I've been driving roundabouts for decades, and think they're great - they really help with traffic flow. I've never found them confusing.
I had to drive this specific Kirkland roundabout the other day, and ended up missing my offramp and going in completely the wrong direction. It's the most confusing roundabout I've ever seen.
I'm waiting until my kids are out of the house (just a couple of years now) to repurchase the 3-volume set. The first purchase didn't survive my kids' childhood - which, yes, I think Watterson would have approved.
I'd be more likely to assume that it's a really great one, where folks like and trust each other enough that they can give each other crap like this and everyone understands and appreciates it, like good friends trash-talking each other on the basketball court. I've thought for years that the way you can tell acquaintances have become friends is if they start insulting each other.
Or, as you say, it could be a really horrible environment - but I don't think you can tell from one anecdote.
Good point. The pager attack on Hezbollah was risky because it involved physically changing the pagers enough to put explosives in them. Quite a lot easier just to ship devices with some subtly insecure code.
Consider yourself lucky. He's a beyond-sleazy, probably criminal online influencer, important in what's called the "manosphere". He's made a lot of money exploiting and abusing women and encouraging other people to do so. If the online clips of him bragging about how to manipulate women into doing pornography don't make you want to throw up, you're pretty far gone. Unfortunately, lots of folks are, in fact, pretty far gone.
If I were in charge of, say, the Mossad, I would have as a significant part of my budget purchasing every single bluetooth device on the market, and set a bunch of underemployed Israeli CS grads to work at finding these vulnerabilities, and then putting them into an easily deployed toolkit. You want an asset with access to, say, an Iranian government office, to be able to walk through the building with a phone and take control of as many machines as possible.
Now that I think about it, I think you have to assume that they probably DO do this...
Until we have robot police officers, there will always be a human in the loop. But right now, there's a whole category of "drunk and disorderly" / "breach of the peace" kinds of laws that are ~100% up to the discretion of a police officer to enforce. I won't say you CAN'T have it any other way, because I can imagine alternatives like "don't have those laws", but I will say that you don't WANT to have it any other way.
> The camera can have different ways of seeing encoded in it, including kinds of gazes that enforce social agreements about what kinds of behavior and people are considered “normal”
The phrase "kinds of gazes" strikes me as the sort of thing that's only going to make sense to people trained in a very particular and idiosyncratic flavor of ethical critique. What a normal person sees here is, "These cameras can detect if people are acting bizarre and dangerous," which is probably something most people would appreciate. In Seattle, the problem, of course, is that the streets are full of people acting bizarre and dangerous, it doesn't take a camera network to find them, and the police seem to be under strict orders not to do anything about it.
I honestly think this is going to be a big part of what remains when AI is doing everything we currently think of as our work. Legally or morally, some things need a human in the loop.
Am I the only one who is mystified by this whole idea? People aren't CPU's. Good luck getting them to follow the code that you thought you were using to define their roles. On the contrary, what makes any complex system work is flexibility. And yes, if that calls into question the whole regulatory regime some companies (believe they) live under ... well, yes.
That does raise the question of what the value is of a "skill" vs a "command". Claude Code supports both, and it's not entirely clear to me when we should use one vs the other - especially if skills work best as, well, commands.
That doesn't work very well if your developers are on Windows (and most are). Uneven Git support for symbolic links across platforms is going to end up causing more problems than it solves.
It's all about managing context. The bitter lesson applies over the long haul - and yes, over the long haul, as context windows get larger or go away entirely with different architectures, this sort of thing won't be needed. But we've defined enough skills in the last month or two that if we were to put them all in CLAUDE.md, we wouldn't have any context left for coding. I can only imagine that this will be a temporary standard, but given the current state of the art, it's a helpful one.
I'm one of those really odd beasts that feels some sort of loyalty to Microsoft, so I started out on Copilot and was very reluctant to try Claude Code. But as soon as I did, I figured out what the hype was about. It's just able to work over larger code bases and over longer time horizons than Copilot. The last time I tried Copilot, just to compare, I noticed that it would make some number of tool calls (not even involving tokens!) and then decide, "Nah, that's too many. We're just not going to do any work for a while." It was bizarre. And sometimes it would decide that a given bog-standard tool call (like read a file or something) needed to get my permission every. single. time. I couldn't do anything to convince it otherwise. I eventually gave up. And since then, we've built all our LLM support infrastructure around Claude Code, so it would be painful to go back to anything else.
This is an N of 1, of course, but I can relate to the other folks who've been expressing their frustration with the state of Claude over the last couple weeks. Maybe it's just that I have higher expectations, but... I dunno, it really seems like Claude Code is just a lot WORSE right now than it was a couple weeks ago. It has constant bugs in the app itself, I have to babysit it a lot tighter, and it just seems ... dumber somehow. For instance, at the moment, it's literally trying to tell me, "No, it's fine that we've got 500 failing tests on our feature branch, because those same tests are passing in development."