Interesting Links for 26-07-2021

[andrewducker]

The UK government isn't including reinfections in their Covid stats

(tags:pandemic statistics OhForFucksSake )

The Best Time Travel Movies of … All Time

(tags:timetravel movies )

What's the moral (and pragmatic) approach to dealing with vaccine deniers and the damage they cause?

(tags:vaccination psychology conspiracy )

DNA Has Four Bases. Some Viruses Swap in a Fifth

(tags:virus dna )

Comments

[nancylebov]

Part of the situation is "social death"-- shaming people for admitting mistakes.

[calimac]

Yeah, the tweetstorm says that.

But who is it who is demanding we empathize and respect the vaccine deniers? I sure haven't been seeing any of that.

[andrewducker]

https://twitter.com/EBHarrington/status/1418287918695817222?s=19

The bit highlighted in yellow. I've seen other similar things saying we won't convince them if we call them idiots.

[calimac]

That was what I read that made me ask the question.

[ninetydegrees]

Well, in general, you won't convince anyone by calling them idiots :) Also, people often confuse empathy with sympathy. Understanding why deniers believe what they believe and the anger, fear, etc. that often comes with it is one thing, sympathizing with them is another. You can also tell someone they're wrong and still be kind. One doesn't negate the other.

Anyway, very interesting read.

[dewline]

I've come to believe that some people accidentally confuse empathy with sympathy, while others pointedly conflate the two concepts for effect.

[ninetydegrees]

Well put, thanks for giving me more food for thought. Also great pic :)

[mountainkiss]

Here is one perspective.

[calimac]

2) I am mystified by the line in the Groundhog Day writeup, "(and, yes, that is where the term comes from)." What term? What is the "where" that it comes from? Do they mean that the holiday Groundhog Day was invented by the movie? (It certainly wasn't.) Or are they trying to explain why the movie was named for the holiday? (Because it takes place then: is that so hard to figure out?) Those both being stupid answers, it probably means something else, but I can't figure out what.

[davidcook]

"Groundhog Day" now refers to two things - the holiday, which did indeed exist before the movie, and a situation where someone is trapped in a time loop until they perform the right actions / grow in the right way to end the loop, which is also referred to as Groundhog Day, and that term-for-that-situation came from the movie named after that particular holiday. Clear? :)

[calimac]

Really? I've oft seen the movie referred to in cases of constantly-repeating situations, usually IRL and consequently not involving actual time loops, with nothing said about how you get out of it (e.g. "every time I called the help line I had to go through the same routine: it was like Groundhog Day"), but note that the movie is used as a simile and not the name as an actual term ("it was a Groundhog Day"). That type of usage I have not seen.

[matgb]

It's so well used it's got a trope page:
https://tvtropes.org/pmwiki/pmwiki.php/Main/GroundhogDayLoop

Lots of shows and showwriters talk about groundhog day episodes or similar, I really like the Dark Matter one

[calimac]

Oh, TV Tropes. Yeah, they're famous for connecting things that it had never occurred to anybody to connect before they wrote about it. (Sometimes their connections are brilliant, some strained: this has got some of both. Waiting for Godot? Noises Off?)

I've read or seen a number of their items, and seen many discussions of them, but never have I seen anyone refer to one of these as "a Groundhog Day" without specifically identifying the movie (and often enough not even referring to it at all). Which was my point.

The UK government isn't including reinfections in their Covid stats

[doug]

It's true that reinfections are not included in the daily dashboard - but it says this clearly at the top of the dashboard and has done for as long as I can remember. But reinfections are monitored and tracked pretty closely and published regularly. They are less than 1% at the moment. This is not anywhere near a big deal, and is not having a material effect on the headline figures at the moment

See e.g.
https://twitter.com/kallmemeg/status/1419246354224848897
https://twitter.com/RP131/status/1419236888934764547

There is an interesting geeky issue about how best to account for reinfections in those headline figures - which must be published rapidly. All they get rapidly is essentially name (plus a few demographics and address/telephone if they're lucky) and date tested positive. The issue with just counting the number of positive tests is that one person can - and often does - test positive multiple times during a single infection. So they might get a positive on a home LFT test, then they get a confirmatory PCR, and then maybe they're one of those who track themselves with daily LFTs to see when it comes clear. You obviously only want to count all of those as a single infection, not two, or ten. When the proportion of reinfections was tiny, simplicity was the best option: you dedupe by only reporting the number of people with a positive test who've never had a positive test before. The great benefit of this is that it's simple and clear and relatively quick. But even this in itself is tricky more often than you'd think, e.g. because of imperfect matches for names/dob/etc. Problems with the process are one of the things that can lead to the daily stats issue being delayed. Which always creates utter havoc with people shouting that it's a cover-up and they're all evil - as if it's not easier and quicker to make up numbers than to take care to get them as right as you can.

Anyway. At the start of the pandemic and even until relatively recently, reinfections were negligible. (I remember when there was doubt they happened at all!) But now reinfections are rising, it's worth revisiting that decision. Anything you can do robustly and reliably on a daily basis is a judgement call. The current thinking is maybe to count a fresh positive test >90d from the last one as a fresh infection. But that time cut-off is quite tricky - long covid is a thing, although I think viral load tends to be undetectable later, but obviously not always. And of course there's the issue with continuity with the dataset published so far. In geeky stats terms you can just issue a new series and deprecate the old one while continuing to publish it for continuity, but in political terms it will be dynamite whatever you do. So there's an interesting and tricky issue under here ... but it's not one that remotely warrants what Peston was implying.

Don't trust Peston as your source on anything. He's well connected, but he sees his job as being first to pass on exciting things, with no proper analysis or checking. As a result, most of his exciting new things are completely misleading or outright wrong. I know that spreading misinformation is the last thing you want to be doing, so I'd suggest never passing on anything from Peston without doing the analysis/checking first. It is usually true that someone has told him what he's saying, but that's very much not the same as it being true and useful to pass on rapidly.

Vaccine deniers

[anna_wing]

I suppose it depends on how large a percentage of the population they are, how much social influence they have, and the attitude of both the political leadership and the majority of the population. In countries where they are a minority without significant popular or governmental support, it is sufficient to no-platform them, and ensure that access to public facilities and international travel for unvaccinated people is restricted, in the interest of everyone's safety. Eventually, once enough people are vaccinated and continue to be vaccinated for herd immunity to actually happen, they will be irrelevant. Though obviously the requirement of vaccination for travel should be retained, as for yellow fever.