Yeah, it’s good enough that it even had me fooled, despite all my “it just correlates words” comments. It was getting to the desired result, so I was starting to think that the framework around the agentic coding AIs was able to give it enough useful context to make the correlations useful, even if it wasn’t really thinking.

But it’s really just a bunch of duct tape slapped over cracks in a leaky tank they want to put more water in. While it’s impressive how far it has come, the fundamental issues will always be there because it’s still accurate to call LLMs massive text predictors.

The people who believe LLMs have achieved AGI are either just lying to try to prolong the bubble in the hopes of actually getting it to the singularity before it pops or are revealing their own lack of expertise because they either haven’t noticed the fundamental issues or think they are minor things that can be solved because any instance can be patched.

But a) they can only be patched by people who know the correction (so the patches won’t happen in the bleeding edge until humans solve the problem they wanted AI to solve), and b) it will require an infinite number of these patches even to just cover all permutations of everything we do know.

Here’s an example I ran into, since work wants us to use AI to produce work stuff, whatever, they get to deal with the result.

But I had asked it to add some debug code to verify that a process was working by saving the in memory result of that process to a file, so I could ensure the next step was even possible to do based on the output of the first step (because the second step was failing). Get the file output and it looks fine, other than missing some whitespace, but that’s ok.

And then while debugging, it says the issue is the data for step 1 isn’t being passed to the function the calls if all. Wait, how can this be, the file looks fine? Oh when it added the debug code, it added a new code path that just calls the step 1 code (properly). Which does work for verifying step 1 on its own but not for verifying the actual code path.

The code for this task is full of examples like that, almost as if it is intelligent but it’s using the genie model of being helpful where it tries to technically follow directions while subverting expectations anywhere it isn’t specified.

Thinking about my overall task, I’m not sure using AI has saved time. It produces code that looks more like final code, but adds a lot of subtle unexpected issues on the way.

An alternative that will avoid the user agent trick is to curl | cat, which just prints the result of the first command to the console. curl >> filename.sh will write it to a script file that you can review and then mark executable and run if you deem it safe, which is safer than doing a curl | cat followed by a curl | bash (because it’s still possible for the 2nd curl to return a different set of commands).

You can control the user agent with curl and spoof a browser’s user agent for one fetch, then a second fetch using the normal curl user agent and compare the results to detect malicious urls in an automated way.

A command line analyzer tool would be nice for people who aren’t as familiar with the commands (and to defeat obfuscation) and arguments, though I believe the problem is NP, so it won’t likely ever be completely foolproof. Though maybe it can be if it is run in a sandbox to see what it does instead of just analyzed.

I think those are where the name “desktop” comes from, though that term now refers to other computer things.

I refer to them as “tower”, “case” (which is technically just the shell and frame, but can include the contents), “computer”, or “machine”.

It might be sufficient if the case airflow is good. Not sure if you could avoid any heat throttling that way, but I’d guess it wouldn’t need to shut down because of heat.

This one is even worse than just removing the CPU cooler, because that cooler is now blocking the hot air from leaving the case via the rear fan.

No, they pulled the cooler off thr CPU and decided to use it to block airflow entirely to the CPU case fan. Best guess is that they are trying to build an expensive smart oven.

Back in the 00s, a story about CPUs getting so hot they’d start on fire went viral. In it was a video of someone removing the cooler while it was running and then a few seconds later a flame appears.

On the one hand, obviously you shouldn’t remove your CPU cooler while it was running.

But on the other hand, fans and mounts can fail, so this was still a risk even for people who were smarter than removing the cooler entirely.

It prompted CPU makers to add thermal protections that started out as “if CPU hits threshold, cut power”, but over time more sophisticated heat management was integrated with more sophisticated performance and power management.

So these days, if you aren’t sufficiently cooling your CPU, it won’t die much quicker, instead it will throttle performance to keep heat at safe levels. OP would have gotten better performance out of it after removing that plastic. Assuming it was CPU bottlenecked in the first place. Things like RAM choice and settings can make it a moot point because the RAM can’t keep up with the CPU at 100% power anyways.

That “we” isn’t global. Some called it “the CPU”, some called it “the hard drive”, some made fun of those two groups for not knowing what they were talking about.

I believe it was a product of the earlier conflict between copyright owners and AIs on the training side. The compromise was that they could train on copyright data but lose any copyright protections on the output of the AI.

On yeah, the little mouse puzzles. I always figured it wouldn’t be that hard to give cursor movement a more natural curve, just give it an interpolation that clamps the first 3 derivatives of position and adds jitter and a little overshoot and correction or clamps the derivatives even harder at the end to mimic slowing down for precision.

I’d say countdown to programs that pretend to be webcams and display an AI video of the requested action has started but I bet at least someone has already done it. And then the arms race between actions to be requested and what AI can do will start until eventually passing the test will be a fail because the actions requested are either too difficult for humans to understand or too difficult for humans to perform, at which point AIs will be trained on knowing the physical limitations of humans.

This will come in handy for when they get tired of our shit.

Or just do what I did where I have one of those wall mounted plastic channels with some of the cables hidden in it but two other cables just go to the TV without being hidden because the channel got full and I decided that I was done managing my cables for now.

The lady that owned this place before me had one of those in wall cable runs on a wall I didn’t want to put my TV on. Not even sure how you get the cable out the other side, so I’ve left it there with the broken HDMI cable she left there, in case I want to run a different cable (so I can just attach it to the current one and pull it through). I probably should just patch the wall up though lol.

Isn’t masking tape the one that looks kinda like sticky yellow/beige paper? I think you’re right about tape being used, but it’s one of the clear ones.

I also counted at least three of the plastic ones that get nailed to the wall. One near your first picture, (can’t see it while I write this comment (really need to get a better client than Voyager) so I forget if it’s visible in the pic or if you need to check the big pic), on the cable from the PS4 to the TV, just after the bend to a diagonal, it’s one of the ones you clip the cable into after mounting. And two on the xbox psu cable, they are both the type that you nail over the wire and a curved bit of plastic holds the wire against the wall.

But those were the only ones I could see, plus even those ones don’t explain how it keeps the hard angles instead of the cable settling into a rounded shape. Guessing they intended to use the plastic ones but quickly realized it wouldn’t hold the shape they wanted without needing to put a ton of holes in the wall and switched to tape after that to solve both problems.

Edit: looking at it again, I don’t think the first one I mentioned is one of those clips anymore, so just the two on the xbox psu cable (should be obvious once you see them).

Not gonna gatekeep posts, but this one seems more like a bug than asshole design IMO. Though “pay attention to me!”-style notifications existing for the app at all is assholey, so it’s probably a bug in their asshole design.

Yeah, not sure why yours in particular was the only one, probably timing.

This was reported but I don’t think it warrants censorship.

If you’re taking this more seriously than it should be, just keep in mind that there will likely be legal consequences since gas stations frequently have plenty of cameras and ask yourself if this hill is worth dying on. You’ll get charged and they’ll just replace the damn things anyways.

Yes. They just introduced a paradox to our logic, therefore everything is true.

But also no, because that contradiction also implies everything is false.

Yeah, Java’s enforcement of everything must be a class put me off of the language right from the start. I was already used to C++ at that point and hated that I couldn’t just write a quick little test function to check something, it needed a bunch of boilerplate to even get started.

I still think C++ has a great balance between object oriented and sequential programming. Not sure if it’s the best, but I can’t think of ways to improve on it, other than built in concurrency object stuff (like monitor classes that have built in locks that prevent more than one thread from accessing any of its functions at the same time, basically guaranteeing any code in it has mutual exclusion).

Sometimes tech presentations make me feel really bad for the person giving it. They are up there trying their best but clearly don’t have the skills to do more than just communicate information but still try to make their presentation cool and fun and it just falls flat.

Anyone can be cool, but not everyone can be cool on demand or on stage.

Though on the other hand, just because a presenter can pull off the cool factor, it doesn’t mean what they are presenting is actually cool. The coolness of a presentation has no correlation with the coolness of what is being presented, unless that coolness is just information about the product (though even then, they are probably skipping over the flaws and enshitification).