joshuahedlund

> I liked season 1 of the Expanse - the settings and production were fantastic, but I found the characters just became overwritten and annoying as the show went on.

I recommend reading the original book series. In my opinion the writing only gets better as it goes on, and it’s obvious the writers had a plan in mind from the beginning rather than just writing more books for the sake of continuing a series where the ideas were all used up in the first book or two

It’s a bit cherry-picking, but right now BTC’s 5-year growth is 28% while the S&P is 67%.

To be fair bitcoin’s initial years were extraordinary. By any metric you use though the rate of increase has clearly slowed.

As a birdwatcher I strongly expect the benefit is in the identification experience and agree with this sibling comment:

> A big part of birdwatching is identifying small, fast-moving animals that exhibit specific combinations of behaviors at specific times of the year based on specific climatic conditions. That sort of rapidfire pattern matching from video and audio info is probably a big contributor to what keeps the brain sharp.

It takes time and experience to build up your identification skills but it’s really rewarding and exciting and modern tech makes it easier than ever to learn

> If there are no remaining items

In every software job I’ve ever had, every released new feature leads to maybe three new requests to modify the new feature, two unintentional bug tickets, plus some tech debt. Work leads to more work, not less. I don’t see how increasing the velocity of the work should change that.

Understanding is needed when something breaks and you need to fix it immediately. Otherwise you become a meat proxy and now you’ve added Anthropic’s uptime as a single point of failure to your application.

> It's still valuable to deeply understand parts of a program, but we don't have any tooling that helps us do that. We just have to raw-dog it by thinking really really hard and remembering how all the code connects together.

I deeply relate to this. When engineers were writing all the code that meant every part was deeply understood by _someone_ on the team, and they could valuably contribute to maintenance and further development. It wasn’t perfect, people leave, people forget things, etc, but the overall coverage was high and valuable.

Now, every agent-produced MR introduces code that is deeply understood by _no one_. It’s the “original developer left five years ago” problem, but now growing on every single new piece of code. Reviewing doesn’t give you the same depth of understanding, and the continually increasing impulse is to just approve, maybe nudge it about some isolated enum types or something, but don’t take the time to understand it, just keep the train going.

But then what happens when something breaks and the cloud agents are down…

> all the various hobbies that people can spend a lifetime enjoying

I disagree. AI took chess literally decades ago now, and more humans enjoy playing it now than ever before.

It (hopefully) might not be that easy. The world of bacteria and viruses is a complicated war zone with an arms race of defenses and billions of deaths every single day. It might take thousands of attempts to seed something that doesn’t just instantly die off.

Given that the outages are caused by record traffic, so far they are only cooked in the Yogi Berra sense of “Nobody goes there anymore, it’s too crowded”

Is this a result of too much activity on GitHub (AI-driven jump in repos), though, rather than a fall?

I may be a dumb layman, but that argument seems too tidy and absolute to me - wouldn’t it suggest, for example, that viruses could never appear because they’d be outcompeted by more advanced bacteria? Earth seems to have room for an incredible diversity of types of life.

Why haven’t I heard of this? Is it available in IDEs like Cursor?

> For equal capability tokens, there has been about a 10x drop in cost every 6 months

Is this still happening? Opus 4.5 was six months ago, can you get its capabilities for 1/10 cost now? Are we on track to get the same for 4.6 in a couple months?

For someone who purports not to care about the film, you seem very familiar with the discourse about it. (I have seen zero experts, or anyone really, discussing boat design.)

> thinkism

Thank you for this term. In my view, the belief that AGI singularly will rapidly destroy us because it will think 10,000 times faster than us is a form of thinkism.

> Last mile is already “solved” with the little robots that drive around cities, no need for hands.

And yet we haven’t seen widespread adoption because they can’t handle stairs, steep slopes, streets without sidewalks, sidewalks with mud, or a hundred other real world challenges

Sometimes it’s hard to objectively tell whether two animals don’t appear to reproduce because they are unable genetically, or technically able still but behaviorally unwilling in normal natural circumstances, or we don’t know but we just haven’t observed it for that particular combo, etc

It has, tho the rate of new record highs have been reducing from peak to peak: 10x > 3x > 1.5x

Any ideas why verified has stagnated? It was increasing rapidly and then basically stopped.

Scott Alexander blogged about it today: https://www.astralcodexten.com/p/best-of-moltbook

> If your job is to translate requirements into code manually - and that's it - you're the generalist travel agent.

I’ve been a full-stack web programmer at five different companies over the last fifteen years, big and small, e-commerce and B2B, junior to senior to staff, and that has never fully described my responsibilities.

I would love for SWE Verified to put out a set of fresh but comparable problems and see how the top performing models do, to test against overfitting.

> most normal people don't know what Claude or Gemini are

“Google Gemini” is the No 2 ranked app in the Apple App Store (behind ChatGTP) and has been for some time

https://en.wikipedia.org/wiki/Goodhart%27s_law "When a measure becomes a target, it ceases to be a good measure"

I'm also curious what results we would get if SWE came up with a new set of 500 problems to run all these models against, to guard against overfitting.

Won’t those models gradually become outdated (for anything related to events that happen after the model was trained, new code languages or framework versions, etc) if no one is around to continually re-train them?

How about denying the Fourth Amendment rights of US citizens to be secure in their homes in the recent Chicago apartment raid? https://www.notesfromthecircus.com/p/the-sufferable-evil

How about detaining US citizens without warrants for days at a time and then releasing with no charges? https://www.theatlantic.com/politics/archive/2025/09/george-...

There is a hard limit on the number of atomic elements, and an even smaller limit on the number of soluble compounds that facilitate chemical reactions, and water is demonstrably both the best and the most common in the universe.

So while it may be possible for life to exist without water, any alternatives should be reasonably expected to be even more rare than water-based life

I have stared at a Blue Jay mimicking while Merlin repeatedly labeled it Red-Shouldered Hawk. I’ve seen it pop up a bunch of suspicious one-offs around a mockingbird as well (same with Gray Catbird, another mimic).

But I agree with you that it gets those things correct most of the time and it also seems to be improving over time.

As a heavy user of Merlin, it definitely isn’t perfect - especially with all the mimic birds - to the point that there are complaints about beginners polluting the citizen science database with erroneous IDs from Merlin.

But it’s very very good.

Doesn't "average" = "middle class" by definition?

US median individual income in 2022 was $48k

https://en.wikipedia.org/wiki/Personal_income_in_the_United_...

> The “average person” doesn’t make enough money to pay rent or afford groceries

<1% of the US population is homeless, and ~10% receive food stamps. The average person makes enough money to pay rent and buy groceries.