Yep, they (almost) literally wrote the book. I found their manual super useful when trying to deal with jerk articles that only post portions of their results. I don’t care that it’s “bad practice” to post raw data, it needs to become the standard.
You found the fifth? I keep having trouble with my #10s. Sneaky little buggers.
I mean, the biggest issue with me for the great googlio isn’t the ads and the ai, both of which I hate, but the actual shit-infested results. It’s not removing the ads full of SEO that are posing as websites, it’s just giving you an old UI for the new 2025™ search.
Repair steps are one of the few tasks that I feel videos are better than words (and sometimes pictures). It definitely helps to see the motions they’re taking and a single capture of the location from walking up to the car (or other repairable object) all the way to looking at the part that needs fixing.
I prefer the dratini->gyrados / magikarp->dragonite pipeline theory, myself, but I can gestate with this caterpie branch. The venonat->butterfree coloring and eyes do seem too coincidental. Maybe the artists that worked on each one had their favorite colors, so pokemon they did all ended up schemed similarly?
The ‘not significant’ part could be due to low numbers in general, so they can’t get the variances small enough to get low p numbers. It’s a quote that I can’t quite remember perfectly that is well known in sociology/psychology: “The only reason our findings aren’t significant is because we’re too damn lazy to drag enough people in for the study.”