Thursday Throughput: Big Black Hole Edition

Avatar

Michael Siegel

Michael Siegel is an astronomer living in Pennsylvania. He is on Twitter, blogs at his own site, and has written a novel.

Related Post Roulette

17 Responses

  1. Avatar Michael Cain
    Ignored
    says:

    ThTh3: Not only given permission, but provided a site on federal land that bypasses the state having any say in the the project, and makes cooling water available using the federal government’s authority to preempt state water laws. This will be tied up in court for a few years settling whether the reactor violates the INL agreement with the state of Idaho to not bring additional commercial nuclear waste into the state until the INL cleanup is complete. Myself, I expect the group of small Utah utilities that have agreed to fund this will go bankrupt before a single watt is delivered.Report

  2. Avatar Jaybird
    Ignored
    says:

    ThTh4: While I would be more than happy enough to believe that large gatherings of people are likely to be superspreader events, the whole issue of how it’s seems to only be worth reporting that *FUN* gatherings of people transmit the virus seems to undercut the case.Report

    • Avatar Kazzy in reply to Jaybird
      Ignored
      says:

      I also am not seeing much digging into correlation versus causation. If X people went to Sturgis already sick and X people left Sturgis still sick and no new people got sick, then Sturgis caused zero cases, not X cases. But that isn’t how those events are being talked about everywhere I look.

      How many people got sick at Sturgis who wouldn’t have gotten sick had they never gone? And how many of those people got other people sick who wouldn’t have gotten sick if the initial people never went to Sturgis? That seems to be the important numbers and maybe we don’t have the tools in place to find those numbers… which is a problem. But the solution isn’t to just make up numbers.Report

    • Avatar Brandon Berg in reply to Jaybird
      Ignored
      says:

      I like pointing out the sheer hypocrisy of the Narrative pushers as much as the next guy, but I still think people are underestimating how important the indoors-outdoors distinction is. Indoors the virus just keeps bouncing off walls and getting recirculated by the air conditioning system until it flies up someone’s nose. Outdoors it blows away.

      I don’t doubt that you can get some spreading outdoors, especially in very dense crowds, and to be fair some of the Fiery But Peaceful Protesters did do a fair bit of indoor looting (after breaking windows to keep everyone safe and healthy!), but I do think it’s plausible that the Sturgis rally, which apparently involved a lot of large indoor gatherings, really was a more significant cause of transmission than the Peaceful Protests, at least on a per-capita basis.Report

      • Avatar Jaybird in reply to Brandon Berg
        Ignored
        says:

        I haven’t seen anybody outside of amateur epidemiologists (like us in comment sections!) make that distinction, though. Has anyone argued that it’s probably okay to do stuff like go outside, hang in a park, go to the beach?

        I’ve only seen: Stay Safer At Home and Wear A Mask If You Must Leave versus Racism Is Important Enough To Risk Disease.

        No “hey, go someplace with good ventilation!”Report

        • Avatar Brandon Berg in reply to Jaybird
          Ignored
          says:

          I can’t remember where I originally saw this (scroll down to the red/yellow/green matrix), but this model explicitly takes ventilation into account.

          I found this NYT article as well, which stresses the inside-outside angle pretty hard.

          This chart got picked up by local news all over the country, though I didn’t find a national news source.Report

          • Avatar Kazzy in reply to Brandon Berg
            Ignored
            says:

            Well, in NY/NJ outdoor dining was allowed well before indoor dining was. So someone somewhere decided that an activity that requires the removal of masks can probably be done safely outdoors but not indoors.

            My (NY-state based) camp had different rules for indoor versus outdoor activities, so we did everything outdoors.

            So while we are still learning about this coronavirus, I do believe there is enough established evidence about how airborne respiratory viruses spread to determine policy about indoors versus outdoors.

            My understanding is that the reason we have a cold-and-flu season is not to do with the weather itself but because the weather often forces us indoors to spaces with limited air flow.

            So, least ’round these parts, the people with the power are saying, “Indoors bad, outdoors good,” as it relates to coronavirus transmission.

            Related to cold-and-flu season, I heard (though didn’t dig into) that the southern hemisphere was experiencing record low rates of the flu during their cold-and-flu season because of all the social distancing practices employed to halt the spread of Covid. So we may just avoid the double whammy of a winter wave and a flu wave? We’ll see.Report

  3. Avatar Oscar Gordon
    Ignored
    says:

    ThTh4: I’m not sure I’d say COVID is to blame for this. Peer Review By Press Release was a thing before COVID.Report

    • Avatar Brandon Berg in reply to Oscar Gordon
      Ignored
      says:

      And honestly, peer review isn’t all its cracked up to be. It helps, of course, but deeply flawed papers make it past peer review all the time.Report

      • Avatar Michael Cain in reply to Brandon Berg
        Ignored
        says:

        I did peer review for the IEEE for a couple of years. Doing it well is an enormous amount of work. My managers were willing to let me do most of it on company time, which helped. Doesn’t help that so many of the papers are written by people in publish-or-parish mode, and you feel guilty about rejecting them.Report

        • Avatar Brandon Berg in reply to Michael Cain
          Ignored
          says:

          Under what circumstances is a paper flat-out rejected? Major methodological issues that can’t be addressed without starting all over? Just not very interesting?Report

          • Avatar Michael Cain in reply to Brandon Berg
            Ignored
            says:

            Speaking just for myself, my short list for recommending a paper be completely rejected were (a) so badly written I couldn’t tell if there was anything new and interesting in it, (b) didn’t have anything new and interesting, and (c) didn’t fit the theme an editor had set for a special-topic issue. In the last case, I’d add a note that said it might be perfectly fine in another situation.

            In hindsight I was fortunate that I was reviewing engineering papers where measurements of smaller/faster/stronger/more accurate are generally straightforward (which is not the same as simple or easy). I would really dislike having to judge papers where the effects being measured are so small that most of the discussion is about measurement methodologies and/or subtle statistics.

            Another point that Oscar Gordon raises from time to time is the publishing bias towards positive results, particularly for medical stuff. It’s much harder to publish papers whose conclusion is “the null hypothesis that there’s no effect/benefit is not rejected.” I suspect that one of the most valuable tools some of the big drug companies have is a database built over decades identifying molecules that didn’t work, almost none of which has been published.Report

            • Avatar Brandon Berg in reply to Michael Cain
              Ignored
              says:

              It seems to me that one problem with eliminating the publication bias in favor of positive results is that the correct interpretation of small studies with negative results is that they’re inconclusive.

              There was a recent study of the effects of Arkansas’ Medicaid work requirements on employment, and the abstract says “work requirements did not increase employment over eighteen months of follow-up.” But if you read the full study, you’ll see that what they actually found for the employment effect was a 95% confidence interval ranging from -8.5% to +6.5%. The correct interpretation of that is that we have no idea what the effect on employment was, other than that it was probably in that range.

              Big studies with negative results generally get published because they’re expensive and generally have enough statistical power to rule out a substantial intervention effect.

              Maybe the solution here is not to pay more attention to studies with negative results, but to stop taking small studies with positive results so seriously, and just treat them as preliminary results that need larger preregistered replications before we start acting like the results are true.Report

              • Avatar Oscar Gordon in reply to Brandon Berg
                Ignored
                says:

                We still need to pay attention to negative results. If nothing else,we should have a public database of negative results, if only to avoid blindly treading over the same ground over and over.

                Otherwise, yes, small studies that suggest positive results, but are inadequate to confirm such should not get much press. When the institution pushes the PR hard, that’s when I suspect they are trying for peer review by press release.

                I mean, I give a lot of science reporters credit in that they do state clearly that it’s a small study, or a preliminary study, and it should not be taken as gospel. But others are not so honest, and a lot of media consumers are just fecking clueless regarding what constitutes significant results, and what is just a positive lead.Report

  4. Avatar Swami
    Ignored
    says:

    ThTh3 is some of the best news I have heard in a while. Not only is nuclear the best short term option for combatting AGW, I also like the way the new reactors are smaller, cheaper and more flexible.

    In Ridley’s new book How Innovation Works, he makes an interesting observation on nuclear. He notes that successful innovation usually comes about via trial and error and incremental tinkering, and that this is the antithesis of the nuclear energy model.Report

    • Avatar Michael Cain in reply to Swami
      Ignored
      says:

      More today’s model of how electric power delivery is going to be organized, for which we can blame FERC. FERC’s particular market (and to a lesser extent, reliability) models leave no room for nuclear tinkering. They leave no room for the kinds of network management needed to make renewables successful. There’s no hope of a no- or low-carbon power grid unless Congress imposes drastically different goals on FERC.Report

  5. Avatar Gabriel Conroy
    Ignored
    says:

    ThTh5 [“heartbeat”]:

    You may have heard something about a “heartbeat” being detected in space. It’s not an alien. It may be something more amazing:

    I don’t know. I think a heartbeat from an actual alien would be pretty amazing.Report

Leave a Reply

Your email address will not be published. Required fields are marked *