And it's not just the US, I had multiple conversations with various french friends when I was studying there where the idea that none of this (prayer/RE in schools, the king being head of the CofE, bishops in the lords) was that big a deal in our politics was clearly just bizarre to them.

I can't quite remember which of these I actually played in 2015 if any but probably fallout 4 then, mgsv or Undertale now.

Meanwhile outsourcing it to the LPC while ChX takes credit for the upsides also doesn't seem ideal, and since Osborne hiked it it's been the thing to do to say you're doing something for people on low incomes (which, hm) without having to spend money which doesn't seem a great set of incentives.

The issue is that there's a strong case for it being one of the very few success stories of British economic policymaking over that period so there's temptation to keep pulling the lever - but agree we're now in uncharted territory and continuing to pull it likely comes with unpleasant side effects.

I think there's a strong case for pausing here and more research on 2nd order effects. But I worry about the idea it was all a big mistake when we have pretty good evidence up until 2019 that it put a lot of money in pockets without any significant hit on employment. share.google/fN3U5mgyvBMS...

The important caveat is that evaluations of the UK MW haven't found an employment effect, but the most recent only look up to 2019 iirc and it's gone up since. There is evidence that min wages increase rates of industrial accidents etc though so the 'jobs are harder' story feels very plausible.

That makes sense thanks - and matches my somewhat mixed experience with copilot tbh

Is Claude code significantly different to just using Claude in vs code via GitHub copilot? haven't looked into other options properly because I get copilot pro free from GitHub through their education program

the distinguishing characteristic is that under some circumstances it makes sense to take risks not that have high volatility compensated by high expected value, but risks that have negative expected value compensated by long right tails, ie the “risk” compensates expected loss. 2/

As soon as politicians stop enforcing norms things can unravel very, very quickly. Currently the Starmer legacy based off the last six months of total capitulation and ceding of ground, bar one decent speech in September.