4 ms·

>I don't disagree with any of them! Well you sure "didn't disagree" at great length above! >you're gonna need a much stronger argument for this than a list of

>I don't disagree with any of them! Well you sure "didn't disagree" at great length above! >you're gonna need a much stronger argument for this than a list of things from 50 years ago that are obvious huge inventions and...what, a much shorter list for the last 50 years? I don't think I do need a much stronger argument than this: you just said above you agree with all of my examples. Now you say ... my examples aren't enough. Yet you "feel" I am right anyway? What would be "enough?" Simply looking at 1969's vision of the future is pretty damning as well. Re your Kelvin goalposting; you don't have an example of technological pessimism from 1969, do you? You've been a reasonably honest interlocutor thus far; don't goalpoast me, bro. "So what makes 1959-2009 "not progress"?" was bad enough. >Propose a mechanism of action, ask why is this happening? I have my suspicions, which is why I consider writing a book on it. First, though, I'd like to get you, who agrees with me on one line, then disagrees on another (you assure me this is not routine bay area tone policing, and I'll just assume so for the sake of argument), to admit it is actually happening. What evidence besides your subjective feelings on the matter would you accept? What evidence do you, who claims to agree with me, yet disagrees on every detail, have? I think naming a series of bonkers technological revolutions that happened in one time period with not even one comparable example in a later time period is a reasonable and strong argument. If you disagree with the time span in question, I invite you to suggest a better time span, or one going back another 50 years. 1869 to 1919 was super amazing, and nobody thought technology was over in that time period, even if Kelvin thought physics was done (which was a reasonable thing to think back then; oddly people no longer think this, despite the parallel lack of experimental results which contradict theory). Same with the 50 years prior to 1869; technological progress was vast then. Any of those 50 year periods had more important breakthroughs than today. It's the recent era which has slowed down. You'll get the same result breaking it down by decade, by 20 years or 40 as I did in the original takimag article.

zamfi 7y ago
I'm going to meta-level with you: 1. No one is saying that the inventions of 1909-1959 were not amazing. Yes, the jet engine is fantastic. Flying is bonkers. Antibiotics, great. No one is seriously arguing that these are not spectacular inventions that transformed humanity. You don't need to keep repeating the argument that 1909-1959 was great, by listing great things from then, because no one is arguing against that. 2. I (at least) am making an argument about the proof you offer that progress is slowing, I am saying it is weak proof. I am doing so in two ways: (A) by pointing out that the ways you are arguing 1909-1959 things are great can also be applied to post-1959 things, and the ways you are arguing post-1959 things aren't great can also be applied to things from 1909-1959. This argument looks like this: jets are great compared with props in the same way that iPhones are great compared with ENIAC. So why is the former transformation categorically different from the latter? You are interpreting A as me saying that things from 1909-1959 weren't great. I'm not saying that. No one is saying that. You are saying that my examples are absurd: yes, that is the whole point of this argument. It's absurd to say that jets are not great because we already had props, just as it's absurd to say that iPhones are not great because we had ENIAC (or whatever appropriate computer from 1950 you're thinking of). (B) by pointing out that it is really hard to know the impact of many really amazing, crazy foundational things until 50 years after the fact. Not all things, of course, which is why this thread is full of examples and counterexamples. People being optimistic or pessimistic about tech in 1969 is irrelevant -- if I found someone who in 1969 thought that 1919-1969 was lame compared with 1869-1919, would that convince you of anything? No, because it's irrelevant. The whole Kelvin thing is apparently apocryphal anyway. With that out of the way, you ask: what would be enough evidence to convince me that progress is slowing? First, one problem is that we'd need a definition of progress. "Rate of invention of things Scott thinks are obviously amazing with 50 years of hindsight"--aka your list---is not a great definition because we don't have 50 years of hindsight on 2009. See point (B) above. Other people have looked at counts of patents, per-capita research papers, etc., all of which continue to accelerate. But these are all also problematic, because these things can be gamed and end up correlated with other things that are irrelevant to that ineffable "actual progress" we're both arguing around. We could go with lives saved, by which measure 1959-2009 looks amazing except that most of those lives aren't in the USA or Europe, compared with 1909-1959. In another post, you say (paraphrasing, but I think I got it) "things that give a society that possesses them a guaranteed win in a war over societies that don't" but this is a very pre-atom-bomb type of definition. At this point, the atom bomb is that thing for everyone all the time, so it makes sense that societies don't prioritize weapons like that anymore: you don't need them! We already have the bomb. Who needs a second world-annihilating type of device? So technological progress doesn't look like this anymore, post-bomb. (Even the things you list as examples of post-1969 definitely don't fit. Coding theory feels like a stretch here IMO.) I speculate that it's because societies tend to fight more in economic terms today than physical terms. So we get information-related inventions and financing inventions, not physical-dominance or energy-extraction inventions. But I digress. I don't have a good definition of progress. But I'd love to come up with one that fits. Second, without a definition, things are tricky, but I tend to look at proxy measures like productivity. Technology lets people do more, so, how much can people do per unit time? The rate of productivity growth has been slowing a bunch since 1970-1980. Another proxy measure you could look at is health outcomes: life expectancy at birth (reflecting childhood diseases, mostly) has basically been stagnant since 1980. These proxy measures are what make me think that progress is slowing. But it could be the proxy measures losing their proxy-value instead. I welcome a new definition of progress from you to argue about.
- scottlocklin 7y ago
  Thanks for continuing the discussion in a constructive way! > it is really hard to know the impact of many really amazing, crazy foundational things until 50 years after the fact. That's absolutely untrue in all of the examples I listed. Nukes, antibiotics, supersonic jets, space flight, the invention of computers; all of these things were immediately recognized as of titanic importance. There is nothing recognizable in the last 50 years as being of comparable importance. Cell phones and search engines are cool, but not hugely different from what we had before, and again: not comparable to the previous 50 years. They were minor revolutions, and how I make my living, but compared to antibiotics? Feh! Sure there may be things in the lab now or back a few years which will eventually prove amazing. That's pretty impossible to talk about; I doubt there is even a single decent example from the prior 50 years you could name that was as hidden as these alleged breakthroughs are now. Important things aren't often kept secret. I mean, maybe the air force has antigravity UFOs like the tinfoil hat brigades say; I kind of doubt it! There is a distinct lack of obvious breakthroughs of the last 50 years. That's what I keep saying! Maybe CRISPR or whatever will change everything; but they've been saying stuff like this since the 70s, and again, the biggest biotech revolution thus far is ... as I said ... cheap insulin. Not even close to antibiotics. Your point about us not developing super-weapons is not obviously true. The US has basically failed at this, just as it continues to fail at most things involving large scale industry: but the Russians just fielded a nuclear ramjet cruise missile (the US was working on this in the 50s) as well as a crazy doomsday torpedo and some weird hypersonic things I haven't fully parsed yet. Also, antibiotics wasn't a super weapon, nor was computers, nor any number of other examples; but these technologies conveyed huge advantages on the possessors of such technologies that have military implications. The point of measuring breakthrough techs via their military advantages is to have some obvious metric. It's a pretty good one! Nothing is perfect, but it's a decent way of pointing out why some technological innovations are more important than others; evolutionary fitness. Re productivity; if you remove chip improvements from economic measures of productivity; things look super grim. That's what I said in the original article!
  - zamfi 7y ago
    > Nukes [1940's], antibiotics [1920's], supersonic jets [1940's], space flight [1960's], the invention of computers [what are we calling this, 1950's?]. Supersonic jets do not belong on that list. I get that you like them; the world would be pretty similar if no one had ever invented them. Personal computers ('70s-'80s) definitely belong on that list, and I'd argue that smartphones (~2007) do too but you apparently disagree. Maybe the Web too ('90s), though I imagine you definitely disagree there. So, now that we have a list that's more evenly distributed across decades for the last 90 years, what? We just argue about what's on the list? Who cares? Is it just what I see as humanity-changing vs. you? This is exactly the kind of useless argument I don't want to have, because it tells me nothing except what Scott thinks is important. Your military-usefulness definition is interesting, but I think limited, and ignores how competition between societies has changed in non-military ways. So, good at pointing out why some inventions are more important than others? Yeah, for some subset of inventions, but conveniently not recent ones. So I don't really think it strengthens your argument. Further: > Your point about us not developing super-weapons is not obviously true. I made no such point. I said that the incentives for people to work on breakthrough inventions has shifted, and thus as a result that progress would look different post-bomb. The US rewards invention differently than Russia, so it's not surprising that Americans aren't building superweapons, that military funding no longer produces amazing breakthroughs at the rate it did, and that the Russian military is still working on new ways to annihilate everyone. > Re productivity; if you remove chip improvements from economic measures of productivity; things look super grim. That's what I said in the original article! But this is wrong? Productivity stalled in the '70s, just as chip improvements really started to skyrocket!

zamfi 7y ago

I'm going to meta-level with you: 1. No one is saying that the inventions of 1909-1959 were not amazing. Yes, the jet engine is fantastic. Flying is bonkers. Antibiotics, great. No one is seriously arguing that these are not spectacular inventions that transformed humanity. You don't need to keep repeating the argument that 1909-1959 was great, by listing great things from then, because no one is arguing against that. 2. I (at least) am making an argument about the proof you offer that progress is slowing, I am saying it is weak proof. I am doing so in two ways: (A) by pointing out that the ways you are arguing 1909-1959 things are great can also be applied to post-1959 things, and the ways you are arguing post-1959 things aren't great can also be applied to things from 1909-1959. This argument looks like this: jets are great compared with props in the same way that iPhones are great compared with ENIAC. So why is the former transformation categorically different from the latter? You are interpreting A as me saying that things from 1909-1959 weren't great. I'm not saying that. No one is saying that. You are saying that my examples are absurd: yes, that is the whole point of this argument. It's absurd to say that jets are not great because we already had props, just as it's absurd to say that iPhones are not great because we had ENIAC (or whatever appropriate computer from 1950 you're thinking of). (B) by pointing out that it is really hard to know the impact of many really amazing, crazy foundational things until 50 years after the fact. Not all things, of course, which is why this thread is full of examples and counterexamples. People being optimistic or pessimistic about tech in 1969 is irrelevant -- if I found someone who in 1969 thought that 1919-1969 was lame compared with 1869-1919, would that convince you of anything? No, because it's irrelevant. The whole Kelvin thing is apparently apocryphal anyway. With that out of the way, you ask: what would be enough evidence to convince me that progress is slowing? First, one problem is that we'd need a definition of progress. "Rate of invention of things Scott thinks are obviously amazing with 50 years of hindsight"--aka your list---is not a great definition because we don't have 50 years of hindsight on 2009. See point (B) above. Other people have looked at counts of patents, per-capita research papers, etc., all of which continue to accelerate. But these are all also problematic, because these things can be gamed and end up correlated with other things that are irrelevant to that ineffable "actual progress" we're both arguing around. We could go with lives saved, by which measure 1959-2009 looks amazing except that most of those lives aren't in the USA or Europe, compared with 1909-1959. In another post, you say (paraphrasing, but I think I got it) "things that give a society that possesses them a guaranteed win in a war over societies that don't" but this is a very pre-atom-bomb type of definition. At this point, the atom bomb is that thing for everyone all the time, so it makes sense that societies don't prioritize weapons like that anymore: you don't need them! We already have the bomb. Who needs a second world-annihilating type of device? So technological progress doesn't look like this anymore, post-bomb. (Even the things you list as examples of post-1969 definitely don't fit. Coding theory feels like a stretch here IMO.) I speculate that it's because societies tend to fight more in economic terms today than physical terms. So we get information-related inventions and financing inventions, not physical-dominance or energy-extraction inventions. But I digress. I don't have a good definition of progress. But I'd love to come up with one that fits. Second, without a definition, things are tricky, but I tend to look at proxy measures like productivity. Technology lets people do more, so, how much can people do per unit time? The rate of productivity growth has been slowing a bunch since 1970-1980. Another proxy measure you could look at is health outcomes: life expectancy at birth (reflecting childhood diseases, mostly) has basically been stagnant since 1980. These proxy measures are what make me think that progress is slowing. But it could be the proxy measures losing their proxy-value instead. I welcome a new definition of progress from you to argue about.

scottlocklin 7y ago
Thanks for continuing the discussion in a constructive way! > it is really hard to know the impact of many really amazing, crazy foundational things until 50 years after the fact. That's absolutely untrue in all of the examples I listed. Nukes, antibiotics, supersonic jets, space flight, the invention of computers; all of these things were immediately recognized as of titanic importance. There is nothing recognizable in the last 50 years as being of comparable importance. Cell phones and search engines are cool, but not hugely different from what we had before, and again: not comparable to the previous 50 years. They were minor revolutions, and how I make my living, but compared to antibiotics? Feh! Sure there may be things in the lab now or back a few years which will eventually prove amazing. That's pretty impossible to talk about; I doubt there is even a single decent example from the prior 50 years you could name that was as hidden as these alleged breakthroughs are now. Important things aren't often kept secret. I mean, maybe the air force has antigravity UFOs like the tinfoil hat brigades say; I kind of doubt it! There is a distinct lack of obvious breakthroughs of the last 50 years. That's what I keep saying! Maybe CRISPR or whatever will change everything; but they've been saying stuff like this since the 70s, and again, the biggest biotech revolution thus far is ... as I said ... cheap insulin. Not even close to antibiotics. Your point about us not developing super-weapons is not obviously true. The US has basically failed at this, just as it continues to fail at most things involving large scale industry: but the Russians just fielded a nuclear ramjet cruise missile (the US was working on this in the 50s) as well as a crazy doomsday torpedo and some weird hypersonic things I haven't fully parsed yet. Also, antibiotics wasn't a super weapon, nor was computers, nor any number of other examples; but these technologies conveyed huge advantages on the possessors of such technologies that have military implications. The point of measuring breakthrough techs via their military advantages is to have some obvious metric. It's a pretty good one! Nothing is perfect, but it's a decent way of pointing out why some technological innovations are more important than others; evolutionary fitness. Re productivity; if you remove chip improvements from economic measures of productivity; things look super grim. That's what I said in the original article!
- zamfi 7y ago
  > Nukes [1940's], antibiotics [1920's], supersonic jets [1940's], space flight [1960's], the invention of computers [what are we calling this, 1950's?]. Supersonic jets do not belong on that list. I get that you like them; the world would be pretty similar if no one had ever invented them. Personal computers ('70s-'80s) definitely belong on that list, and I'd argue that smartphones (~2007) do too but you apparently disagree. Maybe the Web too ('90s), though I imagine you definitely disagree there. So, now that we have a list that's more evenly distributed across decades for the last 90 years, what? We just argue about what's on the list? Who cares? Is it just what I see as humanity-changing vs. you? This is exactly the kind of useless argument I don't want to have, because it tells me nothing except what Scott thinks is important. Your military-usefulness definition is interesting, but I think limited, and ignores how competition between societies has changed in non-military ways. So, good at pointing out why some inventions are more important than others? Yeah, for some subset of inventions, but conveniently not recent ones. So I don't really think it strengthens your argument. Further: > Your point about us not developing super-weapons is not obviously true. I made no such point. I said that the incentives for people to work on breakthrough inventions has shifted, and thus as a result that progress would look different post-bomb. The US rewards invention differently than Russia, so it's not surprising that Americans aren't building superweapons, that military funding no longer produces amazing breakthroughs at the rate it did, and that the Russian military is still working on new ways to annihilate everyone. > Re productivity; if you remove chip improvements from economic measures of productivity; things look super grim. That's what I said in the original article! But this is wrong? Productivity stalled in the '70s, just as chip improvements really started to skyrocket!