really we should all move to Fife

merci! I probably would not have been aware of it

This is very cool and I could imagine research on within-area accessibilityinequalities. Looking at the data/model they have (whatif.sonycsl.it/15mincity/15...) - I'm not sure everyone would agree going up your road is a 2 minute cycle @meganjames.bsky.social

Lack of cars and yet a considerably wider street - or is it a different angle/perspective?

This is rude - I was gone 30 seconds

The main source of error is having to infer who is actually a long-term immigrant/emigrant, vs who is coming/leaving but will depart/return soon. They use past behaviours for it, but those are poor predictors in fast-changing contexts

There's nothing hidden in the publication, though you have to go dig for it because it's a mammoth. Data is a combination of border control and tax, with some corrections for populations that don't show up on them.

There are a lot of reasons that stack on top of one another - could ramble about it for hours (I've worked on this specifically). Unreliable and changing data, methodology still being developed at a time when the numbers have increased massively, some unknowns that are out of the ONS's control, etc