substation13

I am finding the combination of LLMs and strongly-typed functional programming to be incredibly potent.

By building programs out of small, single-purpose pure functions, it is much easier to review the output of the LLM, whilst also giving me the ability to grok and steer the overall direction of the code. A traditional type-checker catches many of the looks-right-but-is-subtly-wrong hallucinations too.

Yep. And then they wonder why the government can't deliver anything.

(Government Digital Service being an exception, great work to those working there and reading this!)

There comes a time in every company's evolution when they switch from supporting openness (an opportunity) to protecting the walled garden from openness (now a threat)

> preforming a search is basically RO

You don't know this. Google results are not the same for all users. How do you know there isn't R/W going on, particularly when signed-in to Google?

(Unless you work at Google on search, in which case I stand corrected!)

I get it, but I also wish they could make one app that gets all of this stuff right. They could start by improving on the data model. "Unknown Artist" would never pass code review!

I think you are nitpicking. My point is that companies (including Microsoft!) are capable of running large scale infra with much higher uptimes than GitHub. They want to put themselves at the center of our workflows (e.g. GitHub Actions) yet they are not delivering uptimes that are commensurate with that. What is their excuse?

How do you know that?

Google search is basically never down.

AWS is basically never down.

WhatsApp is basically never down.

Time for GitHub to grow up?

You will need to maintain MS Build files alongside the Bazel definitions. In theory the MS Build files can be generated from the Bazel definitions but someone has to write that tool.

The article explains why Cargo was insufficient for them.

You may be right, but this:

> Cities could convert these office spaces into homeless shelters and save money.

does not follow from this:

> Unhoused homeless people actually have a non-zero cost to the cities they’re in, whether it comes to incarceration or other costs.

I think there is a feedback loop with self-driving cars.

More self-driving cars on the road -> more driving data for that company -> better self driving cars

The book was way over optimistic but many of his predictions are coming true, a few decades late.

I might have to revisit it!

This is the big question isn't it. With self driving cars we have been on the final 20% for what seems like forever.

I think replacing a SWE with AI means we have reached AGI levels.

However, GPT 4, which is amazing but not able to do what a SWE does, is already making concept artists, voice actors and content marketers obsolete. The arts are far more susceptible to this technology.

Anyone who thinks SWE is generating Python snippets doesn't really understand the role.

It's a fun action novel; I loved the part with the cult on the island. But yeah, it has nothing on Player of Games.

OCaml is having a bit of a moment

Only had 10 mins so far, but I think it's useless.

It generated some (incorrect) code for me. Just ask!

If it is simply pricing power then... why now?

If companies could raise prices 20% all along, why did they wait till now to do so?

Why not in 2019?

Trouble is, wild areas are being destroyed to make room for suburbia.

There's still time.

Rate hikes may be cancelled if there is a debt crisis.

This is why cities are better for the environment. I don't think rural car usage has nearly as much negative impact as driving in a big city though, so this shouldn't be a strong policy focus. Awesome that the kids can ride to school!

I think GPT is more useful to people writing self-help books than programmers.

Did you ask a skilled pen-tester?

Containers can be broken out of. They are not a security tool.

However, from a configuration perspective I totally agree. Python ecosystem is broken beyond repair. Just chuck it all in Docker and get back to programming!

I mean low EV compared to other careers right now. Obviously automation benefits the ownership class massively. Buy tech stocks; they are discounted at the moment.

Low EV. Some make it very big, but most earn nothing and retrain.

That's not really how it works.

At least in the UK (and most western countries are similar), the government decides (with ministers) what the curriculum should be and how it will be assessed. They decided that rote learning is what students should do. The schools have no funding for anything innovative - again, a decision by the government on how much to allocate. They can barely afford text-books, let along support an edu-tech start-up ecosystem. VCs won't touch edu-tech with a barge pole. Meanwhile, the government assessors ensure that things are taught in a particular way. Again, decided by the government and the bureaucrats. The teachers have zero control over this.

Now universities should know better. They have more funding and more resources. But there are some leaders here, like MIT.