we must regulate our population because there are too many people on Earth. Let Nature take its course and allow the population to get regulated through the market forces in a free association of human beings in society.

The source blog is gone, so I'm not able to look into it in more detail than that. This might be an example of using "regulated" to mean "orderly", but that looks extremely unlikely; it's more likely that it refers to the general existence of rules, in this case a code of conduct. (Which would place it in the "other type of control" category.)

It is in fact fair to say that the usage of "regulated" seen in the second amendment has entirely disappeared from modern usage, and that's why people don't understand it and are willing to deny that using the word that way was even a possibility. It is not a possibility now, but it was then.

The problem here is that you have not proven in any way that the Second Amendment's use of "well-regulated" does not refer to "government law" or "some other type of control". If you're suggesting that they expected militias to be self-organizing, you need to present at least one instance of a self-organized militia being referred to as "regulated". The evidence you've presented so far makes it clear that contemporary thought understood militias to be organized by law.

If you could find a single example of a militia being "regulated" by anything other than a legal mandate, I have to assume you'd have presented it by now. 1 gazillion points to Calvin; game over.

> The problem here is that you have not proven in any way that the Second Amendment's use of "well-regulated" does not refer to "government law" or "some other type of control".

You're really set on ignoring the George Clinton speech, huh?

The second amendment begins like this:

> A well-regulated militia being necessary to the security of a free state...

This usage is obsolete; translated into modern English, the same clause would read:

> A well-functioning militia being necessary to the security of a free state...

And it is obvious that this does not refer to the existence of laws, because that's just not part of the meaning of the word. What do you think needs to be proven? Why?

Anyway, here is another letter from George Washington: https://founders.archives.gov/documents/Washington/03-07-02-...

> I am unacquainted with the Extent of Your Works and consequently of the Number of Men necessary to man them. If your present Numbers should be insufficient for that purpose, I would then by all Means advise Your making up the Deficiency out of the best regulated and equipped Militia. [4]

> [4] The draft reads: “out of the best regulated Militia that could be got.”

This is pretty explicit: George Washington anticipates that Philip Schuyler may not have enough troops under his command to achieve his objectives. He recommends mustering troops from the local militia. But he is more detailed than that: he envisions ranking the local militia along a scale of how "regulated" they are, and choosing the men who are best regulated. All of these men are subject to the same laws, but they differ in their degree of regulation.

As I said: "Perhaps the more common usage is "regular" (as in "regular season"), but it's not some massive difference in meaning where they mean the opposite."

Nevertheless, from the same page you referenced here are some where it means "organized":

> future studies may show that the P2X 7 R could be regulated by a range of ligands

> A possible mechanism for this is that it is regulated epigenetically

> while medical staff monitored his condition and regulated his anti - rejection drug

> Let Nature take its course and allow the population to get regulated through the market forces in a free association of human beings in society.

> This is why the regulated evenings - at least this way I know he will be getting one evening

> the practice of Orthodox Judaism, whose followers maintain a life strictly regulated by their interpretation of holy scripture

> a person's conduct is regulated for the most part by socially approved dispositions that are physically realized in the brain

> But if a man's efficiency is not guided and regulated by a moral sense, then the more efficient he is the worse he is

I dunno - I could keep going too.

> Nevertheless, from the same page you referenced here are some where it means "organized":

But those aren't examples where the word means "organized". Except perhaps the one I already called out, but again, probably not even that one.

> We can see this in the US Constitution where the common understanding of the word "regulate" in the late 1700s means to "to make regular" -- which in modern language would be rendered as un-regulated. Reading the Interstate Commerce Clause with the understanding of the words as they were used at the time gives almost an opposite understanding to what was meant at the time.

I've seen this rhetoric used somewhat frequently by 2A advocates but it's completely bogus. The word has had the same common meaning in English since at least the early 1600s.[1] Even in Latin the word regula (from which "regulate" originated) was used in the context of governing, control and regulation as we know it today. As in regulae iuris which was literally, the regulation of interpreting the law.[2]

I've yet to see anyone present any argument at all, let alone a reasonable one, that would even form the basis of such a theory for the "changed" meaning. It's always just stated as a fact when, in fact, it seems to be a completely fabricated talking point.

[1]https://www.etymonline.com/word/regulate

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

Thank you.

Backing you up with some more sources: Webster's 1828 dictionary [1], which the grandparent cited, does not have the "make regular" definition, but rather definitions consistent with modern usage:

>1. To adjust by rule, method or established mode; as, to regulate weights and measures; to regulate the assize of bread; to regulate our moral conduct by the laws of God and of society; to regulate our manners by the customary forms.

>2. To put in good order; as, to regulate the disordered state of a nation or its finances.

>3. To subject to rules or restrictions; as, to regulate trade; to regulate diet.

Samuel Johnson's 1755 dictionary [2] has much briefer definitions, but also defines regulate in an interventionist, controlling sense:

>1. To adjust by rule or method.

>2. To direct.

There is absolutely no basis for the claim that "regulate" used to mean something like the opposite of what it now means.

[1] https://webstersdictionary1828.com/Dictionary/regulate

[2] https://johnsonsdictionaryonline.com/views/search.php?term=r...

Idioms are not in general the sum of their parts.

I sometimes see a similar thing when people are arguing. They sometimes use a different current definition of the same word, and don't even realize. Sometimes I would guess it's intentional.

I can't think of a good example, but I'm reminded of when climate science emails were and leaked and one of the emails had someone applying a 'hack' to the data. One side decided they were doing something underhanded. But hack I believe the use of the term was 'clever method'.

Anyway, I see this often. People are arguing over the definition of words, but they think they're arguing about something else more profound.

A huge part of "culture war", uh, discourse, seems to be driven by inability to agree on common terms and a complete ignorance of how labels relate to reality (so most participants don't even act like they know it's a problem). A better understanding of how slippery identity and "is" are, even for everyday things that seem very certain, would really improve communication there, I think. I'm thinking not just of the extremely obvious topic related to this (trans issues) but also things like abortion, where arguing over essentially arbitrary definitions by treating those definitions as per se important almost entirely crowds out substantial discussion.

What a great point. I've been struggling with the epistemological implications of the copula for the last few months. The existence of the copula and its use seems to imply a shared objective reality, when it's perhaps more explainatory to label the copula as a reality constructing mechanism.

Either way it's use as a universal quantifier eschews nuance. Patterns like "A is B" necessarily lead to differences in reality that are easily resolved when phrased like "I percieve that A is B". Copula-induced fights are ubiquitous.

> Noah Webster created his first dictionary as a resource to use alongside the King James Bible because he had observed, in his lifetime, the drift of the meanings of words to the point that what the words meant when the KJV was written were being understood differently.

This sounds off; the KJV was written in an intentionally archaic style, which guarantees that, at the time of writing, popular understanding of some of the usage would have already diverged from the meaning as written. That was a goal of the project, which makes it a strange "problem" to address with a dictionary.

> This sounds off; the KJV was written in an intentionally archaic style, which guarantees that, at the time of writing, popular understanding of some of the usage would have already diverged from the meaning as written.

This is frequently stated but when it gets down to brass tacks seems to be somewhat exaggerated as the source of all archaisms from the POV of later readers. From what I can tell, while some things (like second person pronoun use) where archaic at the time of writing (and egregiously bad choice of archaism because they created distinctions not made into the source material by using a distinction no longer currently in use in English, either), many others were things that were in transition at the time of writing and which didn't reflect later (and also, in some cases, earlier) common usage (thereof to avoid possessive pronouns as gendered ones relating to inanimate objects were on the way out, but “its” was still not firmly caught on).

But I didn't say every archaism in the KJV was due to an intentional style choice. I said the KJV was intentionally in an archaic style. That fact makes it strange that "this is archaic usage" would be considered a problem in the KJV, regardless of whether the particular archaism was written to be an archaism or just developed into one.

> gregiously bad choice of archaism because they created distinctions not made into the source material by using a distinction no longer currently in use in English, either

All the source languages for the Bible have a second person singular/plural distinction. Or did you mean something else?

Daemon is also simply the original latin spelling of the word.

In this context its interesting to note that the original meaning of the greek daímōn and latin daemon had no intrinsic connotation of evil. It simply was a kind of "spirit", a non-corporeal entity that somehow interacted with the human plain of existence.

Looked at it this way, the use of the word in *nix systems makes perfect sense: A non-corporeal, neutral entity, that interacts with the world at large, goes mostly unseen, and its interactions are (for most daemons) close to userland.

Note the demon of the Maxwell demon was actually William Thompson's (Lord Kelvin) term. Maxwell himself used valve and doorkeeper.

There are earlier references elsewhere, but the oldest Internet RFC I could find mention of it is RFC114 (Apr 1971), a Multics related doc called "A FILE TRANSFER PROTOCOL"

https://datatracker.ietf.org/doc/html/rfc114

An excerpt:

The cooperating processes may be "daemon" processes which "listen" to agreed-upon sockets, and follow the initial connection protocol much in the same way as a "logger" does.

Related to MIT's "Project MAC", with the same Professor Fernando José Corbató involved as the Stack Exchange answer.

Surely the right question to ask is what is the origin of computing with regards to the mythological tale of "daemons". The Greeks built a mechanical computer called the antikythera mechanism, and tales of clockwork automotons have existed for millenia. So my argument would be that the concept of a "daemon" doing your computational work for you entirely precedes modern computation.

Edit: The second response to the stack exchange question consists of pretty much this argument, and the Wikipedia page [1] has more information!

[1] https://en.wikipedia.org/wiki/Daemon_%28computing%29#Etymolo...

I like this version the best regardless of what the originators thought

> As early as Homer, Greeks were imagining robotic servants, animated statues, and even ancient versions of Artificial Intelligence, while in Indian legend, Buddha’s precious relics were defended by robot warriors copied from Greco-Roman designs for real automata. Mythic automata appear in tales about Jason and the Argonauts, Medea, Daedalus, Prometheus, and Pandora, and many of these machines are described as being built with the same materials and methods that human artisans used to make tools and statues. And, indeed, many sophisticated animated devices were actually built in antiquity, reaching a climax with the creation of a host of automata in the ancient city of learning, Alexandria, the original Silicon Valley.

Gods and Robots: Myths, Machines, and Ancient Dreams of Technology - Adrienne Mayor

https://classics.stanford.edu/publications/gods-and-robots-m...

Key point from the answers: in original usage, daemons are spirits and can be good, bad, or neutral. As an illustration, "eudaimonia", good spirits, is the Greek word for happiness.

Huh, interesting. I always thought it was because, like me, D&D nerds were the ones in charge of naming things, so their gaming vocabulary trickled into their working vocabulary.

I'm not sure if this is more or less nerdy than my believed version, TBH.

EDIT: Forgot a very important word.... I need to have some caffeine =)

Does anyone else always think of Socrates' Daemon?

I'd just assumed it was from that. Greek-tradition minor deities and helper spirits. Like "genius".

Yes. Socrates' daimon listened in the background and would warn or forbid him from doing certain things.

I strongly thing about it!

Socrates's demon, Descartes's demon, Laplace's demon, Maxwell's demon, and there are many more, it's a common notion

Interesting! Now what is the correct pronunciation…

Day-muhn?

My colleagues all seem to say it differently, and I’ve always been curious which one of us sounds the most foolish.

To me it's daemon as in encyclopaedia, or paediatric

According to the dictionary deamon and demon are the same word with the former an archaic spelling. How do you say the latter?

> deamon

Is that a typo? If not then I see how those would be pronounced the same, but daemon in American English would usually be pronounced "Day-mon". This also helps to distinguish between the long running process and an evil entity, which is good when talking to less technically minded folks.

Yes, a typo.

That should be what they call a deacon in the Church of Satan.

According to Wikipedia it can be either dee-mun (demon) or day-mun

I've heard both in my personal life from different people

https://en.m.wikipedia.org/wiki/Daemon_(computing)

Wikipedia is an encyclopaedia, the answer is closer than you think ;)

Whoever it was that decided “simplified” spellings were an improvement didn’t take into account how difficult they’d make pronunciation. I often hear highly educated Americans coming up with very strange ways to pronounce words, well beyond the normal realms of accent and variation, that are clearly the product of this. English spelling is bad enough but cutting out the hints to a word’s origin makes things worse, not better.

That is how I have always said it and heard it said

In the audiobook for Daniel Suarez's Daemon, Jeff Gurner pronounces it dee-mun. I changed the way I pronounced it after listening to that book. Most people I know pronounce it day-mun, though.

I'm listening to it now. Quite clever, and a real page-turner! The writing and story-telling are awkward, and its plot holes are numerous, but fun concept for the Hacker News crowd.

I pronounce like Day-Mon (those influential and wealthy Ferengi).

Or De-Men when feeling less nerdy.

Exhibit A: "Daemon" is just an archaic spelling of "demon" and we already have a common pronunciation for that.

Exhibit B: In Old English, "æ" was a vowel that did not survive to modern English. There is no direct equivalent sound or letter for it today. Other words that contiained æ have generally taken the long E pronunciation instead, e.g. encylopaedia -> encyclopedia and aether -> ether.

So people can say what they want but in my opinion, the "dee-mon" pronunciation is the most correct. :)

In European languages, æ eventually evolved into ä.

German, Swedish, Finnish: Ä

Norwegian, Danish, French, Icelandic, Faroese: Æ.

According to Wikipedia: “Ä occurs as an independent letter in Finnish, Swedish, Skolt Sami, Karelian, Estonian, Luxembourgish, North Frisian, Saterlandic, Emiliano-Romagnolo, Rotuman, Slovak, Tatar, Gagauz, German, and Turkmen”

Æ is, according to Wikipedia, used in Latin, Icelandic, Danish, Norwegian, Faroese, Ossetic, Kawésqar, and Yaghan. (It is not used in French or English except in loanwords from Latin.)

While it's true that today we would pronounce de- and dae- each as /dee/ without differentiation, TFA points out an interesting notion:

> By the late 16th century, the general supernatural meaning was being distinguished with the spelling daemon, while the evil meaning remained with demon.

Given that both spellings were used simultaneously back then for the goal of differentiation that transcends mere archaic-vs-modern, I have to wonder if there was verbal differentiation accompanying it as well. I suppose this depends on a more precise estimate of æ's deprecation through the transition from Old to Middle to Early Modern to Modern English (specifically, whether its original /æ/ [0] sound persisted into Early Modern or was already /ee/ by then).

[0] Which, confusingly, has several possibilities [1]. Since we're talking about the 16th century though, we might go with the sound it also represents in IPA, as in ash, fan, happy, last, etc. although I can't say I've ever heard anyone utter /dam ən/...

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

I say day-mon because it's less alarming for non-technical people and less ambiguous for technical people.

But, in accord with the robustness principle, I'm happy to accept any pronunciation.

I say day-mon because it's less alarming for non-technical people and less ambiguous for technical people.

Same. Even though it's probably technically incorrect in some pedantic sense, this pronunciation just makes more sense to me.

In the interest of starting a proper flamewar, I propose that we also discuss the pronunciation of 'sudo' and 'sysctl'.

> the pronunciation of 'sudo'

SOO-dough (like pseudo)

> and 'sysctl'.

see-ISSSSS-cuttle (obviously!!)

soo-doo

Siss cee tee ell

I will die on this hill

I'm guessing you're also a my-ess-queue-el person, then? Lack of vowels means discrete pronunciation of letters? How about the ctrl key or fn key?

Oddly, no strong feelings about those, but I am a my-ess-queue-el person. I can't claim to be consistent, but filling in vowel-less acronyms feels wrong. I think that's the difference - ctrl and fn are abbreviations, not acronyms. That might be the first time I've expressed it that way, actually. TIL, chiefly about my own thought processes.

I do believe SQL is not an acronym, but an initialism, precisely when it's spoken as you say it: initials rather than a word. It's when the string of letters is pronounced as a word that "acronym" is accurate. So SQL is either, and thus both...

But yeah, the filling in is odd unless, as in the case of ctl/ctrl, knowledge of the original guides us into lossless compression.

sequel, but the margin was too narrow

sudo = "s.u. do" sysctl = "sys control"

sudo: "sue dough"

sysctl: "sys control"

It's pronounced "GIF."

In seriousness it's pronounced DEE-mon, but you'll never get consensus on this, so best just to say it the way you want to say it and leave it alone when someone else says it differently.

DAEMON (day'mun, dee'mun) [archaic form of "demon", which has slightly different connotations (q.v.)] n. A program which is not invoked explicitly, but which lays dormant waiting for some condition(s) to occur. The idea is that the perpetrator of the condition need not be aware that a daemon is lurking (though often a program will commit an action only because it knows that it will implicitly invoke a daemon). For example, writing a file on the lpt spooler's directory will invoke the spooling daemon, which prints the file. The advantage is that programs which want (in this example) files printed need not compete for access to the lpt. They simply enter their implicit requests and let the daemon decide what to do with them. Daemons are usually spawned automatically by the system, and may either live forever or be regenerated at intervals. Usage: DAEMON and DEMON (q.v.) are often used interchangeably, but seem to have distinct connotations. DAEMON was introduced to computing by CTSS people (who pronounced it dee'mon) and used it to refer to what is now called a DRAGON or PHANTOM (q.v.). The meaning and pronunciation have drifted, and we think this glossary reflects current usage.

https://www.dourish.com/goodies/jargon.html

One meaning of "daemon" is that it's a supernatural being that is kind of halfway between a human and a god.

Daemons often run with privilege, and so are (god-like), but are not the operating system itself (god). Daemons can do things on behalf of the human user which no program could do that the user could write and execute under his or her privilege level.

Maxwell's demon is really a demon in the evil sense because it perpetrates something "wrong": it causes all the gas to move from one compartment to another, thereby wrecking god-given thermodynamics.

> thereby wrecking god-given thermodynamics

I once got into an argument with a creationist who insisted that evolution was impossible because it contradicted the second law of thermodynamics.

I’m guessing he’d view a refrigerator as the devil’s work.

My research shows that the etymology of daemon is based on the concept similar to angels and devas where there existed “messengers” of divinity/god.

Which does connect similarly and semantically to background processes.

The earliest reference to daemon in computing that I could find is from 1962 (a year before Corbato's use in the stackexchange discussion). The Daemon spacecraft computer (Data Adaptive Evaluator and Monitor) was named after the "deity of Greek mythology who served as a guardian over the lives and fortunes of men."

This computer system is interesting because it proposed using a TV-type monitor to display all the significant data in the space vehicle cockpit, way back in 1962. It took 35 years for this to become a reality with the Space Shuttle's MEDS display (1998).

This was published in Aviation Week: https://archive.org/details/Aviation_Week_1962-05-21/page/n3...

Interesting addendum on the difference between kernel-space daemons and user-space daemons:

https://unix.stackexchange.com/questions/266434/are-kernel-t...

Someone please help me finish a joke: what's the difference between a speed demon, and a speed daemon? (I've had this premise for a while, but I keep crit failing the charisma check to make a good punchline.)

How about something like

a speed demon sprints while a speed daemon sprintfs

Another use is the "printers daemon" [0] which sits on the shoulder, and points out (or not) the typos as you compose the line of type.

There is a rather nice door lintel at the Queensland Government Printery which has a printers daemon leering at the public as they pass.

given some flexibility, you can see linkage to "bugs" in "typos"

[0] https://www.timetravelclub.com.au/2015/07/devils.html

Daemon is also how it's spelled in Mary Shelley's Frankenstein. I always wondered if there was some connection to its use in computing. She uses that spelling of the word to imply that the "monster" is not evil. I can't find a good source, but there are plenty of things which have similar jists to this: "A daemon was originally a spirit somewhere between gods and men. Later, with the spelling ‘demon', it acquired the meaning of a devil or evil spirit."

As a side, always loved the hacker adoption of daemon, and usage in hacker lexicon. Also in the show Mr.Robot, I think they use it at times.

Here is the usage of it - https://mrrobot.fandom.com/wiki/Eps1.3_da3m0ns.mp4?li_source...

Previously:

https://news.ycombinator.com/item?id=2691752

https://news.ycombinator.com/item?id=16299583

On a somewhat related note, I distinctly recall in the mid 90s running Doom for the first time (DOS 6 I think?), and its text-based load screen that initialised everything said “refresh daemon”.

At the time I thought it was something to do with refreshing the baddies in the game! ^_^

I can confirm Corby's statement and more so: many of the daeemons running on the ITS systems had names from Zelazny's Amber. And of course the DM machine was the source of Zork.

UNIVAC once called background tasks "parasites". This was changed to "symbiont" for marketing reasons, and the change was visible as a paste job in the manuals.

Horse's mouth (Fernando J. Corbato):

https://ei.cs.vt.edu/~history/Daemon.html

anecdote time - Months ago this same question arose in my head reading something about graphs of cause and effects over other dimension (like time), If I remember well, they mentioned something about information pockets and how you can take advantage of them to change the course of events if you are an almighty observer, then searching about it I read something about daemons being things that changes things and move through those pockets.

I seem to recall it stems from the Ancient Greek word “eudaimonia” = good, helpful spirit. It is a core part of modern ethics (see Aristotle).

Also see https://en.wikipedia.org/wiki/Eudaimonia

Amodern variant would be "microservice"

I wonder why the preference of the less common spelling.

Twas a great sci-fi book called daemon

keep the daemons alive and keep the zombies out.

Sysadmin

Reply on news.ycombinator.com