Hacker News new | comments | ask | show | jobs | submit login
If you put chalk under a powerful microscope (twitter.com)
380 points by DanBC 6 months ago | hide | past | web | favorite | 76 comments

Contrast this with the Hubble Deep Field image, which is an extremely tiny patch of the night sky. It’s sizeis equivalent to a tennis ball at a distance of 100 metres, or ~110 yards.

And my god, it’s full of stars. There’s endless detail in the universe, whether you zoom in, or zoom out.

See: https://upload.wikimedia.org/wikipedia/commons/6/69/NASA-HS2...

Each "star" in the Hubble Deep Field image is actually an entire galaxy... The vastness of the universe is difficult to comprehend.

I read Dragon's Egg recently, and one of the key leap from barbarian to civilization that the neutron star aliens make is (simplistically) to count past "many" and think of quantities / numbers abstractly.

Our ability to track distinct objects is impressive by our own standard, and we're pretty good at understanding "big" numbers, say somewhere between a million and a billion. But in the scale of things, we're much closer to the hunter knowing his kill will feed his tribe for many moons versus being a starfaring race.

Re "big" numbers, I can't help thinking of Richard Feynman's comment "There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers."

The entire universe probably contains around 10^80 protons, yet Google's founders casually referred to the number 10^100 to indicate their ambitions for information processing.

A carbon atom is 0.3 nm in diameter, so only about 3.3 million can fit in 1 mm. There are well over twice the number of people in NYC!

All in all, numbers that describe the natural world aren't all that impressive compared to the human world.

> A carbon atom is 0.3 nm in diameter, so only about 3.3 million can fit in 1 mm.

Sure, but about 70 quintillion of them fit in 1 mm^3.

>The entire universe probably contains around 10^80 protons

The entire visible universe that is. It could well be infinite and hence contain a larger number of them.

What does it mean to say something "exists" but is not "observable"?

The inherent problem is that the universe has been transparent for a limited amount of time. Since this time the light from this era has been travelling in all kinds of directions, some of it in ours.

We can only see the light that has had time to reach us. Everything that is farther away than about a dozen billion lightyears is not visible.

Technically new stuff is always becoming visible to use, the visible universe expands naturally as new light reaches us.

Of course, the stuff is now farther away since the universe is expanding and at some point the expansion will likely overtake the speed of light and over some distance may have, which means beyond some distance we will never be able to see or know.

What is beyond the visible universe is quite unknown. We can't see it.

Skepticism. What's not observable now might become observable eventually.

Googol isn't the estimated ambition for Google's database, it's just the biggest number with a cute, short name.

Excellent comparison. Puts things into perspective

> All in all, numbers that describe the natural world aren't all that impressive compared to the human world.

But you're comparing mere cardinality with combinations and permutations. We know the latter always produces huge numbers.

> we're pretty good at understanding "big" numbers, say somewhere between a million and a billion.

When we stop and think about it, maybe. But try this experiment with a friend (or a few times with different friends): take a sheet of A4 or "letter" sized paper and a pencil, turn it in landscape orientation and mark the far left side of it with '0', and the far right side as '1B' telling them it's an even scale from zero to one billion, and ask them "without thinking about it just mark where a million would be." -- it's important to get a response from "the top of their head" to understand the difference between what we can recall, and what we habitually think. Most people are way off. If paper isn't handy use fingers and a large object you can touch - e.g. top edge of monitor or TV etc.

I asked someone. Their approach was to divide by half a few times. Looks like they made a good guess.

Dividing in half 10 times comes pretty close (976562), but kinda defeats the point of getting a spontaneous answer as the what they feel it would be. On a 1 meter scale the 1 million mark would be just 1mm along. On a piece of paper, it's virtually indistinguishable from the starting point.

Why would you use a million and a billion instead of 1 and 1000?

The point of the exercise is that many people don't really grok that a billion is a thousand times bigger than a million, - or rather just how much bigger a factor of 1000 actually is, when talking about big numbers that they're not as used to... no matter whether using paper, or a desk, or a wall for scale.

Depends on whether the scale is linear or logarithmic.

I should have specified linear instead of even.

That works by using "anchoring" by priming them with the length scale of the paper.

here is a perspective video: https://www.youtube.com/watch?v=q1mkjkTqg0Y

The star UY Scuti is 2,500,000,000 kilometres in diameter.

It would take light nearly 2.5hrs to travel that far.

One of better videos of this type. Anyone know what's the music from background? Hair raising.

That thread has some beautiful photos. This one is my favorite[1], it's amazing to me how intricate the math of an object can be at a microscopic level.

I mean look at the black and white 2x4, top one 2nd from the left, my mind instantly just wanted to plot that in Matlab and gaze at how complex the equation is!

[1] https://twitter.com/ferrisjabr/status/1022890070531960837?s=...

To be honest, I found that and the coccolith pictures quite unpleasant/uncomfortable to look at --- it could be because I'm somewhat trypophobic, since they caused me to feel the same "brain-melting" sensation and have the similar characteristic of many almost-regular-but-irregular edges. The "neural network dream" generated images did the same thing to me; I feel my brain heat up and sweat a little.

In contrast, something like a box of gears or related machinery parts, which theoretically is also full of edges, doesn't provoke the same feeling...

That’s interesting because my first reaction seeing the photos was to think that “they look like a bunch of gears, huh that’s neat”.

Deep dream also made me feel calm, like I was watching a natural process, like a plant growing, or the rain falling. I hadn't had that kind of reaction to "computer" art before.

you will love the movie 'Annihilation' then, in case you haven't already watched it

Have you tried to see the computer-aided reconstructions of how your own cells function?

E.g. starting with the DNA:


Only shadowed by the fact that it was highly surprising, but yeah I'm not sure I'll use chalk ever again. I damn hate this kind of pictures.

Thanks for the warning. I’m trypophobic as well.

It's easier to read this as an single unrolled post than as a set of tweets:


This looks like a very obvious cowpath that Twitter could pave. I just don't get why they haven't done it yet.

Twitter doesn’t understand what makes it successful so it is loath to change anything.

The sad part is that Twitter started as a SMS relay, ans SMS has had a system for splitting and joining long messages for ages...

That's what Moments is supposed to be for. People can make their own, but I rarely see it.

Because twitter is Craigslist.

It’s a slippery slope. If they do this, what’s next? Where do they stop? What’s it going to take for users to finally be satisfied?

(Edit: Don’t understand the downvotes. Don’t understand a lot of downvotes lately. Do I have a target on my back?)

It's entirely reasonable to want to be able to read a series of tweets that are meant to be read as one long tweet/block of text the way it was intended to be read. I don't understand how you think this is an unreasonable request, especially when tools exist to correct this issue with the max-character limitation Twitter has, or what "slippery slope" this could have.

You're getting downvotes because you're advocating something that makes entirely 0 sense from a usability perspective. And commenting on why you're being voted in a particular way is wholly against the guidelines:

> Please don't comment about the voting on comments. It never does any good, and it makes boring reading.

If a series of tweets can easily be put into a defined order for consumption then why not just lift the max character limitation and let people make tweets as long as they want? Seems like it’d be an easy slide down that slope once you’re on it.

There is no slope. Either the limitation is there, or it isn't. What comes after to call this a "slippery slope"?

And clearly the existing limitation is a pain in the ass, or else services that "unroll" multi-tweet text wouldn't exist. The limitation is completely arbitrary and clearly not entirely helpful, so a request to lift it is entirely reasonable.

I can easily imagine a series of changes that ultimately turns Twitter into something like Medium.

> Either the limitation is there, or it isn't.

Well, if you can compose multiple Twitter messages into a long one, the limitation is kind of just half there.

The difference is all about the contributors.

Each having a low limit, or one having a low limit makes for specific conversation dynamics.

Long form, or free form is a different thing.

Equating them is over simplifying both the request and dynamics in play.

Would that be an issue? The character limit is clearly a pain point for some users (graph here[1] of average tweet lengths -- many people abandon their tweets if they hit the limit), so why not alleviate it? Especially given that it doesn't appear to impact other users' behaviour: only 5% of users actually used the extra space during testing of the increase from 140 to 280 characters [2].

This discussion reminds me of when Instagram stopped enforcing that all pictures be square, a (mostly) ideological limit that made using the service harder.

[1] https://blog.twitter.com/official/en_us/topics/product/2017/...

[2] https://www.zdnet.com/article/twitter-opens-floodgates-on-28...

Would it be a bad slope for them to go down?

PS: I can't tell, because I have no idea what people see in Twitter in the first place.

Those coccoliths also make for fantastic oil bearing reservoir rock, provided certain conditions are met, otherwise chalk tends to make a good seal. Some of the most prolific oil fields in the world, in the north sea, come from fractured chalk beds.

For example, massive salt diapirs (think of them as giant, mountain sized fingers) 5-15 miles beneath the earth's surface are gradually squeezed, like a fluid, moving upwards and fracturing rock (and chalk) above, creating porosity and permeability necessary for oil production.

Reminds me of a quick, classic read from GK Chesterton: "A Piece of Chalk" https://www.chesterton.org/a-piece-of-chalk/

Also the composition of diatomaceous earth https://en.m.wikipedia.org/wiki/Diatomaceous_earth which is used in certain types of water filters, like pool filters.

In a discussion of the economics of a far future spacefaring human civilization, someone once suggested diatomaceous earth as one of the few raw materials that might be exported from Earth, since it might well not occur anywhere else!

Especially since diatomaceous earth is good for preventing infestations of insect-sized organisms without depending on the biochemistry of those organisms.

Careful: Not Safe For Trypophobics

So reading Huxley's lecture, https://ebooks.adelaide.edu.au/h/huxley/thomas_henry/piece-o... are we supposed to understand that chalky places like Dover were the bottom of the sea long long time ago? (He is talking about the mud at the bottom of the ocean having the remains of the same creatures.)

> But if the Radiolaria and Diatoms are thus rained upon the bottom of the sea, from the superficial layer of its waters in which they pass their lives, it is obviously possible that the Globigerinae may be similarly derived; and if they were so, it would be much more easy to understand how they obtain their supply of food than it is at present. Nevertheless, the positive and negative evidence all points the other way. The skeletons of the full-grown, deep-sea Globigerinae are so remarkably solid and heavy in proportion to their surface as to seem little fitted for floating; and, as a matter of fact, they are not to be found along with the Diatoms and Radiolaria, in the uppermost stratum of the open ocean.

> It has been observed, again, that the abundance of Globigerinae, in proportion to other organisms, of like kind, increases with the depth of the sea; and that deep-water Globigerinae are larger than those which live in shallower parts of the sea; and such facts negative the supposition that these organisms have been swept by currents from the shallows into the deeps of the Atlantic.

> It therefore seems to be hardly doubtful that these wonderful creatures live and die at the depths in which they are found.

> However, the important points for us are, that the living Globigerinae are exclusively marine animals, the skeletons of which abound at the bottom of deep seas; and that there is not a shadow of reason for believing that the habits of the Globigerinae of the chalk differed from those of the existing species. But if this be true, there is no escaping the conclusion that the chalk itself is the dried mud of an ancient deep sea.

From the comments in the Twitter thread:

> Only tangentially relevant, but I love the fact that sand from Normandy still consists of up to 4% microscopic war materiel from WWII.

Looking at some of those pictures gave me a slightly light-headed feeling - the same kind I get when I look at the Hubble ultra deep-field.

Newbie here with these types of images. In the image there are spheres and flat discs. I take it the life forms were spherical and flat discs have broken off rather than vice versa? Of course I find it hard to imagine discs forming spheres but at the same time, that sphere doesn't seem single cell to me. Fascinating none the less.

> I take it the life forms were spherical and flat discs have broken off

Exactly, yes:


Thanks, sorry for the silly question as well, I aggressively block social media on my laptop and only seen the image. Visiting your link on a PC answered all my questions.

For those of you enjoy good quality things under a microscope, take a look at https://www.nikonsmallworld.com/

the world is so beautiful

and to think that most of the space in an atom is vacant. Seems to be a tremendous waste of space. incomprehensible that nature would permit this. surely something must exist in that nothingness where the electron vibrates around the nucleus. and we are blind, deaf, insensate to it.

You just said it... the electron is there. At any given moment, the electrons occupy that “empty” space, which means it isn’t really empty, but full of electrons. The wave function of the electrons is a “smeared” out geometry which occupies that space, with various probabilities of localizing the electron at any given point. When you touch something and experience mutual electromagnetic repulsion (i.e. what touching actually is) you’re very much sensing that. No need to invoke magic and mystery to fill a gap that doesn’t exist.

That does very much sound like magic. The wave function is in principle just describing the probability of measuring a value, if I recall correctly. What you make of it is but an attractive theory to interpret that wavefunction after the fact. And it's not particularly satisfying. It's completely meaningless that it would have been in all places at once, if in fact, when you look, it's only ever in one place.

“When you look” is the key phrase there. When you measure you’ve perturbed the system and a superposition of probabilities collapses to a single possibility. You can try to ignore the problem, but it keeps cropping up. Take for example the Delayed Choice Quantum Eraser:


I am not going to even click that. "superposition of probabilities" is not a real thing, merely a mathematical tool. The definition of probability still involves something that is going to happen. Hey that's amazing, it's the probe-ability.

edit: It's very hard for people to admit we really don't know, is what I am trying to say. I mean that's the real definition of chance, we don't know for certain. Whereas, if there is more to the field theories, more than ether, I'd really like to know, but I'm not holding my breath. Between measurement uncertainty and observer uncertainty, the models will remain just that.

Well you’re allowed to believe whatever you want of course, however divorced from observation, experiment and theory it happens to be. I would just caution you against drawing such strong conclusions about a field you seem to know very little about based on what you think should be, or your intuition.

Well, now I read the paper and, while I do find it interesting but don't follow the formulas, I wonder what you are trying to show me.

Removing the Beamsplitter (BSA) would only remove the ability to correlate the measurements, so how do you know that there are actually no instances of Gaussian distribution happening already at the origin, which would cause the fork at the splitter (instead of information traveling back in time, for example)?

> observation, experiment and theory

we agree on the observation and theory parts. The predictive power of the theory for experiments is duly noted, but here the object under scrutiny is way bigger than a single atom. Whereas the science around the materials used, crystallography to begin with, is way above my grade.

The language in the paper caused me a bit of trouble: "It is easy to see ...", "at the same time", "a quantum".

Aww, I was gearing up for learning about blackboard chalk.

When elon musk says it's possible that we're living in a simulation, it's fun to think about the computer that can handle this level of detail and complexity from the very minute to the billions of galaxies

And then it's funny to realize a simulation would not have to compute those details until someone looks at them.

You're definitely right about that! But who is that someone? A human, a bateria, any sentient being?

Anybody who cares enough to ask the question. Everything else can be abstracted over.

And then the simulation only has to be detailed enough to provide you with a reasonably-plausible answer. As soon as you stop looking, the algorithm can just strip out the detail.

This trick isn't even that hard, graphics processors do it all the time.


Me. Who else ?

Not you, you're part of the simulation. It's just superasn, Elon Musk, and a bunch of goldfish used to generate entropy.

Applications are open for YC Summer 2019

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