Forever watching people reinvent CSS classes for no reason. (I appreciate that this article actually does show doing it with classes, but it’s still grating to me that these “frameworks” just keep reinventing style="...")

This was written by ai, right?

Reduce is harder to read. Maps and filters are deeply simple. The problem space a reduce can solve is, I’m pretty sure, strictly larger than both of them. Its generality is its downfall.

I’ll happily use a reduce, but it has to be either as a well named function or immediately stored in a well named variable. If the reduce is complex enough that you can’t name it, that’s a smell to me, reach for a different tool.

That doesn’t change the fact that it’s harder to properly review code than to write code, especially when that code is ten times longer than it needs to be, makes no sensible decisions about how to separate concerns, and might, if you’re lucky, pay lip service to naming things well.

I don’t necessarily think this will always be the case with generated code, but I expect we’ll need a proper breakthrough in the technology before it can write good code (it needs a number of things: actual world models, the ability to go back and edit its output when it realises a mistake – which is crucially different from saying a new thing that contradicts the mistake – and continual learning, to name the three that spring to mind).

Even if it started writing perfect code, I’d be dubious of signing your org’s ability to write code away to a service that you don’t own and has a notable history of intellectual property infringement and an insane disregard for both the law and morality. At the very least we should all be using open weights models.

It is a very well known truism that it is harder to debug code than it is to write it. This is, in my experience, doubly true of the convoluted code generated by current frontier LLMs.

Allowing LLM coding at our org basically stole most of this year’s progress from us as every PR made this way still has yet to be merged because the code quality just never reaches anywhere near our minimum requirement.

We are investigating ways to improve this (a style guide for agents etc) but the best step we’ve taken so far is just to ask people to stop using it and see what happens. (Code quality jumped up and PRs started getting merged, though the LLM ones are still languishing and probably will need rewriting from scratch before we can merge them)

Yes, I’m well aware of this. There are plenty of locations where that’s not really a problem, though. And generally when they’re hotter and therefore less efficient it means there’s also relatively abundant sunlight, somewhat counteracting the effect of the heat. Cars also aren’t necessarily that hot, particularly if they aren’t powered by burning stuff. And putting all of that under the shade of a canopy would reduce the heat reaching the ground. If you used a brighter paving material like concrete, you could even benefit from bifacial panels using the reflected energy.

There are problems and there are solutions, maybe the problems outweigh the solutions but it’s still a worthwhile avenue of research to find that out.

Also a great idea. One of the things I think is good about putting it everywhere, though, is that it cuts transmission requirements+losses and increases resilience by having the generation be localised.

All fair. And, to be clear, I’m suggesting it be mandated or incentivised by governments, I’m not suggesting businesses would do it on its own merits. Though I wouldn’t be surprised if it ends up being a good idea for them in, say, twenty years time, even with all of the complications.

The USA has a stupid relationship with car parks, yes, but I’m not American and I’m not encouraging car parks at all, I just think we should put the space where they already exist and are actually useful to better use.

Ideally, yes, we’d have much better public transit infrastructure but it’d need to be every ten minutes on every route and not stop overnight or for Sundays or public holidays for it to be a viable replacement for a car for me. Which is very feasible in a big city but not so much out in the countryside.

Ultimately, some people will either always need personal cars (or perhaps some other solution, but no public transit I’ve ever seen will do it) for a huge variety of reasons, including disabilities and house locations (and I don’t mean suburbia, that’s generally solvable with public transit and also generally a bad idea).

Yeah, I’m largely spitballing. Perhaps the numbers don’t work out, perhaps they do.

That said, solar over parking is a source of income. It may well pay for itself whilst providing parking that is both shaded and rain sheltering and improving energy security and helping fight climate change and probably powering a bunch of charge points underneath it, which you could either charge for or just leave free to encourage people to come to whatever the parking is attached to (or just the parking itself if it’s a paid car park). My local Sainsbury’s has a free charging point and it’s a big part of why I shop there.

Also design the canopies to be their own scaffolding so the elevated maintenance is moot.

You can resurface under a canopy. Hell, petrol stations are almost always under a canopy and they definitely get resurfaced sometimes.

There’s a risk of someone crashing into any building, too, but we still build things that are useful beside roads.

I’m just saying that, yes, it’s not cut and dry, but I’m pretty sure the problems with the idea are generally solvable.

It’s probably a good idea to put solar panels on car parks where we’re going to have car parks anyway, though. In addition to agrivoltaics and using, as you say, substandard land for large scale solar. Also put it on roofs. Basically anywhere it doesn’t do any harm, I say.

I wish journalists would include stuff like this (because it is sadly often left out of research, particularly when people have an agenda).

Does this study account for the obvious: poor people eat more UPFs and are more depressed?

IIRC, third cousin is pretty close, n times removed is just generation gap. I mean it makes little difference, genetically, but being that close to a royal within written history is probably a good indication of generational wealth.

People who believe (because they’ve been lead to believe) that driverless cars are safe still have a right to live without injury. Being credulous or fooled doesn’t remove your rights.

I assume the exponential doesn’t hold up for some reason, maybe because of the amount of wheels they have. Otherwise it would seem that an LGV (maximum weight to cross borders in the EU is 44tonnes), would deal 16^44 times as much damage as a normal car, which is a large enough ratio that I think it would be safe to say that either the road would be immediately obliterated under any vehicle of that weight or normal cars do so little damage that it would be misleading to represent it as anything other than zero.

To be clear: I’m not trying to say you’re wrong. There’s another option for the growth rate, given the two data points given, that probably makes more sense. I just found it amusing to entertain the thought and go through the process.

Given how much heavier freight vehicles are than any passenger car, what percentage of road damage do they cause, given the damage seems to grow at O(n^16) ?

IIRC, their existing stuff is a white labelled mullvad, which is based in Sweden and, as far as I know, generally considered a good or even the best VPN for privacy.

Far Away by Jose Gonzalez.

Could well be that that was the effective treatment.