HWFO Reading List, July 2024

And some discussion about spiking

I really have been writing articles all this time, you just don’t see them because I keep spiking them.

I wrote a really great one called the “Devon Eriksen Toaster-Fucker Supposition” outlining a variant on echo chamber amplification stylized as “basic toaster fucker theory” and applying that to modern internet media engagement mechanics, but then some readers proofing it correctly indicated that a large body of my reference material could be considered homophobic, and it would need a rewrite. So I spiked it.

I got half way done with “Project 2025 Has Some Cool Stuff In It,” but then I realized that in order to do a full accounting of the 950 some odd pages of Project 2025 it was going to take me a month, so I spiked it.

I got half way through a treatise entitled “Welfare Needs Walls,” about how uncontrolled immigration undermines welfare states, and the only way to make welfare work is to put bounds on the group you were trying to give welfare, mathematically, figuring that’d be a fun way to trigger both the reds and the blues. But deep into my research on that one, I discovered that one of the foundational examples I’d planned on using was wrong. According to the latest sound research, most fentanyl street junkies in San Francisco are not out-of-state transplants like I’d heard, but instead appear to be 90% home grown. Which means SF has created an entirely different class of problem with their junkie-amelioration policies, but that class of problem didn’t support my thesis, so I spiked it.

I apologize for not producing more good material in the past month and a half, but one thing I want to stick to is to only produce things that are defensible, instead of just things that sound right or cool, which means sometime I have to spike things. I can honestly say I’m trying, and if you want a window into the efforts to produce more, join the Slack server where we talk about this stuff.

If you would like me to revisit one of the above spiked articles then subscribe, join Slack, let me know, and I’ll revise the one you pick to be publishable. Toaster-Fucker is probably the closest to a publishable state. You’ll also gain access to a group of over a hundred people all talking about stuff like outlined below, as well as things like presidential assassination attempts failing so miserably that they knock the other guy out instead.

Reading List

Curious Grant Proposals

I’ve never written a grant proposal, but I imagine proposals to confirm the dominant narrative probably get approved more often than those which disconfirm it. We have a few disconfirmers this month.

A scientific paper looked into the hypothesis that banning sulfur from marine diesel increased global warming by decreasing aerosols. It found significant effects.

“Gun people have smaller penises,” the great mantra of the bigoted antigun left, has now been scientifically disproven. Imagine writing that grant proposal.

This preprint is fairly interesting, which assayed thousands of academics on twitter and tried to determine whether they were more or less freakouterist than the general population. It shows lots of interesting results, including that STEM academics are more muted in their online opinions while humanities academics are pure nutballs.

In contrast, the grant proposal for this study was assuredly much easier to write. Historical tornado mapping shows that the hot spot for tornados has shifted away from Oklahoma and into Mississippi, much to the chagrin of tornado chasers who’d prefer to chase where there are fewer trees.

But of all the grant proposals difficult to obtain, the one for this study had to be the hardest. The study identifies a systemic policy by the British Medical Journal to viciously self-censor and eliminate any possibility for debate about Covid-19 policies, particularly when it came to lockdowns. Vinay Prasad goes into it in depth here:

During the COVID pandemic, the British Medical Journal massively published zero COVID zealot views & seldom the other side

Nordic Fun

This fascinating article in WSJ (paywall breaking link here) goes deep on how Sweden maintains a waiting list for military service.

This Swedish Substack author outlines in some detail the reasons I write my stuff - because writing is the act of thinking things through.

How to think in writing

This Danish quantum algorithm specialist buried a Large Language Model AI into a Truetype font.

More AI Stuff

Wired wrote a story about how Perplexity.ai was a “bullshit machine,” which I personally find to be objectionable because Perplexity.ai is the best Google replacement going. Then Wired came back and used Perplexity to ask Perplexity about Wired’s article and kited Perplexity into plagiarizing Wired’s article. three takeaways from this, for me, are these:

1. Perplexity may be doing some unethical things, specifically in regards to scraping things they say they’re not scraping or aren’t supposed to be scraping, in order to make themselves the best AI engine out there for answering questions and providing references to those answers, but also..

2. Perplexity.ai is the best AI engine out there for answering questions and providing references to those answers. You should try it out as a Google replacement, it’s very good.

3. Someone at WIRED probably got fired from Perplexity and they’ve got a hard-on for writing these articles.

A Facebook friend of mine is one of the big wigs at Palisade Research, which has just released the weirdest Chrome extension I’ve ever seen. FoxVox is a button that you can install which will use AI trained against conservative and liberal news media to convert the language of any page you go to into a conservative or liberal coded variant of the page. They have demos on that link where you can convert people’s tweets, CNN articles, and similar to completely different language intended to target and persuade political demographics.

While my buddy Jeffery at Palisade is assuredly noble in mind, soul, and intent, and the FoxVox website states that it’s a demo to try and warn people about AI’s capabilities to exacerbate siloed realities, improve propaganda, expand hidden biases, and contribute to the further breakdown of trust, I view this project as Torment Nexus Adjacent.

Someone at Palisades Research is going to end up making a shitpile of money carrying this code off to CNN or Fox and doing the exact thing with it that Palisades is trying to warn us about.

Esoterica

We share a lot of Jeff Asher stuff around here, and this one specifically debunks something I’ve seen floating around my media circles. Yes, FBI crime reporting began to suck in 2020, but no that’s not the reason crime appears to be dropping.

Did 6,000 Agencies Fail To Report Crime Data in 2022?

…and related material from Jeff:

Is The NIBRS Transition To Blame For Our Current Crime Trends?

This TERF war update shows the Biden administration flexing their muscles to twist the international standards of transgender care more towards what would win them more Twitter arguments in the USA.

Unsealed Court Documents Show That Admiral Rachel Levine Pressured WPATH To Remove Age Guidelines From The Latest Standards Of Care

One day I’ll write the full HWFO analysis of school voucher systems. But the short version is this:

1. The liberals are right that most or all of the ones I’ve seen are basically just a private school tax break,

2. Private schools look better mostly because they choose not to let in problem kids,

3. In order for vouchers to work as intended, they’d have to be full-value only, and any school accepting a voucher must accept all vouchers without academic or other discrimination.

That’s not how the Arizona voucher system was set up.