bsmithers

Really? I think we have exactly the same problem in the UK. Granted, my eyes are more susceptible post laser surgery, but headlights are definitely getting brighter and higher as the years go by.

> And yet there have been several recent studies that show the younger someone is, the more likely they are to be scammed online.

I think you are misreading the post. Pretty sure they meant

you could trick me into believing I was talking with Elon, but you could not trick me into believing I was talking with my son

To which I agree personally, though I don't know how universal this is.

One big difference is that there are lots of reviews for $CommonPosition. With this, you'd have to expect that most reviews are the only one for a given property/landlord

I find that storing the jars upside down works surprisingly well.

From the article:

> Unfortunately, if you’ve been observing the p-values, you’ve noticed that most have been very high, and therefore that test is not enough evidence that the tips/threats change the distribution

It doesn't look like these p values have been corrected for multiple hypothesis testing either. Overall, I would conclude that this is evidence that tipping does _not_ impact the distribution of lengths.

4.5L of 0.5% ABV beer would not make you drunk. For comparison, this is equivalent to less than a pint of a typical mainstream beer. Indeed a 0.5% ABV drink can be branded alcohol free (at least in the UK).

Would 4.5L of 2% ABV make you drunk? Depends how quickly you can drink it. It's 9 units of alcohol. We commonly assume the body can process 1 unit/hour so if this quantity was consumed over an entire day, I don't think you'd get merry.

I'm not saying it's a good idea to drink like this, just highlighting the impact of ABV on the outcome.

Love the game. Couple of nits on the site:

- changing puzzle size seems to break the back button - scroll behaviour is a bit strange

(Pixel 5, Chrome, if it helps)

Excellent article. Faceted visualisation is an incredibly powerful technique.

Something the author hints at but isn't quite explicit: manual inspection of individual examples from your dataset can help you understand what questions to ask, what category to facet on, or the bug in your aggregation.

Not to be pedantic, but that's a bit of a non-argument. _Of course_ you can do it with xargs and shell, but imho parallel is generally more convenient, especially for remote execution. It provides a higher level of abstraction for such tasks.

I don't get this at all. What happens if they reply to you - maybe hours later - when you are the one presenting?

I think it's on you - as the receiver of a message - to appropriately deal with your messaging app of choice such that it doesn't interupt your flow. Otherwise, you are placing an impossible burden on the sender of the message to some how telepathically know if it's a good time. IMHO, your preferred workflow makes it seem that your time is more valuable than the other person's as you're now asking them to hang around waiting until they can actually ask their question.

The asynchronous nature of messaging is important to preserve.

I assumed the intention is to alert a driver who really needs to take a break. I don't see what other use it has, other than to annoy me when overtaking a cyclist.

The refusal to lock was probably that you'd not fully switched off the engine. I've had this on my Qashqai. The fuel-saving system that automatically turns off the engine when you're stopped and out-of-gear can be confusing. This behaviour often happens when parking but it seems you still need to press the stop-start button to "completely" turn it off in order to lock the car.

This is the first car I've owned with "modern" driver aids and it is true that they are somewhat temperamental. For example, the auto full-beam headlights get very confused in fog (which can be dangerous!), whilst I often get a warning noise when approaching sharp bends at a perfectly safe speed.