In theory roundabouts resolve this sort of issue because they give priority based on where a car is rather than who arrived first, so the bike would have been in the wrong if it pulled out in the way of a vehicle already on the roundabout and the vehicle would be at fault if it hit a cyclist already on the roundabout. The tradeoff is that roundabouts can end up with big tailbacks in some directions if flow of traffic in a different direction is constant as if traffic flowing N to S is constant then traffic travelling E to W will not be able to go until other traffic interupts the N to S flow.

Checkout Hemel's 'magic roundabout' if you want to see a set of closely located mini-roundabouts that see very heavy traffic. https://en.wikipedia.org/wiki/...

We don't have enough detail from a reliable enough narrator to make any blanket statements like this about most cases. It's possible the cyclist did something illegal but there certainly isn't sufficient information here to determine that with confidence. Regardless, someone doing something illegal does not mean they are at fault if the other party proceeds to make an avoidable mistake. You can't just run a pedestrian over if they start crossing on a red light and you could easily avoid it for example. Again we don't have nearly enough info in this case to know if that is the case here.

The main issue with coverage of self-driving vehicles is the human logic shortcoming that can't handle big numbers and ratios. We see a single story about a cyclist who had an accident and rode off with what seems likel superficial, if any, injury and can't contextualise that around 300 cyclists a year die in collisions in the US each year. Sure there's a responsibility/accountability question but we should expect some accidents where self-driving cars make errors because it isn't realistic or helpful to expect them to be perfect when what they would replace is as crap as the average human driver.

I've always found it bizarre that even in the EU/UK where safety testing includes pedestrian safety there is basically no consequence to buying a car that is dangerous for pedestrians. If you took a car with a really high safety rating and put spikes on the front then hit someone even by accident you'd be prosecuted for reckless endangerment or some such but buy a car that just comes that dangerous to begin with and you're fine!

Look, I dislike the things, I don't own one...but I don't see that the govt should be in the business of guiding public behavior.

Oh bollocks you don't. You think if you gave the people of your country a free vote on whether to give themselves $1,000,000 each tomorrow they wouldn't vote overwhelming for it and how it would be paid for be damned? You think if the US gave them a free vote on whether to bomb the shit out of Iran the day after 3 US soldiers were killed they wouldn't do the same? The whole and only point of having representative democracies over direct democracy is to moderate the public behaviour.

And you need to keep in mind that your logic actually encourages what is happening here (contrary to the post you are responding to). Parisians support anti-SUV measures therefore by your logic it is right that there are anti-SUV measures in Paris to do otherwise would be politicians interfering with the publics will.

There are plenty of alternatives to Facebook. Look at the proportion of under 30s that either don't have FB accounts or who never/virtually never use it and that becomes obvious. I'd argue there's less competition for YouTube because a lot of the content on YouTube isn't anywhere else and unlike Social media where people often use more than one platform there are very few people who regularly use more than one video service like YouTube.

There's plenty of evidence to the contrary with major YouTube content creators sharing figures about earnings vs alternatives; most recently MrBeast when discussing relative earnings from X. YouTube seems to have gotten much worse imo over the years but if you want a lot of content it's pretty much the only place to find it currently, and this means it's where everyone looks for content.

Awefully snarky for someone who can't manage to comprehend that something being the least important thing about a language doesn't make it irrelevant. If Python only ran on 8 bit processors, was 10,000x less efficient, had to be programmed in a specific godaweful IDE, and had restrictive licensing controlled by a known bad actor, then no one would be using it and how it interprets whitespace wouldn't be discussed because it's trivial by comparison.

I've worked in companies that have categorically saved money from moving all, or virtually all services, to cloud providers. A lot of the points you make about not optomising as much in practice on cloud infrastructure are equally true for on-premise. I've literally sold off hardware that was less than a year old to buy new hardware because the original equipment was so poorly optimised for the type of licensing required; almost every time I go into a business they either have shitloads of excess on-premise hardware with all the related costs or don't have sufficient to offer decent resilience and haven't done something about it because of the upfront cost.

Which is absolutely not to say cloud is better, and certainly not better in all cases, but I'd have to see a larger meta-analysis to really buy into any claims that staying on-premise is cheaper for most companies.

To be clear, there is no evidence that legislation is the problem in this case anyway. The vendor is trying to claim regulation justifies what they are doing but that doesn't mean it's true.

There are plenty of experiments that show direct observation isn't entirely reliable if you really want to vanish up a philosophical debate but I'm inclined to think most people want to discuss the story.

I never ceased to be amazed by how astoundingly ignorant or naive some anti-regulation posts manage to be. Anyone who really think a world without contracts, and no consequences for defamation or misrepresentation is not going to be a shitshow needs their head examined.

No, because copyright may be relevant here (or not) but contract law would allow this kind of restriction on use anyway. I can sell you something with a contract specifying you must pay me £1 billion if you get the train serviced by someone else, and it would be enforced by law, regardless of copyright on the code itself unless you regulate what can be enforced in a contract. The lack of copyright law doesn't stop me from including terms in a contract that are equivalent to copyright law either unless you regulate what can be included in contracts.

In short: if you allow contracts then regulation can only reduce what could be done because without regulation anything could be in a contract.

You seem to not understand what regulation is; and it looks like confusing it with contract law. Less regulation doesn't mean no contract law, and regulation restricts what can be enforced by a contract because by default anything could be in a contract. Thus less regulation would mean the manufacturer is free to require what they like from a buyer making enforcing things like this even easier.

The clear sign you are missing the point here is that the summary, and article, don't claim that the trains were disabled due to any regulation based requirement; they were disabled because the vendor didn't want the trains servicing by someone else.

There are valid arguments against some regulation, and over regulation generally, but you aren't helping make them when you try and spin unrelated stories that way; especially when regulation to stop exactly this issue has been used to stop similar issues.

I think you've made a mistaken assumption that they were arguing for copyright when it seems pretty clear to me they were highlighting that what happened here is a consequence of the copyright system, that it is bad, and are themselves suggesting people choose software with permissive licenses as a solution.

I'm not a fan of copyright but I also think it's easy to claim the idea is entirely problematic when there isn't really a way to know when can't tell what things would be like without it. I'll focus on games here as it's the point of this story and suggest that given it is easy to distribute software without restrictive licensing if you want to then if that was an effective way to develop and distribute world class games then surely many of the most popular or highly rated games played would have their source distributed with permissive licenses? I'd also suggest that consumers, regardless of whether they should or not, simply don't value whether they can run a game 5+ years after they buy it when making decisions on what to play or buy.

I'd also love to see copyright protection require those claiming it to provide their product in a usable format to some form of central archive for retention etc. The main obstacle I can see to this is that it would either restrict how developers could operate or is impractical with some practices. Can you imagine how confusing/hard/implausible it would be to 'archive' WoW with it's constant codes changes to server and client side in a way that would allow someone other than Blizzard to just pick it up and immediately operate it.

Sadly you see similar harmful, pull up the ladder, style systems in lots of places. In the UK there are 3-4 different rules that make owning larger property than you need attractive, for example housing is exempt from inheritance tax so if your well off you don't want to downsize because you would pay tax on the money released.