nathanielvirgo.bsky.social
Researcher in applied category theory at Hertfordshire, UK, formerly at ELSI, Japan. Maths, science and random creative projects.
67 posts
308 followers
150 following
Getting Started
Active Commenter
comment in response to
post
But there are a lot of people who don't seem to see it that way and instead talk about it as if there is something called "the" level of selection, which you have to identify before you can talk about natural selection at all. I agree with you that this is silly.
comment in response to
post
But of course it does happen, and in some circumstances it has easily observable effects, e.g. eusocial insects. This is also sort of like stat mech, where sometimes long range forces dominate and sometimes short ones do, etc.
comment in response to
post
easy for it to be outweighed by individual level effects, such that the effect of group selection would be basically negligible, at least according to the models people were using at the time. This led to a backlash where people would say (incorrectly!) that group selection doesn't exist at all.
comment in response to
post
The issue is which of them will have observable effects. A rough potted history is that people used to say things like "trait X was selected because it benefits the group even though it costs the individual", but then it was shown mathematically that although that effect does occur, it's very
comment in response to
post
experiment in light and texture, possibly unsuccessful
comment in response to
post
first attempt
comment in response to
post
What's funny is that if I log in to that site I basically only see posts like that, about how amazing AI is. Even the posts about the site's owner are mostly gone. The algorithm is really pushing that topic because the bubble really needs to stay inflated just that little bit longer.
comment in response to
post
We don't know for sure that it doesn't
comment in response to
post
#astrobiology #speculativefiction
comment in response to
post
(To be continued or not, we'll see.)
comment in response to
post
The third was the discovery of clear, unambiguous and utterly compelling evidence of intelligent extraterrestrial civilisations, in their trillions, throughout the universe. But we'll get to that later.
comment in response to
post
The second was the discoveries in the oceans of Europa and Enceladus and on Titan and the asteroids, and the resulting long-raging disagreements about what exactly had been discovered and the nature of the boundary between chemistry and biology, if there even was one at all.
comment in response to
post
There were three things that began to dispel that idea. The first was increasing evidence of biosignatures on exoplanets, although there was much debate about that and it would take more than 20 years to reach even a tentative agreement on the right interpretation of the data.
comment in response to
post
Although the discovery of Mars-life had taught us much about the nature of life and its history on Earth, it hadn't done much to dispel the idea that the origin of life was a unique event that only occurred once. Both biospheres shared a common ancestor, after all.
comment in response to
post
In many ways Mars-life was better at evolution than Earth-life; by some measures Mars had a higher biodiversity than Earth. But Earth-life had an advantage despite that. It had found a set of 20 amino acids that worked and fixed them once and for all, allowing it to innovate in other areas.
comment in response to
post
This meant that Martian life could use many more than the 21 amino acids used by Earth-life, but it came at the cost that the code was more error prone and constantly subject to change, making it difficult for genes to be transferred between lineages.
comment in response to
post
They shared much of the same machinery for DNA replication and translation into proteins, as well as some of the core metabolic reactions, but the Martian genetic code used four base pairs to code for an amino acid rather than three.
comment in response to
post
Either way, some of the adaptations that made life so successful on Earth had only arisen after that, with the result that Earth-life and Mars-life differed in fundamental ways.
comment in response to
post
(This is a reminder that you're listening to Orson Welles' radio adaptation of War of the Worlds. Life on Mars has not in fact been announced.)
comment in response to
post
Life must have hitched a lift from Earth to Mars on debris from one of the many meteorites that impacted earth in the first epoch of Earth's biosphere. Or perhaps life originated on Mars, only to find a far more clement home on Earth. We'll never know for sure.
comment in response to
post
However, over time it became clear that we were looking at something else. Life on Mars did share an ancestor with life on Earth, but they had diverged billions of years ago, *before* the last common ancestor of life on Earth.
comment in response to
post
But the instrument was small and was able to piggyback on another mission. Its success was profound. It not only detected DNA but could partially sequence it. At first the sequences seemed random, though there was concern about contamination because a few resembled known sequences from Earth.
comment in response to
post
The first evidence against this came from Mars, in the form of an experiment to detect nucleic acids via technology developed for gene editing. It was a stroke of luck that this was sent to Mars at all, as there was no guarantee of success: even if life existed there it might not have DNA at all.
comment in response to
post
As a result, in the early 21st century the view was widely held that we are alone: life, and especially intelligent life, was rare in the universe if it existed at all. (At least, this was the view often presented to the public. The private opinions of scientists were in truth rather diverse.)
comment in response to
post
Because of this, all the life we knew of prior to the 21st century was descended from a single common ancestor. Space missions of the mid-late 20th century had revealed what appeared to be a dead Solar system, and the telescopes of the time revealed no obvious signs of life among the stars.
comment in response to
post
The origin of life on Earth took place a long time ago. We don't know how it happened and we never will, though not for the reasons you think. It gave rise to many weird and wonderful lineages of microbes, but one was so successful it drove all the others to extinction. (Microbes can go extinct!)
comment in response to
post
Don't ask me how that works exactly - I only know it's done by one of the most impressive pieces of molecular machinery in the biological world, along with the machines (i.e. enzymes) that replicate DNA, translate it into proteins and pump stuff across the membrane.
comment in response to
post
Photosynthesis involves the reverse of this, where a photon of the right frequency is absorbed, causing an electron to jump up to an orbit with a higher energy. Somehow the plant manages to capture the energy in this electron before it jumps back down again and uses it to make sugar.
comment in response to
post
In a quantum world this doesn't happen because light can only be emitted in discrete jumps, i.e. single photons, where the energy of the photon is the difference between the energy levels of the different electron orbits.
comment in response to
post
I'm no photosynthesis expert but maybe I can help. It's hard to explain anything in chemistry without quantum mechanics, because in a classical world electrons would emit light waves as they orbit the nucleus, so they'd lose energy and spiral into it.
comment in response to
post
Also our biomass ultimately comes from plants, so we're made of air too.
(Plus hydrogen from water and a small amount of nutrients from soil, which make up only a small fraction of our mass but we can't live without them.)
comment in response to
post
Yes, the more precise statement would be that wet plant biomass is about half air and half water, plus a few percent nutrients from soil. But biologists usually think in terms of dry biomass, and that really is mostly made of air.
comment in response to
post
Sure, but by weight it's mostly air, at least if you only count dry biomass. Plant biomass is mostly carbon and oxygen, and those both come from fixing CO2.
comment in response to
post
Surprisingly little of it comes from water either, at least if you only count dry biomass. The hydrogen in biomass comes from water but the carbon and oxygen come from fixing CO2, and since H atoms are so light they count for very little of the biomass.
comment in response to
post
Sure, but by mass it's not very much. Most of a plant's dry mass is carbon and oxygen, and that comes from air. This is why plants generate soil in the long run, instead of using it up
comment in response to
post
Sure - photosynthesis takes carbon and oxygen from the air, plus hydrogen from water, and turns it into sugar. The remaining oxygen from the water gets released into the air as O2. Proteins and DNA and all the other molecules in the plant are made from that sugar, plus trace nutrients from the soil.
comment in response to
post
Plants are made of air, not soil
comment in response to
post
📌
comment in response to
post
I'm a native English speaker from England I'm not even sure I can pronounce the word "squirrel" to be honest
comment in response to
post
I really like this. Do you have a picture of the original painting? Google seems not to have heard of it!
comment in response to
post
Lauritzen (1988) Extremal Families and Systems of Sufficient Statistics, I'm not sure if there's a freely available pdf or not
comment in response to
post
Langton's ant en.m.wikipedia.org/wiki/Langton...
comment in response to
post
📌
comment in response to
post
I hadn't come across Krashov estimators before, that's interesting. For Gaussians it seems like there should be an analytic estimator for the effective correlation - except that I'm not sure it's even defined in that case, because then the variables are unbounded and there's no uniform distribution.
comment in response to
post
But how do you calculate the mutual information from that data - wouldn't you need to bin it anyway? In that case you might as well bin all the data first, then calculate the effective information using the binned distribution. Or is there another way to calculate the MI in the continuous case?