osigurdson

Incredible that SO sold for $1.8B at one point.

>> If you have suggestions on how to name our products better, I'm all ears

<version>-<size>

Where <version>: 6.0, 6.1, etc., and <size>: xs | sm | md | lg

E.g. 6.1 Terra = 6.1-sm

If you eventually make specializations like "math" and "jokes" then I would call it something like 6.1-jokes-sm.

>> Claude Code lives in its own repository, separate from the monorepo

I thought mono meant one.

What are some use cases (if any) that the language is not suitable for?

Even in the M7, I doubt AI is having much impact other than benefiting from AI spending. Its hard to point to any non-AI products in the M7 that have meaningfully improved since 2023.

The problem isn't that "AI doesn't understand..." this or that, as concluded by the article though. Its more a lack of creativity and incorrect reasoning about what AI can and cannot do. For one, whole role replacement is pure fiction and likely will never happen with LLMs. I expect most businesses will look very different in the next 10 years due to AI and it will not be smooth as the needed creativity will mostly happen in startups (and perhaps companies that can truly operate like startups) and diffuse out from there. I'd expect a lot of churn in the S&P493 and even the M7 most likely.

I think you've answered your own question. The models simply aren't that good yet. If they were, the prompt would be something like "Find the most lucrative product that we can build. Then build it and start selling it. No mistakes".

The might actually know real design principles vs what ever we made up in the 00s.

Although I was personally fairly invested in UML, OO, etc., in the 2010s, it is undeniably true that classes, inheritance and all that is incredibly out of fashion today - people just found it didn't really help. It's interesting that this is now the solution to AI coding.

When I first saw this I was convinced it was AI slop. Nope! This is absolutely incredible. I've also never really resonated with Tesla aesthetic in the past but they really nailed it here imo. This car looks phenomenal - I hope this style trickles down to other models.

You're thinking the whole AI boom is just a front for crypto mining?

Beyond a certain scale, humans cannot organize themselves in a completely altruistic manner, even in Europe. The best we can hope for is for gatekeepers to mostly wield their power for the good of most people - we have to accept some error rate. Making all such practices illegal would only serve to make things less visible with little actually changing. It might even lead to more corruption.

Answering my own question. Andrew says "we almost had it but ripped it out" ... here: https://youtu.be/zwi5b5xSsKA?t=2711.

I suspect they can't easily tell the difference between a high quality LLM based contribution and a low quality one.

Curious as to the state of evented IO / io_uring in this version.

I can't imagine many situations where you have many cpu bound go routines. Most people aren't doing SPMD / open mp / similar in Go or other GC type languages. The main point is efficiently waiting for IO with an understandable programming model.

Haha! Laughing at my own comment: "a goroutins shares memory". The iPhone keyboard wins again.

For sure. The parent comment was merely asking if it could be done without channels or mutex, not if it was a good idea or not. Of course sometimes it is indeed the right thing to do.

A goroutins shares memory so you can just share variables if you want. It is basically the same as any other language in this regard.

>> in terms of how things can end up happening in any thread

Doesn't that describe pretty much any green thread style concurrency implementation.

Agree. These days, if you think you have a methodology that works better than others, you can actually try it / compare it and publish it so that others can replicate and critique your work. Articles like this one, that merely claim they've found the secret sauce, therefore should not carry much weight.

That wasn't the case with 00s agile / Uncle Bob stuff since proving that any of it was helpful was impossible - you just had to believe (and if you didn't believe there was something wrong with you!).

A recent HN article (below) concluded that asking agents to do TDD wasn't particularly helpful. I hope there is more research on this because TDD will be slower, use more tokens and results in more code to review.

https://news.ycombinator.com/item?id=49605246

The easiest thing for people to do in that situation is just create another Bitcoin. The person that has the 500K original Bitcoin is free to trade with themselves. This is why Bitcoin is much more of a social network than an algorithm.

Bitcoin is more like an asset than a currency. Unless a country makes Bitcoin its only currency it doesn't really matter that it is deflationary.

Agree. Bitcoin is more of a social concept / weighted democracy than an algorithm. If the participants decide that a new / incompatible hashing algo is needed (e.g. post quantum), it will be forked and everyone will use that.

The odd part is, the governance model is not defined. At the end of the day, some person or group has to decide what the new "real" Bitcoin is and the average holder just has to go along with it.

Note that Linux often runs on Lenovos and Dells, some of which may be creaky. I guess you are saying you prefer macOS.

I think for most projects, add an ORM only after you start getting really annoyed and understand what you are trading.

I used .NET for many years with Postgres on a huge project. Honestly, I don't know what you mean. Can you break down "Everything" a little. What is it that expects MSSQL? Who are the thought leaders pushing it?

I'd say Microsoft's revenue related push with .NET these days is to try to subtly nudge you into Azure (where MSSQL is rare and Windows is virtually non-existent). But, I also think they know they will kill .NET if they go overboard with it as the competition is strong.

Obviously, don't use it if you don't want to and don't have to. It is not worse than Go or Java however which is its only meaningful competition.

It is an interesting fact about Bitcoin in general. There are 21M tokens in total AND some percentage are lost every year. Run this simulation long enough and there will be very few active Bitcoins remaining.

I like how the nurse gets to check if the AI is doing the right thing as one of her new tasks. AI is going to have to get to the point where it can do things on its own or it will only transform problem domains where verification is fast and inexpensive. Otherwise something is getting traded and people will notice, eventually.