LESSWRONG
LW

jacob_cannell's Shortform

by jacob_cannell
25th Mar 2020
1 min read
3

6

This is a special post for quick takes by jacob_cannell. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
jacob_cannell's Shortform
9jacob_cannell
4jacob_cannell
2jacob_cannell
3 comments, sorted by
top scoring
Click to highlight new comments since: Today at 10:00 AM
[-]jacob_cannell5y90

Am I one of the few people here who has looked at the covid-19 data and reached the conclusion that it's probably only about as severe/fatal as seasonal influenza?

I have a longer blog post outlining the case here.

TLDR: CFR!=IFR, influenza CFR is similar to covid-19 CFR, and we know from influenza data that typically IFR << CFR due to enormous selection sampling bias from mostly testing only those with more severe disease. We can correct for that by comparing the covid-19 confirmed case age structure to the population age structure using uniform or age-dependent attack rate. The resulting IFR is similar to influenza, which is also the best fit for the Diamond Princess data (where selection bias is mostly avoided so CFR~IFR).

Selection bias can help explain why the CFR is higher in Italy, and probably why it's so much lower in Germany (I'm looking for age structure data on covid-19 cases from Germany, I'm predicting it will be flatter than US or Italy data). South Korea is also another interesting case (which I found some data for but haven't put into the blog post yet) - we can clearly reject a typical attack rate age structure there, which was surprising at first but then made sense given that the outbreak in SK started in a large tight-knit cult with a young median age and they tested everyone in the cult.

Anyway if anyone here has already encountered these thoughts and still believes covid-19 IFR is much higher than influenza IFR I'm curious what the best arguments/evidence are.

Reply
[-]jacob_cannell3y40

Just noting some comment links for future reference:

The sequences are tragically flawed - based on some overconfident assumptions about the brain and AI which turned out to be incorrect.

CEV is probably against suicide, and so is human empowerment.

Evolution succeeded at alignment: humans are massively successful by IGF (inclusive genetic fitness) metrics, and some do actually optimize mentally for IGF.

Reply
[-]jacob_cannell2y*20

Velocity uncertainty of an electron

(A test of latex of sorts)

The uncertainty principle bounds the combined product of std deviations of position and momentum:

σxσp>=ℏ2

The momentum uncertainty is :

σp>=ℏ2σx

As p=mev, the velocity uncertainty (std deviation) is thus:

σv>=ℏ2σxme

For a 1eV electron with σx≈1nm (the electron is confined to a ~1nm cavity on order of de broglie wavelength), then:

σv>=6.5∗10−16eV∗s2∗1nm∗0.5MeV/c2

σv>=6.5∗10−16eV∗s2∗10−9m∗0.5∗106eV/c2

σv>=6.5∗10−16eV∗s10−3m∗eV/c2

σv>=6.5∗10−16eV∗s10−3m∗eV/(3∗108m/s)2

σv>=6.5∗10−16eV∗s10−3m∗eV/(9∗1016m2/s2)

σv>=6.5∗10−16eV∗s1.11∗10−20eV∗s2/m

σv>=5.85∗104m/s

Meanwhile the mean or expectation of the 1eV electron's kinetic velocity is 5.9∗105m/s ....

So the angular std dev and or linear velocity std dev on is on order ~ 10%?

Reply
Moderation Log
More from jacob_cannell
View more
Curated and popular this week
3Comments