g_sch

These cameras may have been originally sold to municipalities as a way to find stolen cars, but from one year to the next, federal agencies have (1) decided that their main goal is finding arbitrary noncitizens to deport, and (2) that they're entitled to the ALPR data collected by municipalities in order to accomplish this goal. The technology isn't any different, but as a result of the way it was deployed (on Flock's centralized platform), it was trivial to flip a switch and turn it into a mass surveillance network.

I think this is equally true of writing. Once you see something written one way, it's very hard to imagine other ways of writing the same thing. The influence of anchoring bias is quite strong.

A strong editor is able to overcome this anchoring bias, imagine alternative approaches to the same problem, and evaluate them against each other. This is not easy and requires experience and practice. I am starting to think that a lot of people who "co-write" with ChatGPT are seriously overestimating their own editing skills.

I sympathize with people who find writing difficult. But, putting myself in GP's shoes, I can't imagine trying to read my father's LLM-generated memoir. How could I possibly understand it as _his_ work? I would be sure that he gave the LLM some amount of information that would render it technically unique, but there's no way I could hear his voice in words that he didn't choose.

If you're writing something for an audience of one, literally nothing matters more than the connection between you and the reader. As someone with a father who's getting on in years, even imagining this scenario is pretty depressing.

What I'd really like to know is:

- how many people that currently work at your company had to go through an AI interviewer to get the job? - do referrals have to go through an AI interviewer too?

To me, this just smacks of a tool that increases the cost of cold-submitting your resume so companies can optimize for "preferred" hiring paths likeinternal referrals.

Fan/AC remotes have some of the worst, most unexpected UX I've ever seen. I still remember an AC remote at an Airbnb I stayed at a few years ago that, instead of having temperature control buttons that said "up"/"down" or "hotter"/"colder", it had two buttons that said "too hot" and "too cold". The "too hot" button decreased the temperature setting on the thermostat and the "too cold" button increased it.

This labeling makes some sense if you stop to think about it, but I can't shake the thought that this was the first time I encountered an interface that asked me to describe how I was currently feeling, rather than me telling it what I wanted it to do.

The author is pretty upfront that the conclusion was meant as a reference to a character in the movie "Hoosiers", not that any of the personalities named were literally drunks or bad fathers.

I also recently had this experience! I remembered a recurring bit from an older comedy film (a customer in a shop keeps saying "Kumquats!") and tried to prompt ChatGPT 4o into getting it. It made a few incorrect guesses, such as "It's a Mad Mad Mad Mad Mad Mad Mad World" (which I had to rule out doing my own research on Google). I found the answer myself (W.C. Fields' "It's a Gift") with a minute or so of Googling.

Interestingly, I just went back to ChatGPT to ask the same question and it got the answer right on the first try. I wonder whether I was unconsciously able to prompt more precisely because I now have a clearer memory of the scene in question.

Notably the jury pool was from North Dakota, a state where many people have direct or indirect ties to oil and gas industries. In fact,

> During jury selection, potential jurors appeared to largely dislike the protests, and many had ties to the fossil fuel industry. In the end, more than half the jurors selected to hear the case had ties to the fossil fuel industry, and most had negative views of anti-pipeline protests or groups that oppose the use of fossil fuels.

Source: https://archive.ph/OMEYP#selection-1559.0-1563.204

The same type of people who name their companies Palantir and Anduril

Matt Levine had a funny similar take recently:

"You could have a model of Harvard Business School that is like:

    1. Harvard Business School teaches you skills that would make you good at running a company.
    2. There are lots of companies that could use those skills.
    3. But you don’t want to run those companies, because they make, like, ball bearings.
    4. You want to run a fancy company; you want to run a hedge fund or a tech startup or something.
    5. Meanwhile, the people currently running the ball bearings company would not be all that excited about you, a fresh-faced business school graduate who has never run anything, coming in to run their company, even if you did learn a lot of useful skills at Harvard.
    6. Therefore various industries exist whose principal business is laundering ball bearings companies into opportunities that appeal to Harvard Business School graduates. You wrap the ball bearings company in a name like “private equity” and suddenly it is legible to the Harvard students, so they flock to it.
    7. Those industries are also in the business of getting the ball bearings companies to accept the Harvard Business School graduates, which in practice means not so much “make the ball bearings company excited about its new Harvard CEO” but rather “buy the ball bearings company and install new management.”

Source: https://archive.is/8IUCA#selection-1795.0-1869.303

Yes, but what I'd really like to see is a visual graph of the dew point over the course of a day. The graph currently only shows RH%, and the only way to see how the dew point will change over the course of a day is to drag your finger over the RH% graph and watch the dew point number change.

The feature I miss most about Dark Sky was that it allowed you to visualize changes in dew point throughout the day.

Where I live (US East Coast), the weather can feel dramatically different depending on the humidity. Relative Humidity has always felt to me like a poor way of measuring how humid the weather will feel. For example, 50% RH at 84ºF will feel lightly humid and generally pleasant, whereas 50% RH at at 97ºF will feel like a swamp. The dew points at those respective points - 63ºF and 75ºF - do a much better job at immediately conveying how humid the air will feel.

Dark Sky used to show hourly dew point graphs that you could browse throughout the week and see when the humidity would break (or return). Apple Weather does show you the dew point, but only when you select a point on the RH graph. So to track the dew point over the coming week, you basically need to drag your finger over each day's graph and observe the changing numbers.

I think this is probably just due to the lack of general awareness about how dew point is a more elegant shorthand for "absolute humidity" than any other weather metric currently in use. I hope there will be more of us in the future!

> in the last 10 years.

From the article:

> In 2014, amid flagging sales and pressure from investors, Darden sold Red Lobster for $2.1 billion to Golden Gate Capital, a San Francisco private-equity firm.

Back when I was in consulting, I used to think of it as a perk (as did many of my peers). Once the travel started to wear on my personal life, I ran the numbers and discovered the miles and points I was earning equated to only around $200-300 per month in cash equivalent value.

It's really surprising to me how intensely some people will pursue relatively worthless airline miles. I suppose if you're going to be traveling anyway, you might as well pick them up. But if you have the choice, it's not really worth the trade-off.

I also stopped being able to read articles on archive.is about a week ago because of this captcha loop. It seems to be related to my device and browser - my Firefox gets hit with the captcha loop, but Chrome on the same machine loads the article fine, and DuckDuckGo Privacy Browser also loads the page on my phone.

As noted in the linked article, people in urban areas are paying many times the price for water that farmers are. If farmers had to pay municipal water prices, it's much less clear whether their farming would be profitable.

The key here is distributed coordination. In the last few decades, the most effective social movements are those that did not rely on a single organization (hierarchical or not) but instead allowed multiple organizations and actors to work together separately and embraced a diversity of tactics.

In the 50s, John Boyd did a lot of really interesting work on the idea of "mission command" and the OODA (observe, orient, decide, act) loop. Although he did his work for the US military, I think his ideas are broadly applicable to social movements as well, as they describe how "swarms" of autonomous actors can easily overwhelm hierarchical organizations due to their ability to make decisions faster.

I don't believe this is true on iOS, unless there was a recent change I'm not aware of.

Right now, tapping power five times on my phone running iOS 16.1.1 opens an "emergency" mode, which allows anyone to call 911 or see my "Medical ID". It also has an option to "cancel" this mode, which returns the phone to a normal locked state but forces a passcode to unlock (i.e. Face ID will not unlock the phone in this state).

The linked article and this entire discussion is addressing the ethical issues that surround grading, originality, and plagiarism. How is AI going to be able to apply the "correct" ethical code when all of us can't seem to agree?

I think it's not fair to fault DB or EVG for failing to run 100% service after a nationwide strike was called off just a few hours before it was supposed to start. An orderly temporary shutdown of a major rail network is going to require plenty of ramp-down and ramp-up as trains and people get in position. Some of that was undoubtedly already underway when the tentative agreement was reached, and you can't just turn it around on a dime.

Archive link: https://web.archive.org/web/20230516100811/https://blog.plov...

Maybe I'm just a newbie when it comes to building search algorithms, but how on earth would you maintain or test something like this? From experience, I know that search algorithms using far fewer ranking factors than this are deployed in production and treated as terrifying black boxes that no one wants to touch lest they break something in an unexpected way.

At most non-fancy corner stores they're still under $4. Usually under $3 if you just want egg & cheese. And they're always made to order - ketchup, mayo, salt/pepper, hot sauce, on a roll or bagel, 50 cents extra for a tomato. I know New Yorkers can be a little insufferable about their bodegas, but I've lived in a few other major cities and none of them have provided me with this much access to a cheap breakfast.

It's true though, good breakfast burritos are still hard to come by here.

As other commenters have mentioned, people living on the street don't comprise the majority of unhoused people. It does seem like California has a higher overall percentage of "unsheltered unhoused" people among the unhoused. But e.g. here in NYC the vast majority of unhoused people move in and out of the (largely corrupt, poorly maintained, highly restrictive) shelter system, and a only small percentage live on the street or in the subway.

Also anecdata, but my upcoming payment is somehow 30% lower. Before that, it had gone up ~15% in two years though.

I think your story carries an unspoken assumption that the police understood and/or cared about the law. I wouldn't be too sure that's the case.

I suppose, although perhaps this analogy would be more apt if there were few to no single family homes in SF (as is the case with Manhattan).

I mean, SHFs barely exist in Manhattan so it's kind of a moot comparison. But the thrust of GP is correct - you can buy a townhome in much of Brooklyn and Queens for less than $1m.

Both What and Waffles were launched in the aftermath of the OPP shutdown. Oink was the only game in town when I first got into torrenting, save for a few small "scene" sites whose main claim to fame was releasing major albums a few minutes sooner than everyone else.

After OPP got shut down, both What and Waffles were stood up in a matter of days.

The writer here makes an error in assuming that what he terms "luxury beliefs" are invented by people with more money or influence. Ideas like police abolition absolutely have their roots in working class movements in places like the South Side of Chicago and the Bronx.

I think a more interesting analysis would have compared people whose lives are at stake in political fights vs. those who are well-off enough that they can participate as a sort of hobby (or, yes, as a way of accumulating cultural capital). For the latter group, their beliefs and their ability to participate in political debates are indeed a luxury. But I don't think this divide would mirror the US culture war as closely as the author seems to think it would.