XorNot

"the West" geographically yes but the Australian public is hoovering up EVs right now.

Whoever hits the right window on an electric ute is going to make bank (it's probably more about getting another battery price reduction at this point so you can throw energy at the aerodynamics and range issues).

True, but then you're into the super big picture questions that The Culture books were generally asking.

Programming has arguably never been the most important part of making video games though, except when people stuff it up and a game doesn't run well on the hardware people want to play it on.

Story, design, game mechanics...these all have something in common, and it's that at a moderate sized studio they're not even slightly the programmers job.

Arguably if you wanted to make games then now is the best time to do it because a lot of the boilerplate won't be in your way.

I really don't. I've realized I never liked coding. I liked designing and watching manual tasks become autonomous.

So AI software development is basically giving me everything I want: see a problem, iterate on the design I think will fix it, try it out. And then it works.

I have personal Android apps which "I developed" by having Claude update the react native hot reload view as I used them for the main application, and the biggest change in my home server infrastructure has been retooling it to support the fact that almost all the apps I use going forwards will be built and hosted locally.

I mean you can be bored all you want, but the technology works. That's the difference. This isn't cryptocurrency where "you gotta get in to make it big (because I need to sell it to you)". It works: the results are real, the code runs, you can read it - the systems it provides are real.

Even the security argument has evaporated: AIs now hack better then people. They're better bug-finders then developers. The tool which exploits your app itself would be another AI system. On a good day you might be smarter then it...but you're not faster, and you don't have the perseverance.

My challenge would simply be: buy a low-tier Claude license. Or some other system - I use Claude so that's the example. Point it at something you wrote. And realize in the span of about an hour how obsolete the human developer writing code truly is.

Why are bioweapons the only thing tech bros worry about, when some bags of nails, fertilizer and diesel in a sedan will defeat any security measure they care to build for themselves quite reliably?

All I know is that I currently pay Anthropic a bunch of money per month for Claude and that Google wasn't even on my radar for that.

Seeing a physics result in there has got me hyped.

Turn this loose on LHC data ASAP.

The trouble is that's doomsday anyway. When I don't use most of the features of your product, you're competing with the monthly cost of my Claude subscription for value - and that's very good at building me exactly what I want.

Code review is the problem I'm currently trying to figure out for myself. I've never liked GitHub PRs but I'm not 100% sure what I want instead (which is possibly just: it feels like it should be a VS Code plugin... But maybe I don't like VS Codes interface that much either?)

That is...not at all my experience of Opus 5.5, which has absolutely both spotted weird workarounds, then spotted the underlying bugs which were leading to the workaround, then spotted the race conditions which were masked by the workarounds.

This is outdated news though. It's cope. People are out there telling themselves "well it's insecure". That was true. The AI models are better hackers then humans now though. Red teaming your own code with AI will produce a better penetration test then any developer trying to implement OIDC.

And frankly...have you tried configuring Keycloak? It is a gigantic "do everything app". The idea that it's going to have a narrower attack surface is basically fantasy.

You are seriously underestimating how good something like Opus 5.5 is at testing.

That's what truly convinced me programming was over: it writes better tests, more thoroughly, then any developer.

It also has to do so without destroying a bunch of other exceptionally well tested theories - it's very easy to explain something if you ignore other requirements, like being compatible with relativity - which has been a big problem for MOND but is no problem at all for dark matter (because it's just extra matter being normal out in space).

No they won't..like, observably they won't. This is a problem almost every entrepreneur who looked at small businesses as a start up opportunity ran into immediately: most small businesses are run by people who are exactly happy with where they got to, and so showing up and promising you'll expand the business goes over as well as a lead balloon.

Conversely for me AI is letting me focus on the part I actually enjoyed: organizing and planning systems, and powers through the parts which were just a grind to get there.

By far my biggest use at the moment is just getting all the software I use to function as I want it too.

As opposed to just exploding from the high pressure tanks of gas onboard due to an otherwise survivable fender bender?

Okay but he also hasn't cured cancer, hasn't delivered full self driving, and this car is not useful as a car in anyway.

Any sign of the tens of thousands of Optimus robots?

It'll be fine. You voted for strong labor protections, universal healthcare and a welfare system right?

Seriously I don't get what everyone else is doing where they say things are getting worse somehow.

There's days I no longer need to sit at my computer: Claude Remote prompts my phone when it needs direction, the hardware stays busy and I can just go out with my family instead.

How timely, I just now pointed Claude at my work laptops MIPI driver and told it to figure it out.

I said "in an hour" to mean, the time it takes to identify the issue.

That fairly obviously doesn't turn into "bugs per hour" to verify, but if you find 10 candidates in an hour and then get reproducers of them out over the next few days, they are real bugs.

Like sure, "not properly reviewing" could be a thing but if I'm looking at a hard crash on my console, that's not behavior as designed.

Okay for feature work but the amount of bugs I can find with Claude in hour, with deterministic reproducers is staggering.

Claude's code review is a lot less interesting then getting Claude to reproduce the bugs it claims to find in code review, which has had an absurdly high hit rate for me.

The biggest problem I see with how a bunch of people use these tools is they go to them as an oracle, rather then letting them be plugged into and interactive with problem.

And it's in that later context that Claude is amazing: it can run tests and setup scenarios which would take days or get stuck in some weird problem loop. And then you can just say "okay, walk me through this problem" and see it yourself right there.

I won't call it common sense, but humans with highly limited technology lived and reproduced to build sophisticated civilizations in almost every biome on the planet.

Our biology is immensely adaptable, so anyone declaring you absolutely need some finely decided very specific thing in order to be "be healthy" should be treated skeptically.

With regards to protein specifically: your body doesn't trade in proteins as inputs, it trades in amino acids which cracked off the protein chains you eat. Everything else is just trying to keep the crowd of gut bacteria from making you feel uncomfortable.

One habit I've really noticed I have to break myself of with Claude is the temptation to say "maybe later" when I have an idea. There's never a reason to do that: if something might work, I can leave Claude in a tmux window to check the details I think should exist and then just...go do something else.

I mean thats not why I have Claude decompile things. I have it so it because my software should work exactly how I want it to.

I just do a lot of DIY construction. Being able to lift even 100kg safely would be a game changer.

The trades would benefit enormously from such a thing.

As a former AI skeptic I get it and still get it.

But now other people's code is just a starting point for me forking an app to do exactly what I want it to - it's inspiration because code is cheap.

Of course for larger projects we will still need solid components underlying them, but what you're really talking about is just maturity - it's very easy to produce a lot of output now but you can't replace actual history or having more then exactly 1 maintainer.

It's because tmux makes session resumption easy. Although for K8S it makes no real sense, for Claude it does.

Of course with all the running around with AI programming, it would be nice if we could task sometime to making remote desktop not suck on Linux.