randy

I feel like she's trying hard to distance herself from Trudeau, but after years being right beside him, I doubt she can shake it that easily.

Note the article is from two months ago, on October 10, but it's still relevant. I was confused when I saw Chrystia Freeland quoted.

You can only pardon someone who was convicted of a crime.

I don't believe that's true. See the pardon of Richard Nixon, who was never even impeached, let alone charged. I believe there's some debate on the validity of such a pardon, but none (including Nixon's pardon) have been challenged in court.

Excuse me, that was Finn's right arm.

I've noticed that, if an equation calls for a number squared, they usually really mean a number multiplied by its complex conjugate.

I’m sure plenty of pedestrians have been killed by cyclists.

I did some quick searching and found 2019 data from Europe. In all of the EU that year, bicycles killed 19 pedestrians while cars killed 3200 pedestrians. Over 168 pedestrians killed by a car for each killed by a bicycle. I know there are plenty of irresponsible cyclists, and yet they are still a tiny fraction as dangerous as a driver.

Yes, but not all clients expose dependent tasks (which is sadly a common issue with open standards: they aren't always properly implemented). I'm using Tasks.org on my phone (which supports dependent tasks), synchronizing to a Nextcloud server with the Tasks app (which supports dependent tasks now, ~~but didn't for a long time~~), which also syncs to Thunderbird (which does not appear to show dependent tasks as dependents).

Edit: remembered that the Nextcloud Tasks app has long supported dependent tasks. I was thinking of recurring tasks, which it does not support. Again, open standards aren't always fully implemented.

I think CalDAV (which uses the iCalendar format) may be the closest thing. It covers calendar items, obviously, but also task and journal items.

It sure feels like we're at the peak of the Gartner hype cycle. If so, the bubble will pop, and we'll end up with AI used where it actually works, not shoved into everything. In the long run, that pop could be a small blip in overall development, like the dot-com bust was to the growth of the internet, but it's difficult to predict that while still in the middle of the hype cycle.

The original blog post (linked in the article) refers to this as a DynaRec, i.e. a dynamic recompiler. So it's not exactly emulating, but nor is it the ahead-of-time recompilation that Rosetta 2 can do.

Relevant XKCD. Humans have always been able to lie. Having a single form of irrefutable proof is the historical exception, not the rule.

Days before the 2016 election, 538 (which Nate Silver founded and was leading at the time) ran an article titled "Trump Is Just A Normal Polling Error Behind Clinton". Nate Silver and 538 did some of the best forecasting of that election. Don't conflate him with others' screwups.