When I buy something from Amazon (.de), during checkout there is often a banner asking if I don't want to not fail to join Prime. It changes from time to time, but it is *always* deliberately confusing. Usually, you are safe to ignore it, but I am pretty sure there was at least one time when I had to specifically uncheck an option to avoid joining.

It's a huge gray zone, of course, but actually tricking customers is illegal. It's time and past time that executives were not allowed to hide behind their corporation. Do something illegal, be personally liable. Let's have more of that, please.

See my comment below. I give three very different examples

Go for it. The water is warm. My last Windows installation disappeared about two years ago. Linux has its own quirks, of course, but it is at least as good as Windows. And doesn't put ads in your start menu.

Mint with Cinammon is an easy start. Personally, I like Xubuntu.

LLMs are incredibly useful, if you use them correctly. Let me give three very different examples:

1. I am fluent in German, but I learned it as an adult. As a result, I will always make silly mistakes with gender and grammar. When I need to write something that is absolutely correct, I can ask ChatGPT to "correct" a text that I have written. It does an extraordinary job of changing as little as possible (i.e., preserving the way I write) while fixing mistakes. I can also ask it to improve a text, and tell it how I want the text changed. Then it alters more, generally in the direction I want. Of course, the result still needs checked - it makes mistakes of its own - but this is an example where it is an incredibly valuable tool. A nice side effect: doing this is resulting in the first real improvements in my German in years.

2. I gave ChatGPT a list of four book series that I had enjoyed, and asked it for recommendations. FWIW I have a strange mix of tastes, so this isn't totally trivial. Anyway, it gave me a list of 10 or so series. Proof that it was probably on target? Three of those were other series that I had really enjoyed. I bought the first series, and it's exactly right.

3. A few times a year, I have to work with a rather complicated API. When I need something, I can either troll through the documentation, look in a book that I have, or hope that StackOverflow has a useful answer (which is increasingly rare). Or, now, I can ask ChatGPT. It provides an answer, together with example code and usually a pretty good explanation.

Now, sure, you can induce hallucinations. You can ask stupid questions. You can misuse any tool. Did you know that I could stomp on my keyboard and produce nonsense? Maybe you could gain some TikTok cred by making a video of it. LLMs are useful, and I wish the tech bros would stop hating on them and just use them for what they are good at.

p.s. I said "ChatGPT" above, but I actually experiment with a mix of LLMs. ChatGPT itself does seem better on the first example, possibly because it has more training material in the language. For the others, open-source LLMs like Mixtral seem to be just as good.

Lots of these guidelines are absolutely counterproductive to what many (most?) of us would like to see in the Internet.

I used to be admin for a couple of small, independent websites. Among other things, they had some decent, independent reference material online. Freshness? The material rarely changed, because it didn't need to change. Embeddings? No, we didn't artificially make lots of internal links to the reference pages. Who has time for that?

These sites were semi-important in their niche market. Back in the 2000s, the sites ranked in place 1 or 2 on Google. Over the course of the 2010s, rankings started to slip. First came commercial interests: shops that paid for SEO. Ok, that's not so bad, our sites were still always on the first page. As time went on, more and more crap sites ranked better. Sites containing semi-literate content, lots of ads, and links to other crap sites. Honestly, sites like that shouldn't be hard to detect, but I suppose they drive ad revenue, so...

I can see it now: swarms of greedy lawyers filing class action lawsuits. The goal will not be any sort of meaningful reimbursement of individuals, or even a legal win. Instead, the goal will be to extort settlements out of companies, with those settlements going to the lawyers.

The interesting technical question is: To what extent have companies reduced or eliminated their usage of PFAS since the initial concerns were raised about 20 years ago? Usage before then are not really relevant, because we didn't know that there was a problem. Continued usage after 2006 or so is more concerning. The lawsuits, of course, will fail to make the distinction, instead relying on tearful testimony by paid actors^H^H^H^H^H^H witnesses.

Depends. We use the cargo space a lot. It's nice that it is out if the weather and lockable.

Anyone with a dog ought to have it in a crate - SUVs are ideal for that as well. If you do anything other than pure passenger transport, they are very practical.

That said, I hate the fact that they keep getting bigger. Our 2008 RAV/4 was a good size. The new ones are fricking huge. Why???

You're giving them a *lot* of benefit of the doubt. By saying "30% chance of above normal temperatures" they implicitly say "70% chance of not above normal". Given annual variations, it is unusual to have precisely normal temperatures - there is always a near 50% change of being above normal, and a corresponding near 50% change of being below normal. So large regions are probably not looking at any sort of problem - but the map sure doesn't tell you that.

Since TFA fails to link to any actual source, one has to take it at face value: deliberately misleading click-bait.

So we have this map, covered with orange and red, which people associate with heat. What does light orange mean? A 33% chance of a warmer-than-normal summer. Meaning a 67% chance that summer will *not* be warmer than normal. White on the map is labeled as "equal chance". How does that differ from dark orange, which is a 50% chance?

Seriously, the coloration on this map is deliberately misleading. It is designed - not to inform - but to induce panic and clicks. Maybe there is real information here, but this kind of presentation must make one skeptical...

This. I assumed from the grayscale that it was e-ink. If it's just monochrome LCD (or any other active tech), then it's a nothing-burger.

Which, unfortunately, would fit with the typical OTT Silicon Valley hype. "Wow, man, let's make a crippled tablet and call it innovative. Awesome!"

That's a weird statement. PowerShell is not a programming language. Sure, you can script in it, like you can write bash-scripts, but I wouldn't list bash as a programming language either.

I suppose I haven't written any sort of Office applications for a long time (15 years or so). I wasn't aware that JS had displaced VBA for things like Excel. Old Visual Basic (through version 6, pre .NET) and VBA were (are) nice languages for their purposes. I'm not sure that JS is any sort of improvement. I suppose the argument is that people already know JS, so it's easier to support just one language rather than two.