I think it’s worth noting that this is evenly distributed precinct level data. The dots are *not* the actual locations where the people who voted live. This probably results in artificially stark boundaries between precincts. I notice some people drawing conclusions based on this artifact.

Someone missed the comms training. 😬

I misspoke. I exhausted the *plus* allocation. Using it for data entry and requirements analysis.

Still a banger: harpers.org/archive/1941...

6. Biggest Polestar 2 design flaw remains the single cup-holder design. 🤣

7. Pilot assist & radar cruise control have been pretty solid. Only conked out once not due to rain/fog. I continue to be a bit unhappy with the level of oversteering. It does require constant supervision, but it helps.

5. ABRP is pretty conservative about efficiency (estimates car is about 25% less efficient than it actually is). Means that I need to get more aggressive about state of charge. Only arrived at one charger under 20%, which is not the most efficient way to go.

3. ABRP has been really helpful for planning, but is buggy as &@#% and pretty terrible UI for navigation. Better to use it for planning and then use your preferred app for navigation.

4. CarPlay is so much better than vehicle UI/apps. And Polestar has a good built-in system, relative to comp.

Guess Becka forgot her Popper. They might want to read up on the “paradox of tolerance”. 🤷🏻‍♂️

Quality AI training input right there!

You cannot effectively program if you are not capable of building a fully functional computer with a soldering iron and a breadboard. 🤣🤷🏻‍♂️ Kids these days! 👴

I mean, better to use dependency injection, but test seams are so much better than random conditionals mucking up the code.

That’s filthy. 🤣

But seriously, this is what test seams are for, isn’t it?

I’m reconsidering my photo backup situation now. 🤣