linearithmic

testing beliefs against reality

Remote work buys time for housing reform.

forecasting bots will transform the field

Yes, great comment. Getting adoption will be a challenge (although the concept might appeal to fledgling applications).

it’s not a technical limitation, it’s a business model

I found the argument of the counterfactual to be persuasive. Do you have a sense of how much of the time would people actually be doing something better with the extra bandwidth?

Possible, but seems high to me. I'm not sure if those planes were really destroyed or just damaged. My knee-jerk estimate would be somewhere in the hundreds of millions. Still a very successful operation by Ukraine though.

I'd be curious about forecaster estimates of monetary damages from the drone strikes.

Good point. For me the issue is the variance which governs the time that I have to leave. For the subway, if I give myself a 10 minute margin beyond the median time, I'm quite likely to be on time, for walking or biking if I give myself a 5 minute margin, I'll almost certainly be on time. For driving on the other hand... as per the article, a trip with a median travel time of 17 minutes can take me over an hour when traffic backs up.

Thanks for the feedback! I'll try to add something about automated cameras and tag violations (which I didn't address) in my next deep dive.

I see these changes as expanding freedom, not reducing it. Car ownership costs thousands annually, which many young and low-income New Yorkers can't afford. Improving buses, creating safe bike lanes, and making parking more efficient gives everyone more affordable transportation choices. From my perspective, the current system restricts freedom, which excludes precisely the young and less affluent.

The idea was not that San Francisco does everything perfectly, but instead that there are things they've tried that have been demonstrated to be successful we can learn from in NYC.

I actually own a car in NYC and think these would make things more practical. I would happily trade time spent looking for parking for a minor fee and would happily trade time spent in traffic for a toll.

Even if every housing reform went into effect tomorrow, construction timelines mean it could be years before we see meaningful improvements in affordability or availability. To improve quality of life and opportunity in the near term, we need high-leverage, low-friction interventions that reshape how people live and move right now.

Mobility is the highest leverage near-term option we have.

seldom needed, but don't neglect it when it is

connecting the dots with network theory

even if progress stalls (and it won't), we've only started to tap its potential

why it's important and a process for doing it

haha, your point is taken that democratic guidance will have its problems. The idea I wanted to push though was that even if AI can do all work, people will still have the option to do something that they will find meaningful and important. The attempt was to illustrate a potential counter-narrative to the idea that human existence will become pointless / unnecessary after the advent of advanced AI.

Our most significant responsibility won't be competing with AI, but rather in deciding what should be done.

A post about predicting tomorrow's predictions.

LLMs can make a developer’s job easier and faster. When might they make them obsolete?