Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

atomsonly.news

Post reply on HN
Print view View original

181–190 of 209 posts

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#181

Post by Neikius »

ForOldHack wrote: post #180
wwarner wrote: post #50

In 1676 Roemer estimated the speed of light by timing the orbit of Jupiter’s moon Io, noting that as the Earth approached Jupiter, Io emerged from behind Jupiter a little earlier every day, and as the Earth traveled away from Jupiter it appeared a little later every day, with the time of day varying by 22 minutes over a year. Knowing the difference between the two distances, he reckoned that light travels that distan…

The speed of light, because it is so fast, is the most precise physical constant known : 299 792 458 m / s. Less than 7/1000ths away from 300,000,000 m / s. I am not going to sweat this in the least. So light travels only 0.3 m / nano second, or 11.802 inches. https://www.youtube.com/watch?v=9eyFDBPk4Yw

Armed with this fact thinking about electronic devices. How the front of the signal travels and how suddenly the distance matters. Was blown away when I first thought about this.

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#182

Post by glial »

ForOldHack wrote: post #180
wwarner wrote: post #50

In 1676 Roemer estimated the speed of light by timing the orbit of Jupiter’s moon Io, noting that as the Earth approached Jupiter, Io emerged from behind Jupiter a little earlier every day, and as the Earth traveled away from Jupiter it appeared a little later every day, with the time of day varying by 22 minutes over a year. Knowing the difference between the two distances, he reckoned that light travels that distan…

The speed of light, because it is so fast, is the most precise physical constant known : 299 792 458 m / s. Less than 7/1000ths away from 300,000,000 m / s. I am not going to sweat this in the least. So light travels only 0.3 m / nano second, or 11.802 inches. https://www.youtube.com/watch?v=9eyFDBPk4Yw

Therefore in one cycle of a 3GHz processor, light travels about 4 inches. Wild.

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#183

Post by bialpio »

tech_ken wrote: post #172
zokier wrote: post #162

Earlier quoted context omitted.

Replying to myself, I found 1899 paper which is more explicit on the matter, and shows how Rayleigh was not all that certain about the results: The comparison of the present with former results throws an interesting light upon molecular magnitudes. It has been shown (Proc. Roy. Soc. March 1890) that the thickness of the film of olive-oil calculated as if continuous, which corresponds to the camphor-point, is about 2.…

> is more of a numerological coincidence, the actual estimate that Rayleigh gives is half of that! I definitely thought so too, in the first paper the 1.67 isn't even really his primary guess so definitely some presentist bias in the OP

From the paper:

"The thickness of oil required to take the life out of the camphor movements lies between one and two millionths of a millimetre, and may be estimated with some precision at 1’6 micromillimetre."

Looks like a primary guess to me, even if the table lists more data points.

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#184

Post by superjan »

akarve wrote: post #129
rkagerer wrote: post #89

This is fascinating, but wasn't it still a bit of a conjecture to assume the oil would spread to a minimum thickness of one molecule? Did he have any doubts, like that surface tension might keep it thicker? Or other clues hinting it was indeed a monolayer?

there has to be some missing information on how he found the area of water that fully consumed exactly that amount of oil as it simply doesn’t make sense without that. for instance one can spread a teaspoon of oil over 1, 2, n square meters and at some point the oil goes from m later thick to one to less than one.

Such an experiment was described in a science book I read as a kid. They dissolved oil in alcohol so as to measure a very small quantity of oil, and then a a knitting needle was used to stretch the oil film across a plastic container filled to the edge with water (until the film breaks).

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#185

Post by xhrpost »

glial wrote: post #182
ForOldHack wrote: post #180

Earlier quoted context omitted.

The speed of light, because it is so fast, is the most precise physical constant known : 299 792 458 m / s. Less than 7/1000ths away from 300,000,000 m / s. I am not going to sweat this in the least. So light travels only 0.3 m / nano second, or 11.802 inches. https://www.youtube.com/watch?v=9eyFDBPk4Yw

Therefore in one cycle of a 3GHz processor, light travels about 4 inches. Wild.

Which is maybe part of the reason RAM needs to be so close to the CPU? Granted though most RAM access takes several CPU cycles.

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#186

Post by Brusco_RF »

glial wrote: post #182
ForOldHack wrote: post #180

Earlier quoted context omitted.

The speed of light, because it is so fast, is the most precise physical constant known : 299 792 458 m / s. Less than 7/1000ths away from 300,000,000 m / s. I am not going to sweat this in the least. So light travels only 0.3 m / nano second, or 11.802 inches. https://www.youtube.com/watch?v=9eyFDBPk4Yw

Therefore in one cycle of a 3GHz processor, light travels about 4 inches. Wild.

My favorite riff on this: at a previous job we worked on a 12GHz SERDES, which meant each inch of the cable had a different data bit on it

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#187

Post by cb321 »

hn_throwaway_99 wrote: post #165
danielam wrote: post #128

Earlier quoted context omitted.

Yes, often what is taught is taught in a manner that seems mysterious in origin, as if it were revealed, certain and final, and developing a sensibility like that concerning scientific matters is not good. You could argue that the viability of science as such rests on certain articles of faith, but the particular findings of science themselves are a matter of demonstration, interpretation of demonstration and argumen…

Yes! Not to get too political, but I saw a lot of this during the Covid debates, e.g. "Trust the science!" Noooo! Science is not just something you're supposed to "trust", but something that's supposed to be supported by evidence. Yes, I definitely understand that most people don't have the training and background to understand complex scientific topics, and in some ways we do have to trust the scientific community i…

It isn't like "scientists" (whatever that means) don't also delegate to equipment or hardware manufacturers or (mathema|statis)ticians or compilers (or grad students, LOL). Sure they calibrate & cross-check, but while oil spread-out over water is easy to replicate at home in a few minutes with an oil dropper (maybe even with precocious (maybe pre-)pubescent kids not just Feynman Lectures on Physics Caltech Undergrads [1]), "more interesting/complex questions" usually have "full stacks" that are impractical to fully vet in general (see, e.g. Ken Thompson's _Reflections On Trusting Trust_). Epidemic disease coupled with human behavior is definitely getting into "holy crap complex" territory.

In its barest essence the problem is this - delegation affords so much that it is basically unavoidable, but trust sure is tricky. The https://en.wikipedia.org/wiki/Demarcation_problem - searches for how to demarcate trustworthiness. Sorry to say, but there is a long history of failure to get consensus. It's notably a competitive game and as long as anything has been deemed valuable there have been cheap knock-offs (e.g. Fool's Gold), but things like The News you seem to complain of (e.g. Crichton's Gell-Mann Amnesia [2]) or "conclusions", being more abstract than metal (often the metaphorically concrete), are trickier still to discern reliability.

It may be the single most central (in latitude-long's of WHAAA? coordinates) problem of today's human condition / experience. I think it's a https://en.wikipedia.org/wiki/Wicked_problem BUT the problem applies recursively to advice from anyone about trust/delegation (or about anything else). So, don't trust me. LOL. ;-) I don't think others can really answer these questions for someone. Part of life is learning to live with uncertainty, however precisely modeled. I'm just trying to share a perspective (and several relevant links!) on some of the principles involved with someone who seems interested in and frustrated by the questions.

[1] https://physics.stackexchange.com/questions/244659/how-did-r...

[2] https://en.wikipedia.org/wiki/Michael_Crichton#Gell-Mann_amn...

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#188

Post by rendall »

We performed this experiment in high school chemistry and it has remained with me as one of my deepest aha moments.

It has become fashionable in foodie circles to mock the idea of adding oil to boiling pasta so as to prevent stickiness. The argument against seems to be that oil floats and cannot possibly affect the pasta, unless you add so much that the pasta becomes slimy. But I maintain that a drop of two in boiling water is enough to coat all the pasta in a single layer of molecules. The agitation of the water spreads the oil evenly as a kind of colloidal suspension.

All these fancy restaurants with elaborate methods to avoid sticky pasta.

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#189

Post by tech_ken »

bialpio wrote: post #183
tech_ken wrote: post #172

Earlier quoted context omitted.

> is more of a numerological coincidence, the actual estimate that Rayleigh gives is half of that! I definitely thought so too, in the first paper the 1.67 isn't even really his primary guess so definitely some presentist bias in the OP

From the paper: "The thickness of oil required to take the life out of the camphor movements lies between one and two millionths of a millimetre, and may be estimated with some precision at 1’6 micromillimetre." Looks like a primary guess to me, even if the table lists more data points.

Oh nice my bad, I was just looking at the tables

Re: In 1870, Lord Rayleigh used oil and water to calculate the size of molecules

#190

Post by qsdf38100 »

ForOldHack wrote: post #180
wwarner wrote: post #50

In 1676 Roemer estimated the speed of light by timing the orbit of Jupiter’s moon Io, noting that as the Earth approached Jupiter, Io emerged from behind Jupiter a little earlier every day, and as the Earth traveled away from Jupiter it appeared a little later every day, with the time of day varying by 22 minutes over a year. Knowing the difference between the two distances, he reckoned that light travels that distan…

The speed of light, because it is so fast, is the most precise physical constant known : 299 792 458 m / s. Less than 7/1000ths away from 300,000,000 m / s. I am not going to sweat this in the least. So light travels only 0.3 m / nano second, or 11.802 inches. https://www.youtube.com/watch?v=9eyFDBPk4Yw

Speed of light has recently been redefined, and is now _exactly_ 299792458 m/s. It’s no longer a measurement, it’s a definition.
Post reply on HN