3 ms·

> You need a mechanism that searches and focuses the space of possible chemistry, that forces life into being. That's exactly what the infinite monkeys on a ty

> You need a mechanism that searches and focuses the space of possible chemistry, that forces life into being. That's exactly what the infinite monkeys on a typewriter are a metaphor for. In literal terms, it's trillions of trillions of protein reactions happening along earth's coastlines. > If you look at the timeline for the emergence of life, I don't think it's one that supports an infinite monkey type of mechanism. It was too fast. If it literally was infinite monkeys, we'd expect life to emerge near-instantaneously. The reality is, of course, that it's not "infinite monkeys" but "a whole f-ing lot of monkeys"--the question is whether it's enough monkeys to produce life in the timeline. There's a lot of research that's been done on this, and we have pretty good data. I'm not an expert in this field, but my outside impression from the research I read is that the open question at this point is actually not why it happened so fast, but why it didn't happen faster. If you're actually interested in the numerical justifications for the timeline, the research is pretty well documented, but given you've ex-nihilo declared that the timeline is too fast, I suspect you aren't actually curious, and are instead leading into an argument you want to make.

andrewflnr 4y ago
My read of the timeline is that life emerged very soon after it was physically possible for it to exist. I'm interested in research wondering why it wasn't even faster. What should I search for? My point is, "infinite monkeys" (as a metaphor, taking them literally is a strawman) describes a brute-force approach. I don't think a physical process that randomly smashes together, say, long RNA molecules will actually result in life, certainly not quickly, probably not at all. The combinatorics of long strings are prohibitive. I think some selection process needed to take over sooner. I don't know if that's the argument you think I "wanted to make". As far as I'm concerned I've said exactly what I mean, just longer and more explicit for a hostile audience this time.
- kerkeslager 4y ago
  > My read of the timeline is that life emerged very soon after it was physically possible for it to exist. I'm interested in research wondering why it wasn't even faster. What should I search for? It's been a while since I researched this, and now I'm trying to search with similar terms to what I used before and it's not working. The reason seems to be that there's a ton of fluff papers about life on Mars and other planets: this sort of speculative modeling isn't without value, but it's lacking much hard evidence, unlike the papers I was reading a decade ago which had a ton of geological information. It's frustrating. I wish I could point you at something good here, if only so I could find stuff for myself again. > My point is, "infinite monkeys" (as a metaphor, taking them literally is a strawman) describes a brute-force approach. I don't think a physical process that randomly smashes together, say, long RNA molecules will actually result in life, certainly not quickly, probably not at all. The combinatorics of long strings are prohibitive. I think some selection process needed to take over sooner. Well, two things: Proteins aren't like cryptographic keys, and I'd encourage you to abandon that way of thinking about them. Finding a single cryptographic key is like finding a needle in a haystack: there's only one key that works. But with proteins there are a lot of combinations that work, and even "work" is a bit hard to define because there are a lot of proteins that aren't RNA or DNA, but exhibit lifelike properties. Some of these proteins are very combinatorially simple compared to DNA. So far I don't know of any research that connects a specific prion with a theorized route toward more complex structures (these seem to have evolved in the context of larger organisms) but it at least shows us that very simple proteins can exhibit lifelike properties. Put another way, partial answers count. So instead of a needle in a haystack, it's a haystack with a bunch of needles, and bits of metal that aren't really needles but count as needles. 2. And keep in mind it's not just the problem space that's large: the amount of "processes" doing the searching is also large. My point is, I don't think it's evident that the combinatorics of long strings are prohibitive, given that there are many "right" answers in the search space, and some partial answers count. To actually put numbers on this to say how long it would take would require information I simply don't have, but I'm skeptical that you have that information either.
  - andrewflnr 4y ago
    I don't think of proteins as cryptographic keys. But monkeys-on-typewriters thinking is more closely aligned with cryptography than biochemistry. That's kind of my point.

andrewflnr 4y ago

My read of the timeline is that life emerged very soon after it was physically possible for it to exist. I'm interested in research wondering why it wasn't even faster. What should I search for? My point is, "infinite monkeys" (as a metaphor, taking them literally is a strawman) describes a brute-force approach. I don't think a physical process that randomly smashes together, say, long RNA molecules will actually result in life, certainly not quickly, probably not at all. The combinatorics of long strings are prohibitive. I think some selection process needed to take over sooner. I don't know if that's the argument you think I "wanted to make". As far as I'm concerned I've said exactly what I mean, just longer and more explicit for a hostile audience this time.

kerkeslager 4y ago
> My read of the timeline is that life emerged very soon after it was physically possible for it to exist. I'm interested in research wondering why it wasn't even faster. What should I search for? It's been a while since I researched this, and now I'm trying to search with similar terms to what I used before and it's not working. The reason seems to be that there's a ton of fluff papers about life on Mars and other planets: this sort of speculative modeling isn't without value, but it's lacking much hard evidence, unlike the papers I was reading a decade ago which had a ton of geological information. It's frustrating. I wish I could point you at something good here, if only so I could find stuff for myself again. > My point is, "infinite monkeys" (as a metaphor, taking them literally is a strawman) describes a brute-force approach. I don't think a physical process that randomly smashes together, say, long RNA molecules will actually result in life, certainly not quickly, probably not at all. The combinatorics of long strings are prohibitive. I think some selection process needed to take over sooner. Well, two things: Proteins aren't like cryptographic keys, and I'd encourage you to abandon that way of thinking about them. Finding a single cryptographic key is like finding a needle in a haystack: there's only one key that works. But with proteins there are a lot of combinations that work, and even "work" is a bit hard to define because there are a lot of proteins that aren't RNA or DNA, but exhibit lifelike properties. Some of these proteins are very combinatorially simple compared to DNA. So far I don't know of any research that connects a specific prion with a theorized route toward more complex structures (these seem to have evolved in the context of larger organisms) but it at least shows us that very simple proteins can exhibit lifelike properties. Put another way, partial answers count. So instead of a needle in a haystack, it's a haystack with a bunch of needles, and bits of metal that aren't really needles but count as needles. 2. And keep in mind it's not just the problem space that's large: the amount of "processes" doing the searching is also large. My point is, I don't think it's evident that the combinatorics of long strings are prohibitive, given that there are many "right" answers in the search space, and some partial answers count. To actually put numbers on this to say how long it would take would require information I simply don't have, but I'm skeptical that you have that information either.
- andrewflnr 4y ago
  I don't think of proteins as cryptographic keys. But monkeys-on-typewriters thinking is more closely aligned with cryptography than biochemistry. That's kind of my point.