Month: November 2022

This is Charlie sitting on the fence with a nice patch of sunlight framing his face and showing off his two-tone whiskers. He's such a photogenic cat.

I've written a lot about crime over the past month or so. Here's a summary of the most important bits. First, crime has gone down steadily over the past decade. Property crime continued to go down in 2021 while violent crime remained stable.

In 2022, the largest cities in the US almost all reported lower murder rates and only smallish increases in violent crime. New York City is the sole outlier, and its numbers are iffy.

If there's nevertheless a genuine fear of rising crime, it should show up in concrete actions taken by consumers. But it doesn't. Google searches for home security devices have gone steadily down over the past few years.

Perception of rising crime is highly partisan and very recent. It started among Republicans in 2021, after Joe Biden was inaugurated.

Taking all parties together, overall perceptions of crime have been down consistently over the past decade. This changed only in 2022, when news media reports and Republican campaign ads began to insist that crime was out of control this year even though every indicator suggests that property crime is down and violent crime is up only slightly.

The Gallup poll results are easily explained. Fox News cynically began running sensationalized reports on violent crime beginning in 2021, and then almost instantly pulled back after the midterm elections of 2022 were over.

Bottom line:

• Property crime is down over the past decade and has continued to fall this year.
• Violent crime is also down over the past decade and is up only slightly this year.
• Perceptions of crime were consistently modest during this time.
• Perceptions changed only after Joe Biden took office. This was thanks to deliberate manipulation of crime coverage from Fox News.

This is a sculpture of a pale, hairless hand holding a bunch of colorful marshmallows on sticks.

Wait. No. It's a Jeff Koons sculpture "donated" to the city of Paris after the 11/15 attack that killed 130 people, and those are tulips, not marshmallows, as anyone who follows Koons knows.

Ahem. I guess I don't know my Koons. Anyway, it turns out that "donate" meant Koons wouldn't take a fee for creating the artwork, but neither would he pay to have it constructed. Eventually, however, a bunch of private donors provided about $4 million toward the fabrication and assembly of the 34-ton metal sculpture and Koons added $1 million to that. After a bit of tediously predictable controversy over whether Koons was sincerely memorializing the victims of the terrorist attack or was merely being a publicity hound, Bouquet of Tulips was installed in 2019 near the Petit Palais.

NOTE: This has nothing to do with Thanksgiving. I didn't have pictures related to Thanksgiving to put up, but I suppose you could think of the Koons sculpture as, um, a bunch of turkey gizzards?

Speaking of newspaper stories about Thanksgiving, the Guardian gets it right this year. And so do the turkeys of Woburn, Massachusetts:

This is not the most inspiring Thanksgiving front page I've ever seen. I guess it could be worse, though. At least there are no recipes for cauliflower mashed potatoes or Beyond turkey slices.

It's dex night and I would like to be out in the desert with my telescope since the moon is down all night. Unfortunately I'm sick and I hurt my leg and the weather is apparently lousy, so instead I'm sitting at the computer clicking aimlessly at whatever catches my eye.

This means I'm going to torture you with another Ngram chart. This one compares how often English-language books mention various sources of news and entertainment:

(Actually, the most common thing books mention is books. Surprise! But I took it out so the others would show up better on the y-axis.)

The high point for radio is 1945, which is unsurprising. However, the high point for television is around 1958, after which it dips and then grows only slightly. This is sort of peculiar, no?

Newspaper has been slowly dropping since 1940 but then flattened in 1980. That was about the time the newspaper biz started to crater, so apparently being mentioned in books isn't correlated much with financial success.

Magazine has been pretty steady the entire time.

But then we get to the really surprising thing. With the exception of newspaper, everything takes a huge dive starting around 2000 or so. This includes both web and internet, which hardly seems possible.

Why? Your first guess—and mine—is probably that this is just some artifact of Google's database. But I tried out a whole bunch of ordinary words and there was absolutely no consistent decline starting around 2000. Every word acted fairly normally, which suggests the database is reasonably reflective of reality. The only words I found that plummeted were those referring to information and entertainment sources.¹

I'm sure there's an explanation for this, but I can't think of it. Web and internet are the most spectacular losers, which makes no sense at all, but radio, television, and the normally reliable magazine all start dropping more steeply too.

One possibility for the peculiar drop in mentions of web and internet is that generic references have given way to specific mentions of services like Facebook and Twitter. Sure enough, if you add up every reference to either the internet or a specific internet service, it comes to about 0.019% in 2000 and 0.014% in 2019. That's still down, but I only checked the dozen most popular social media sites, and the number would be much higher if I included other specific websites. Still, none of the specific sites started to get mentioned until about 2005, while references to internet and web began dropping in 2001 and 2003 respectively. I suppose that could have something to do with the dotcom bust, though I feel like I'm stretching here, and anyway, it doesn't explain the decline of all the non-internet words.

So what's going on? Is the lesson here that the Ngram viewer just isn't reliable? Or do books really mention internet half as much today as they did in 2001?

¹Except for book, which books continued to love talking about. In fact, mentions of book went up between 2000 and 2019.

This is off the beaten path, isn't it? But I was over at Google's Ngram Viewer trying to remember what I had planned to look up, and while my neurons were betraying me I started entering dirty words. That was interesting, so I got more rigorous about it. Here are the most popular current obscenities, sorted by how much their use has grown in English-language books:

Generally, speaking, the use of obscenities in books was close to zero before 1960, and didn't really get off the ground until the 1970s. So that's when I started. The ending point is 2019 because that's as far as Google's data extends.

The winner doesn't surprise me. Fuck and its derivatives are far more widely used on an absolute basis (though shit and damn take the two top spots), but asshole has grown more because it was barely used at all in 1970. If we had a way of knowing, I'd bet that its growth has been even more astronomical in spoken language.

As I've said before (but I can't remember when), I believe that this is because asshole is such a tremendously useful word. Unlike fuck, which has good alternates if you want them, asshole really doesn't. There are plenty of close alternates, but none of them say quite the same thing. It's really a great, great word.

NOTE: I included the four words on the far right of the chart, but they all have fairly common, non-obscene uses that overrun Google's database. This is most likely why their growth has been slow. If we had some way of measuring only their use as obscenities, I'll bet their growth rate would be higher.

The White House sent out some talking points to help you deal with your Fox-loving uncle on Thanksgiving Day, and I wish they hadn't. This is one of the most tiresome memes ever.

But if you're going to do it anyway, for God's sake please proofread it:

Two typos in a single sentence? It should be "No NEW taxes on people making LESS THAN $400k."

On the bright side, I can think of plenty of lunatic conservatives and billionaires who would be pretty delighted by the original formulation.

This is Seal Beach, California. Sadly, there are no longer any seals there.

David Dayen reports today on a letter from the "Blockchain Eight" suggesting the SEC should back off any investigation of crypto trading exchanges:

The March letter from eight House members—four Democrats and four Republicans—questioned the SEC’s authority to make informal inquiries to crypto and blockchain companies, and intimated that the requests violated federal law.

Rep. Tom Emmer (R-MN), whom the Republican caucus just elected as majority whip, the number three position in the House GOP leadership, led the letter. In a contemporaneous Twitter thread, Emmer wrote: “My office has received numerous tips from crypto and blockchain firms that SEC Chair @GaryGensler’s information reporting ‘requests’ to the crypto community are overburdensome, don’t feel particularly … voluntary … and are stifling innovation.”

....The SEC also asked FTX about a rewards program that gave depositors interest on their crypto assets, which could make them a security. SEC chair Gary Gensler has been adamant that crypto platforms are trading and minting securities, and that these securities needed to be registered with the agency.

FFS. If it walks like a security and it quacks like a security, it should be regulated like a security. Ditto for banks. And insurance companies.

When are we going to learn this?