Open Thread

This is our monthly place to discuss relevant topics that have not appeared in recent posts.

GD Star Rating
Tagged as:
Trackback URL:
  • Nathan

    Is the universe itself a perpetual motion machine? Why or why not?

    What major consequences result from answering “yes” or “no” to this question?

    • dWj

      My response to the last question is “probably none”. The difficulty here seems more semantic than useful. If there’s a sense in which the universe might be cyclic rather than on a one-way march of irreversible processes, then that is of real interest, but if you’re asking whether the conventional model of big bang -> universe as we know it -> big empty void with a few large black holes in it should be called a “perpetual motion machine”, I’m not optimistic that useful insight will be gained from thinking about that question too hard.

      • richard silliker

        “Could you explain why “why or why not?” is a bad question?”

        No answers to the question why.

    • richard silliker

      “Is the universe itself a perpetual motion machine? ”


      “Why or why not?”

      Bad question.

      • Nathan

        Could you explain why “why or why not?” is a bad question?

    • IVV

      Perpetuity assumes similarity in motion across all time. Time is defined within the confines of the universe. Since time is an element of the universe in which perpetuity is defined, i.e. time is a subset of the universe, then the universe can only be perpetual in motion if time and motion are the only components of the universe. If there is anything within the universe that is not time and motion (such as mass), then such items cannot, by definition, be perpetual.

      In short, no, because there is no perpetuity to the universe.

      • Nathan

        Thanks, your comments really helped, especially your first two sentences.

        I’m still confused by:

        “the universe can only be perpetual in motion if time and motion are the only components of the universe.”

        Can motion exist without mass?

      • IVV

        I could imagine a universe in which motion is independent from mass, but this is probably not OUR universe. Most importantly, though, mass can exist without motion (or, at least, motion is not a pre-existing condition of mass).

        Actually, now that I think about it, perhaps mass does require motion–consider the effects of the Heisenberg Uncertainty Principle, and the famous E=mc^2. If velocity is a required component for position, then perhaps mass requires motion…

  • dWj

    When we automate processes, we usually leave a human to monitor the process at some level in a capacity to intervene, often, it seems, without looking at whether intervention is more likely to be good than bad. Is this actually about improving likely outcomes, or do we just want someone obvious to blame if something goes wrong? (Perhaps the person deciding to leave a human in charge is frequently the person likely to bear the blame if it were fully automated.)

  • Doug

    Robin, your ethics have defended the existence of a huge population of (simulated) humans living at subsistence level as a good thing. The argument is based on that rich lives are not that substantially better than poor lives, and that poor lives are still lives worth living. Therefore having more human alive, at poverty, is far better than having a few humans alive at fabulous wealth levels.

    In light of this two questions:
    1) Would you volunteer to have your brain scanned for the prototype EM? Thus having many many entities that feels continuity to your current consciousness exist in dire poverty, while only a single entity (meatspace Robin) enjoys a wealthy existence?

    2) If the mathematical universe or some very large multi-verse hypothesis were true would it change or reduce your support for EMs? Let me explain. The argument for EMs goodness is predicated upon the benefit of getting the chance to exist, versus never living. However in a very large multi-verse, all possible lives and states of existence will end up existing (in fact infinitely many times), just because of the law of large (infinite) numbers. Thus is there as much ethical benefit to brining a life into existence if it already has and will existed an infinite number of times?

    • Yes, things do seem to break down at infinity. But it all should add up to normality. Though one could argue that natalism is already not quite normality…

  • candy

    Has anyone really been far even as decided to use even go want to do look more like?

    • When one really been far even as decided once to use go want, it is then that he has really been far even as decided to use even go want to do look more like.

      It’s just common sense.

      Can you really be far even as decided half as much to use go wish for that? Cromulent formed this none back like decided no more like look.

  • JamieNYC

    Are Crazy Ideas Good Investment?

    Robin, you are probably aware of some (all?) of these:

    – Cold fusion experiments
    – Mach Effect Propulsion
    – Pre-cognition

    My question is: are ideas like these a good “investment” – in other words, is the product of (funds needed for further research/proof of concept) times (the probability of success) times (potential impact) more or less favorable than for the mainstream research?

    One can argue that it is more favorable, i.e. that these areas represent a fertile field for technology investments, based on the fact that these areas are widely considered to be populated by “cranks”, “loonies” etc. (I assume that most of the readers of this blog think so too), so they are very low status, and thus under-funded compared to their potential benefits.

    Or, since the potential breakthroughs would be so desirable, perhaps we are already “investing ” too much efforts into these areas, because our imagination gets the better of us?

    I’d appreciate your opinion. If you believe these areas are underfunded, may be we can write a proposal to a private equity fund? ; -)

    • It does seem to be much easier for us to imagine huge payoffs than tiny probabilities. No doubt probabilities being between 0 and 1, while potential payoffs are unbounded, contribute to this bias. Perhaps we should think in terms of odds.

      It doesn’t help that they are in different units either (we seem to consider utility and probability separately in our minds: utility is ‘how much goodness’ and probability is ‘how rare’) and are conceptually rather different. Is there some way to commensurately compare utilities with probabilities?

  • Recently a Christian pastor burned a copy of the Koran in a mock-trial. This (predictably enough) caused a riot which killed 12 people.

    I assume the social status of the Koran has a positive correlation with human deaths. I’m curious as to whether an equal or greater number of human lives will be saved due to the relative reduction in the Koran’s social status that this could cause — or whether it might, by some counterintuitive mechanism, raise the Koran’s social status and thus cause more deaths.

    • anonymous

      > Recently a Christian pastor burned a copy of the Koran in a
      > mock-trial. This (predictably enough) caused a riot which killed 12
      > people.

      Certainly the rioters bear no responsibility in rioting. Nope. Some pastor in Florida is the causal factor.

      Biased much?

      • Why the sarcasm? If you have a disagreement with an assumption I seem to be making, there are other ways to express it. In this case, you seem confused as I did not assign blame to the pastor but commented on how predictable the events were based on the action taken. Causality is not the same as responsibility. In this case the rioters are clearly 100% responsible — something that is so obvious in any civilized society that I didn’t even think it worth mentioning.

      • Note that you didn’t respond to my actual question, which was whether burning the Koran saved more lives than it costs. I’m not sure why the proper assignment of blame is more interesting to some people than the saving of human lives.

      • Because which lives are saved or lost is more important (to some) than how many.

  • evtujo

    I’m just a well read/mathed AI amateur but it has bothered me for a long time that the medicine is not more automated. I know that some of the early experiments showed that making a medical diagnostic system was much harder than expected (as was pretty much everything). But we have the IBM Jeopardy system, we have increasingly efficient Bayes based systems we have Google. It seems like for most things that people come to the doctor for a computer could do a faster and more consistent job. So either the technology is just coming and we’ll soon have the golden age of computer based health, this problem is far harder to solve than I imagine or there is no interest / silent conspiracy to *not* have these systems.

    Is it just the case that sending someone to a computer doesn’t signal caring?

    Just curious if anyone has thoughts on this. It’s bothered me forever.

    • Dave

      Most diagnoses are not hard. No computer is needed. Computer diagnosis requires collection of a lot on irrelevant information.This is not fast but is laborious. At times the diagnosis depends on the results of various tests. The computer would have to wait for these results as well as a human. Where is the advantage? Much of healing comes from human interaction,even laying on of hands. How would computer do this?
      What about computer interpretation of imaging studies? This will probably come about but will need human review. We do have robot surgery modules played like a video game by the surgeon over in the corner of the operating room.
      What about robot obstetricians. It is a possibility if you just have to have computers do everything. OK girls,any takers? They already complain about cold metal speculums.

  • William Easterly thinks the higher variance among autocracies may be attributable to missing/faulty statistics:

    I’m also going to reiterate my previously off-topic request for his take on Barry Schwartz’ “The Costs of Living” and possibly other of his takes on psychology. I expect he’d make a better diavlog partner than the “Most Human Human” dude.

    • Douglas Knight

      A lot of what Easterly says in that post is nonsense. Yes, it is difficult to see what is going on in Botswana, Bhutan, and China, but it is easy to see what is going on Singapore and Korea. It’s like he didn’t look at his graphs. I am surprised to see Taiwan listed as the worst grade of data.

      Easterly’s main point “Average growth over 1960-2008 might have zero mistake ON AVERAGE, but there will randomly be some countries with a string of exaggerated growth rates” is 100% false.

      GDP growth is not directly measured, but is the difference in GDP from year to year. So an error in measuring GDP one year has no net effect – it increases growth one year and decreases it in a neighboring year. If many years are missing in the middle it doesn’t change the average growth. The question is how much the countries grew over the past 50 years. The difference between the top country and 10th place is the difference between being 7 times as rich and being 17 times as rich. If the per capital wealth were off by a factor of 2 fifty years ago, it would move it around the list, but wouldn’t change the top 10 that much.

      It is possible that his source of GDP growth data is reported directly and subject gaming, in which case, he’s an idiot. He should be using GDP data directly to avoid this problem. But he explicitly says that he’s subject to these problems, so he isn’t thinking at all about what he’s saying. It makes me distrust everything else he says in these two blog posts.

      Africa is in bad shape. Maybe you shouldn’t trust any numbers coming out of there, including Botswana. But I really doubt Botswana is one of the 15 countries not reporting any data. Throwing out Africa just makes dictators look better.

  • You have said before that near-subsistence lives are worth living. Does the same hold for those animals we breed to eventually slaughter and eat? Presumably if we became vegetarians then we would not breed so many of them and their fitness would plummet. There would be fewer of them. But if their lives are worth living, then this is actually an undesirable state of affairs.

    • Wonks Anonymous

      Robin has already given his perspective on that:

      • If pig farms are replaced by asparagus farms, that leaves more soybeans and grains and so forth to feed vegetarian humans with. Thus the sustainable human population can be higher in a world with no pigs in it, in addition to the extra asparagus it produces.

        Since a subsistence-level vegetarian human actually consumes a lot less food than a pig being raised for meat, the value of being a pig would have to be substantially higher than that of a subsistence-level human in order for this to be a justifiable trade-off. If a subsistence-level human is better off than a factory-farmed pig, it is better to be a vegetarian by Robin’s logic.

        On the other hand, some (the antinatalists) would argue that bringing humans into existence, who have awareness of the prospect of their own looming oblivion, is unbearably cruel. Plenty of them might also be relatively indifferent to the fate of mere beasts who can be born and die in large numbers without significant moral hazard (and indeed do so regularly in the course of nature).

        If this is the correct point of view, we should first of all distribute birth control to everyone to prevent starvation scenarios, and second of all encourage the eating of meat in order to decrease the available food supply and thereby discourage reproduction. Pets such as dogs and cats which act as surrogate children (and as major meat consumers) would be quite laudable to such a perspective.

  • Is the universe itself a perpetual motion machine?

    How could we know. It might be analogous to a microbe on the space shuttle’s hull attempting to understand what they’re actually on. We don’t know what the universe is. And we might never.