The Phoenix by Julia Ecklar is also really good. It makes me sad that there's no good recordings of it and I can't find any covers.

I'm skeptical that this would be good with tallow. Last time I tried it, it smelled bad, tasted bad, and had the texture of a slightly melted candle. I'd imagine bear fat would have a much better texture and plausibly a better taste. Or did I just try a particularly gross brand of tallow and it's normally good?

I've volunteered during an election before and some of the things they do with ballots are:

• Never interact with ballots without the presence of a member of both parties.
• Transport ballots in locked boxes where the transporters don't have keys (and it would be obvious if the lock was cut).
• Record everything on video.
• Keep track of every ballot given out and every ballot that didn't come back.
• Hand-count some or all of the machine-counted ballots to make sure the results match.
• Notify everyone whose vote was counted (in case they didn't actually vote).

For voting machines, I think there are audits involving paper print outs + chain of custody rules, but I don't know the details. I've only ever lived in states with paper ballots.

I think the load-bearing part is "always have a member of both parties present", which is sort-of similar to what I'm suggesting in the post. If you don't have a single party you trust to count ballots, then make both of them do it (and record everything so you can prove it if one of them complains).

Thanks for reminding me that I actually need to read this. It's frustrating that there's no audio version though. I like to switch back and forth so I can listen to books while I do chores. I wonder if AI-read audiobooks still aren't good enough?

Yeah, the problem is that we don't have a (provably) fair source of randomness.

I guess I only really touch on this in the intro, but the context was for elections where there's high stakes and potentially state-level hacking. How certain are you that the RNG you're using wasn't tampered with in the factory by the Chinese government, and that no one has tampered with it since then?

There's similar problems with voting machines, where even though there's intense chain-of-custody rules (and I think voting machines in the US are fair), a lot of people don't trust them and think the manufacturers and/or political parties are tampering with them.

Thanks! I think it's correct now.

The risk I'm worried about isn't insufficient randomness (I think this is what's alleged in the lottery story). I'm worried that the hat-picker could collude with one of the candidates to increase their chance of winning.

This is sort-of a "the AI is smarter than you" situation where I don't know exactly how they'd do it, but I imagine if you gave Penn & Teller a hat, a pen, and a stack of paper, they could convincingly select the same "random" piece of paper over and over again. And even if they couldn't actually do it, if some voters are convinced that they did, then your election still has a legitimacy problem.

Yeah, my friends who knew about the exceptions didn't care at all. The only case where I think it would be a problem is hiding your exceptions from someone you're trying to convert. If you don't actually eat a fully vegan diet, convincing other people to do that based on your example is misleading. But that's presumably a less common thing where explaining the details and why is worth it (and will likely help you convince people since it shows you've actually thought it through).

Going “I am a lacto-bovitarian for the animals because of the inherent uncertainty of nutritional science I’ve calculated that this is the best way to maximize my impact” sounds more like a ACX bay area house party story than something that will make the average person question their own impact.

You don't really have to explain in this much detail. I was vegan with some exceptions for a while and mostly didn't bother explaining the exceptions to people (for example, if my mom got eggs from one of her friends' back yards to make Thanksgiving pie). A friend still tells people he's vegetarian because it's not really worth the effort to explain all of the weird edge cases.

The [injection site reactions] might be a combination of the volume and pH level of the injection.

Do you have any opinions whether increasing or decreasing the volume is likely to help with this? I saw people recommending diluting 20 mg of retatrutide in as little as 1 mL of solution or as much as 4 mL.

Update: I meant 1-4 mL, not 10-40mL.