gbhn

There are roughly a million violent crimes a year. It is not credible there are 3x as many defensive uses of firearms that are somehow being missed.

The page you reference indexes a few thousand over several years. Seems likely an undercount as the page claims but not by that much by probably a couple orders of magnitude.

A good example of this is comments on this post. They're basically awful and reduce worthwhile traffic on this site by driving away more people than they attract.

The reality of every forum, as the original author says, is that unless you censor topics that attract cranks, you'll eventually become a qanon platform (or the equivalent) and drive out all worthwhile participants.

I'm not sure what the global rule is called, but something like volume nonlinearity or kookiness asymmetry it something. If you have a bunch of looks everyone else leaves. You're options are to get rid of them or fail. It sucks but there it is.

Ultimately wouldn't these rates be set politically? As in, there'd be one politician who says "raise the tax and do this" and another who says "lower the tax and stop doing that" and voters just pick what they want to happen.

Another key factor is enforcement cost. Enforcing a land tax is really easy for a government. You just... Go there and get it. There's no moving the land offshore or having complex relationships or offsets. If someone won't pay the government just assigns the ownership to someone who will and goes on.

Compliance costs with many other taxes are quite high.

That's kind of the thing that makes APIs possible, right? It sounds to me like "what if programming were done in a completely flat global namespace in which abstractions, encapsulation, and structure were impossible."

I think the theory is "there are few enough we didn't notice."

There's other weird effects going on -- some countries that look really really similar have vastly different experiences with the virus. (Canonical example is DR and Haiti, although the situation is dynamic enough no-one knows if that'll last.)

My takeaway from the difficulty in measuring this thing is that there is pretty high variance in transmission. Some carriers spread it REALLY well. Some carriers spread it poorly. With a lot of variance it matters a ton to figure out what is the deal with high-transmission situations. My take so far is we have some clues but no certainty. (Seems to spread in public transit and healthcare environments more rapidly, but slower than you'd expect in schools and prisons.)

It'd clearly be a gigantic win in terms of intervention policy to understand this better, and we simply cannot do that without extensive contact tracing so we know what's going on. They're doing a pretty good job of this in China, HK, Taiwan, South Korea. As nearly as I can tell we're doing a dismally poor job in the US and Europe.

You read papers that have diagrams of restaurant tables and bus seating charts from China. In the US it's hard to get stats on aggregate nursing home vs non-nursing home fatality rates. Maybe it takes some time for these papers to come out, but I don't see evidence that high-quality pre-pub contact tracing data is being relied on to develop policy responses.

From the discussion on this I've read, I think a good direction would be to consider statistical tests like this as simply not "publishable" at all the way we currently think of it.

That is, if you have a theory about how a Gene relates to height on tomatoes, and you do a test, that test would show you you're likely on the wrong track if it falls below some p value, but the only thing it tells you by being above is that "there may be something here."

I think this is true for many fields with a replication crisis. The problem isn't statistical, the problem is no theory. If you have a functional theory there's all kinds of things you do to gain confidence in it, and mostly those will contribute to the ability to predict statistical results, but that is completely different on kind than sending out a survey and noting that question 2 and 6 are statistically correlated.

When a field thinks that the kind of early suggestive work like this is worth talking about, they should probably just talk about it in conferences and similar venues, rather than "publish" it where journalists will pick it up in a "science shows" story that 95% (lol) of the time turns out to be wrong.

In other words, I think it is fine that fields talk about early non-theory results -- that can be interesting for specialists to advance faster. "Publishing" this mostly-going-to-be-wrong stuff is leading to confusion among the public about what the scientific process demands and how trustworthy it is. That is not a good outcome in my opinion.

Are you suggesting the Chinese market is more competitive due to the government being big enough that companies in the market have to exist in an impartial regime? This isn't the impression I get from many other sources.

A local coin/stamp person was completely uninterested in my old stamps from like 35 years ago. I think the shrinking number of collectors is outstripping the number of collections getting thrown out, and the whole idea of collecting stamps is just fading away.

By "open standards people" I mean "people who prefer open standards". Nothing more sinister than that. :-) (I am one, BTW)

This is already a very small subset of the population -- they've bothered to take a stand on something virtually no-one seems to care about. (At least not directly.)

This last part seems crucial to me. You're right that people care that they can chat with friends. It's the clause I think they glaze over at. Something about applications? I just want to chat!! Of course standards folks, being experts in this, care a lot, but messaging is a canonical example of this. I've personally witnessed years of strategy aimed at open standards go nowhere while proprietary standards win out. It's not because billions of people have strong preferences for open standards!

I do think that there are enough open standards fans, and enough open standards-adjacent folks, to make a strong play for something like messaging. But it's not "abstract XMPP" that's the problem -- it's "real XMPP" with fractures, federation headaches, delays in having it be caught up with capabilities of closed platforms, etc. That's the actual competitor. :-(

Fatalities per passenger mile is a good metric for passengers, but for the FAA, I think crashes per vehicle mile is a better target.

That is, the plane isn't inherently 100x safer if you just fly it empty.

I have some direct experience with this. It is summed up as "open standards people are extraordinarily hard to work with, uncompromising, and subject to infinite fine splitting of the space. Meanwhile, 5 billion users do not care and want things to be easy, fast, work well, and be well integrated."

The result of these two forces is that it ends up being extremely difficult to work on open standards for any kind of data that people care about. (Users can't see HTTP, so it is fine to standardize. They can see blog titles, so... no dice.)

The authors' point seems to be "if you ask people questions whose real answers correspond to informal statistical judgments, then they will give the right answers."

This isn't really that informative, but the strongest form of this argument is "informal statistics is better in a lot of real-world cases." This sounds to me quite a bit like the case the behavioral economists make -- that these intuitions aren't crazy, or stupid, but they do exist. They were constructing tests specifically to isolate them, and succeeded.

A further point is that these kinds of situations when intuitions are challenged happen regularly. I don't know what behavioral economics says specifically about causes, but it seems to me they take a fairly neutral view: such situations can occur by chance, due to hostile intent, or by mistake. Whether the outcome is positive or negative depends as well. The point of "nudge policy" (opposition to which seems to be the main point of the article) is to see about making these situations positive -- that is, arranging the world through policy such that making intuitive considerations actually yields positive outcomes. I don't think the authors' argue successfully for the point that since everyone is rationale and no-one ever gambles when presented with the opportunity, that therefore arranging policy such that intuitive choices yield positive outcomes is unwarranted interference.

The point of the article is that your intuition is incorrect. By inhibiting zipper merging you are forcing congestion backward, and leaving carrying capacity of the road unutilized during congestion when it is most at a premium.

Pushing congestion backwards is exactly the cause of a lit of traffic jams. I've observed this myself every single GD day on the 210W freeway between Hill exit and the 134 exchange. People get over way early, causing huge congestion to the merge entrances. If they'd stay left and zipper, everyone in the jam would save about 5-10 minutes. This is thousands and thousands of dollars a day of cost in completely avoidable congestion in just one two mile stretch of highway ( albeit it a particularly egregious one)