Co-lead (Near-Term Detection) at the Nucleic Acid Observatory in Boston. Speaking for myself unless I say otherwise.
jefftk
Karma: 21,602
Scaling the NAO’s Stealth Pathogen Early-Warning System
Attending LessOnline
I like this idea a lot, but I’m nervous about setting the right CPU threshold. Too low and it never shuts off, too high and it shuts down in the middle of something when waiting for a slow download. But possibly if I looked at load logs I’d see it’s so clearly either ~zero or >>zero that it’s not fussy?
Auto Shutdown Script
Pictures for 2024
Why?
Silly Time
Counting Objections to Housing
Fixed! I was missing a comma.
Bike Lights are Cheap Enough to Give Away
Penny Whistle in E?
Camps Should List Bands
Contra Dance Pay and Inflation
Interesting! That Boston Public Schools switched from this mechanism to Gale-Shapley seems like it might be useful in convincing our school board (which is separate from the BPS school board, since schools are municipality-level here) to switch.
Middle School Choice
Saving Zest
Real-Time Gigstats
Their definition of “Price gouging occurs in a competitive market when lowering the price from the market-clearing level would increase total Utilitarian welfare” is a bit sneaky: it means that any time I say “here’s an example of where price gouging helps improve disaster response” they can just say “but that’s not real price gouging, since a lower price wouldn’t increase welfare”.
It also doesn’t look to me like the paper’s approach gives a good framework for thinking about long-term investment incentives and preparation for future disasters, or people selling/renting possessions they wouldn’t normally put on the market sell (air purifiers, renting spare rooms).
The paper’s division of circumstances into price gouging vs not isn’t a good match for the real world, and leads them to support policies like the current ones that normally don’t do anything and then suddenly make large impacts in a disaster. Instead I’d like to see recognition that it’s hard to determine welfare-maximizing pricing in real time and that price signals can reach very far, and instead use a mechanism that allows price increases to occur but redistributes some of the profits.
That’s elegant in some sense, but somehow doesn’t feel like the right way to do it.