tomtheelder

100%, I thought about writing that out explicitly. I really feel like we are extremely close to reaching that kind of breaking point for most folks LLM use cases.

Fable is probably at least as good as the average software engineer and costs $50/wk on the max plan vs a software engineer who would cost closer to $4000 a week.

That's because the max plans are _massively_ subsidized. At API pricing the kind of usage to replace the value of a SWE is going to be way, WAY more than $50/wk. Orders of magnitude more. And to remain a frontier model org that kind of pricing has to continue in perpetuity.

I think the question is even a bit more nuanced than that. Even if frontier models can maintain a big gap that gap has to actually _matter_. If a local model satisfies my everyday use cases adequately then I may not really care that a frontier model is 5, 10, 100x better at ultra high order reasoning tasks.

I think that reality is probably not all that far off for a huge swath of use cases.

Zoom out and you'll see the hidden message

You know sorta feels to me like some new physical tech would be warranted here. Something with a good amount of storage, very quick I/O, and read/writable. More like a fancy thumb drive than a disk. Let the version on it be updated by the system to stay current, gain add on content, etc.

Xbox makes an enormous amount of money from a few long running properties (the letter in question mentions Mojang and King), but basically every big move and acquisition they have made for the past 10 years has been catastrophic.

I don't really have an opinion on Sharma since she's very green, but Phil Spencer's tenure was an unmitigated disaster. The strategy of rolling up studios and trying to grow through GamePass both completely backfired, with the studios producing few and generally unsuccessful games, and GamePass failing to convert casual players while demolishing your take from your hardcore players.

Was definitely time for major change at XBOX. Is this the right direction? I have no idea.

Well it was Phil Spencer, who got fired. They basically cleaned house in management already.

Gleam and Rust are really not alike at all aside from the most superficial ways. A couple bits of syntax, and the use of toml are about all I can come up with.

I'm an American and it definitely seems like we are in a significant and worsening loneliness crisis. I have no idea to what degree any of it is unique to Americans. Social connectedness, socialization rates, and companionship have all been declining for quite a while now. Lot's of potential causes and theories about it. [1] is a decent overview.

Like personally I'm doing great, and so are a lot of people I know, and I'm sure you as well. But I think a lot of Americans are struggling badly with their social lives.

[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC9811250/

The Northeast doesn't have a dry season, and I don't think anyone seriously thinks it's going to develop one. It just has occasional dry periods because precipitation is pretty chaotic, and is getting more chaotic due to climate change. When one of those happens there's some fire risk, like has just happened. "The Northeast is becoming fire country" is just unabashed scare mongering.

Local grocery stores have basically been extinct for decades- and realistically that isn't that surprising given they mostly sell commodity products. We don't need to let restaurants suffer the same fate.

The dependency graph is no different for a monorepo vs a polyrepo. It's just a question of how those dependencies get resolved.

I mean it works for Google. Not saying that's a reason to go monorepo, but it at least suggests that it can work for a very large org with very diverse software.

I really don't see why anything you describe would be an issue at all for a monorepo.

I still have a cast iron skillet, but I mostly stopped using it once I got some carbon steel pans. In my experience they beat cast iron in nearly every way. I only use my cast iron now if I need a huge amount of thermal capacity (like pre-heating it to make pizza on or something) or for the presentation value.

I like OCaml a lot as a language, but the tooling is very, very poor by modern standards. Poor enough that I think it’s a total blocker on wider adoption.

For sure- it's just that YC didn't used to operate like that. They have morphed from an interesting higher-touch incubator whose involvement was a strong positive signal into a scattershot VC, but not everyone realizes that so being "YC backed" still carries more prestige than is warranted.

JVM is a hard dealbreaker for a scripting/glue language.

Of course, but I think it's worth seriously evaluating the subset of those activities that have an especially large propensity for harm.

JIRA is a little overbuilt and overconfigurable, and I think that leads to it often being used in confusing ways. I don’t think it’s a bad product or anything, but my experiences on teams that use something simpler like Linear because the tool kind of guides you toward sane behaviors instead of letting you go nuts.

Kind of a convention over configuration thing.

I mean starting a component library by building your own from the ground up is an absurd thing to even attempt.

IMO the only really sane approach is to start with a facade around an existing library. This way you get your single interface and your consistent visual style, but you are free to swap out implementations as necessary.

I personally can't really imagine anything less inspiring than repeating a feat we managed over 50 years ago.

Honestly I used to feel this way, but ended up using Helix for a few months mostly as an experiment. Now the selection-action version feels just as natural as action-selection used to. I’m sure I could flip it back if I wanted to.

I’m pretty sure it’s not a natural wiring of your brain, but rather long ingrained familiarity.

That said, incredibly marginal value to switching so I wouldn’t bother.

My biggest tip would just be to make it abundantly clear in your interview that this is a goal of yours. Some jobs will bounce you for that, but that’s good- you want to filter those out. Others will see it as a positive and you’ll come in with your manager and leadership already understanding the direction you want to grow in.

Honestly this is such an interesting question. Conventional wisdom would definitely say C#, but I’ve always wondered if that’s because imperative programming is easier than functional for a beginner, or because basically everyone starts with imperative. I’d be curious to see what would happen if someone started functional first.

All that said, probably C#.

Exercise is a fantastic idea, and the cardioprotective benefits are doubly important for those of us with hypertension, but it’s probably not going to lower your BP by more than a few points.

True, but I would have to categorize that as an unacceptable workaround haha

It’s actually so painful to go back to languages without destructuring and pattern matching.

Yeah you nailed the limitation. Switch type expression that returns a value is a pretty universal feature in expression based languages, often in the form of a pattern matching based expression.

Check out the ‘case’ statement in elixir for an example.

In languages that support it, it usually becomes an incredibly commonly used expression because it’s just so applicable and practical.

Depends on the language a bit, but a common feature in these languages is the tuple. Using a tuple you would end up with something like:

let (value1, value2) = if (condition) { (8, 16) } else { (16, 256) }

Or else you’d just use some other sort of compound value like a struct or something. Tuple is just convenient for doing it on the fly.