You missed the point.

A friend of mine made a tea-timer app for the iPhone. Where you enter what tea you are making and it sets the correct number of minutes. Today, people would ask ChatGPT before they go on the App Store to search for something like that. That's the point. Not what YOU would do, but what the general public does. And for Jane Doe, ChatGPT has replaced a lot of these specialty apps.

Skills are great until they aren't. The problem is exactly what you say: You can literally create a skill by telling the LLM to create a skill. Which is... why the f do I need to make a skill if the AI already knows how to do it?

Properly created skills, that include reference material, specific instructions, and expert knowledge the LLM doesn't have or that's buried somewhere under 10 layers of reddit posts is genuinely useful. But most skills aren't. I'm constantly underwhelmed by hyped-up skills that, when I look at them, turn out to be bland, basic instructions.

The right way to do this is to hire a group of developers and have test volunteers report workflow problems - then FIX them - then iterate. That's the point of choosing an open source solution.

Eventually there would be no problems remaining.

But so many possible scenarios exist that it's impossible to guess what happened here.

PS _NSA_KEY

There are apparently other unpatched vulnerabilities that are known but 'too busy':

It sounds like this fix got done because it was 'accidentally' posted as a public PR.

https://github.com/tianocore/e...

Yeah, the threat actor would need a foothold on the PXE network, which is a private VLAN (or better) in the competent case.

That's why we worry about APT's and such.

Granted, you'll find people on certain cloud providers who do far too much on a public IP.

What if someone has fifty windows open with fifty tabs each?

Asking for a friend.

Solid effort - good job.

It's a bummer if any of the employees got rug-pulled but many of the best employees will strike out on their own, form new companies with working conditions they value, and likely hasten Skydance's demise.

nah, a RS code can correct a tiny bit of loss but two metadata copies can correct a bad sector or even track if you're lucky.

Different requirement spaces.

The idea this guy has might make embedded devices slightly more long-lasting.

They usually have quantity one of a cheap storage part and a bad block in the wrong place could brick the whole thing.

It would only be useful in the case where maybe the build is 12MB and the chip is 16.

Most vendors won't ever pay extra for this.

I suspect the people who hate it don't have problems it solves.

Maybe a very basic home laptop is fine with SysV today and they don't want to learn anything new.

Personally I've been able to retire thousands of lines of scripts I'd written to handle complex things with many edge cases and ugly devops modules that are no longer needed with more recent releases.

There's just no userland interface to many of the events it can respond to and it's not easy to write a new C++ daemon to handle each need that comes up.

At this point it's largely a functional API to kernel events in the way I use it. That's the DOTADIW perspective for me. I still have services call my scripts where it doesn't handle what I need. It would be crazy to ever expect it to handle everything.

But just because I need to handle complex things at scale sometimes doesn't mean everybody else does. If I want to run debian with X11 and emacs and you want to run Devuan with Wayland and vi, you'll get no argument from me.

Who has time for drama these days?

But again that's backwards. I let my AI coding assistant write tests exactly so that I have a reproducible, deterministic ground truth that I can inspect for logic errors. If you let the AI do the testing directly, it's a toss-up if today it feels like it or it decides that your desired result is "everything green" so it'll just give you that answer without actually testing anything.

LLMs obviously aren't human. But that's not the point. The point is that you don't need to know a programming language to program a computer.

For example: Especially in the early days of smartphones, there were mountains of small tools that did something that's essentially a simple script. Egg timers, remind-me-tomorrow apps, tons of stuff that's basically a wrapper around a web search, and so on. Almost all of those are today something ChatGPT can do, and are easy enough to describe.

I've actually outsourced the first iteration of that to AI. :-)

At the end of coding sessions, I generall ask it to spin up an adversial agent whose task it is to find flaws, inconsistencies, mistakes, bugs, etc.

I'd still strongly recommend to anyone shipping AI-generated code to have a thorough human code review, but for a first pass at surface level, the AI can check itself if it has some kind of separation between the generator and the reviewer.

Personally, after some experimenting, I've concluded that:

• AI is good at implementing known algorithms.
• AI is good in creating code that flows straightforward from a design document.
• AI is good at code reviews and finding bugs.
• AI (unlike me) doesn't mind writing a few hundred tests.
• But, it tends to over-engineering simple tasks to the point of "why does Hello World need enterprise-software levels of abstraction?"
• It struggles with code that doesn't have a known solution.
• It's difficult to get it into a different lane once it started going down the wrong way.

What I use it for these days is rapid prototyping. Using AI I can make a functioning prototype of an idea I have in a few days instead of a few weeks. I can quickly test out an idea for a feature or change in my existing codebase.

What I also have been using it for a while is as a code-reviewer and for someone like me who is mostly working solo, that's an insanely valuable thing. Not just the bugs themselves, but also the ability to discuss code with someone. Basically, AI is a rubber ducky that replies.

Mickeysoft lost the plot again.

The game-changing element of LLMs and agents is that they have a non-computer-language interface - you can talk to them in plain English (or Chinese or whatever). This makes complex computer interactions available to a much wider audience than any previous computing invention. Adapting humans to the AI is getting that backwards. It's a sign you don't understand the first thing about why AI is so widely successful.

I predict this will be an abject failure and a year from now will quietly be pushed to the very back of the bottom-most drawer, never to be spoken off again (though it is getting crowded in there for M$).

Meanwhile, in the real world, everyone I've talked to about these things already has a skills file to teach the AI about their personal coding style. This is what actual humans came up with, independently several times, and it works: Adapt the AI to you. Plus we've been writing coding style guides for humans for a long time now, we've got plenty of experience doing that at least somewhat right.