Fundraiser Update: Only Minor Explosions So Far

It doesn’t always show, but there’s a lot of prep work for these fundraisers. I’ve got a half dozen illustrations to create, a scene to render, and a video file to edit, most of which has to be finished before Saturday. I’m also finding Steam games are finicky; Crypt of the Necrodancer disables my secondary monitors if I try to run it fullscreen, and only properly recognized my controller once. It still works with a controller, mind you, but the digital pad is disabled. Monster Prom fails to load at all! I’m guessing it does something funky when initializing OpenGL, but without proper debugging symbols I can’t say much. I have both a plan and a backup plan for Monster Prom, so it should still be a go.

Last I heard, we were up to $1,629.34 in donations towards FtB. That’s a few days old, but my general impression is that we’re not pulling in as much as we did last month. There’s no shame in that, times are tough and this fundraiser is a marathon. If you’d like to boost those numbers, your two choices are FtB’s PayPal or Skepticon’s Paypal. The GoFundMe is still down, and at this point I doubt it’ll come back up.

Now if you’ll excuse me, I’ve got a hard drive to reinstall.

Fundraiser Update: Cancelled Panel + New Target

Let’s get the bad news out of the way: that panel I was hosting about fake religions is cancelled. Things didn’t gel together as I hoped, and rather than present a half-baked idea it’s smarter to cancel the thing and save it for next time. I’ve deleted the original blog post to prevent confusion, but you can still view what I had planned via this archive.

Which brings us to the worse news: I’m running Princess Kaizo Land! The numbers will be a bit dated when this post goes up, but as of this typing the GoFundMe is at $80,231, an increase of $581, and we’ve had $1,694 donated via other means. That’s well beyond the $1,200 goal I set, so far beyond that I’m blown away by the amount.

But, uh, I’ve just committed to playing a Kaizo game. Yes, my choice is on the easy end of the scale, but I’m no shovda. I don’t have anywhere near the same skillz, so this is going to be an exercise in frustration. Yes, that can be fun to watch, but it ain’t fun to play.

So lemmie offer you a deal: clearly everyone wants to see me squirm a bit, but there’s more than one way to pull that off. I proposed playing a randomizer as my first target, but I said I’d play it at the “Way Cool” setting. That swaps stages and exits, consistently randomizes enemies, switches around Bowser’s castle entrances, and does some other minor scrambling. There are tougher settings, though; “Mondo” will also scramble sublevels and warps, shuffle enemy placement, tweak boss locations and difficulties, change power-ups, and randomly add ice and water physics to stages. This sits at a nice point between a full-on Kaizo game and the boring vanilla version, while still being an advancement over “Way Cool”‘s settings.

Not good enough? I’ve got one more thing to add. As you’ve probably noticed, I’m not big on comment sections. I think they can be problematic, and I don’t need the online feedback. My Twitch chat settings are almost as locked down as my blog comments. So how about I run Twitch chat much more openly than I usually do: I’ll hold comments for moderation, but that’s about it. I’ll also be checking the chat after each level, at minimum, which should keep things interactive.

Time for numbers: we’ve raised about $2,275 over seven days, pro-rating that another four days brings the total to $3,576, and adding 15% to compensate for a rush of last-minute donations brings the total to about $4,112. Add that to $79,650, and you get $83,762 as the new target for Friday the 25th before midnight. Time to fire some cash at the GoFundMe, or check the fundraising page for other ways you can contribute. Thank you so much for the contributions you’ve made so far, they’ve helped quite a bit with the legal bills!

Now would you please mind saving me from playing a Kaizo game? Pleeeeeeaase??

Fundraiser Update: Target Met! Long Live the Target

I know, it doesn’t look like it. Visit the legal GoFundMe, and you’ll see the current total is $79,760. And yet I said earlier that

As I type this, the legal GoFundMe sits at $79,650. Increase that above to $79,950, and I’ll switch from playing Super Mario World to Super Mario World Randomizer at the “Way Cool” difficulty on the 26th.

Some basic math reveals $79,760 is below $79,950, apparently well short of the goal. The missing piece is the very next sentence:

If you insist on donating only to FtB, instead of everyone impacted by Richard Carrier’s lawsuit, I’ll still count your donation towards that $79,950.

And PZ informs me that the PayPal link has raised a whopping $436.98 in one day! That math easily brings the combined total above $79,950, so it’s official. I’m playing Super Mario World Randomizer, thanks to your generosity!

But it’s going to be a pretty boring fundraiser if we hit all our goals in 24 hours. Thankfully, as I hinted at last time, I have quite a bit in reserve. Ever heard of a “Kaizo” game? Wikipedia has.

Kaizo Mario World, also known as Asshole Mario, is a series of three ROM hacks of the Super Nintendo Entertainment System video game Super Mario World, created by T. Takemoto. The term “Kaizo Mario World” is a shortened form of Jisaku no Kaizō Mario (Super Mario World) o Yūjin ni Play Saseru. The series was created by Takemoto for his friend R. Kiba.

Kaizo Mario World features extremely difficult level designs on the Super Mario World engine. The series is notable for deliberately breaking all normal rules of “accepted” level design, and introduced many staples of later Kaizo hacks, such as placing hidden blocks where the player is likely to jump, extremely fast autoscrollers, dying after the goal post and various other traps. This cruelty and the resulting frustration, as well as the skill level required, is both the purpose of the hacks and the appeal of any Let’s Play videos made of them.

The original hack was so popular that “Kaizo” became a generic term to mean any game mod that includes new levels and significantly amps up the difficulty of the original. As the name implies, Princess Kaizo Land is a kaizo game starring Princess Peach that’s significantly shorter than Super Mario World. It’s considered a “light” or easy kaizo, and has netted rave reviews. I’ve played a bit of it, but never passed the first level.

If you raise the GoFundMe to $80,850 before midnight MDT of September 25th, I’ll switch from playing Super Mario Randomizer (“Way Cool” difficulty) to Princess Kaizo Land. I know, that’s an increase of $1,200, but a) you’ve demonstrated you’ve got the funds at hand, and b) I really don’t want to play Pricess Kaizo Land. I haven’t played Super Mario World in ages, so my skills are pretty rusty. Even an easy kaizo game will be a big challenge! To preserve my sanity, I’ll guarantee to play at least four hours of Princess Kaizo Land, should the target be hit, but I can’t guarantee I’ll pass the game.

As before, any donations to the PayPal account also count towards this goal. My preference is that you donate to that GoFundMe instead of the PayPal link, though; the more cash we stuff in there, the more people will be free of legal debt. Still, it’s your money to do with as you see fit. I’m just happy you spent some of it to help others, in the face of all that 2020 has thrown at you.

Fundraiser: Super Mario World Edition

Like many of you, I grew up playing video games. They were my favorite distraction, much better than TV, and the primary way I bonded with some of my family. I went through a long spell where I stopped playing them, but thanks to a crippling Minecraft addiction I’ve been drawn back in.

When I was considering what to do for the fundraiser, my mind quickly settled on something relating to video games. It’s very much in my wheelhouse, yet not something I’ve shown on this blog. Having said that, I immediately crossed off doing a simple Minecraft stream; I do those way too often to be considered a special event.

Fortunately, I have an excellent substitute: Super Mario World. I played it a tonne as a kid, passing it several times, but I’ve barely played it since my teen years. This is a great excuse to chat about this game and my childhood, then sit down and actually play the thing from start to finish. As the speedrunners would say, I’m aiming for any percent warpless; no Star Road to zip straight to Bowser, though I also won’t take every exit of every level.

And I’ll do it all for the low, low price of $0.

No really, if not a single soul transfers a single penny into the fundraiser by midnight Friday the 25th, I’ll sit down and play Super Mario World at 10 AM MDT on the 26th.

Here’s the brilliant part: playing Super Mario World is kinda boring. It was the launch title for the Super Nintendo, sold 20 million copies, and often came bundled with the system. Pretty much everyone has played it, so while you can wring a bit of enjoyment from reflected nostalgia, it isn’t terribly compelling.

But that popularity also means a lot of other people have hit the same wall, and instead of moving on a few of them tried to find ways to make the game interesting again. A “randomizer,” for instance, is a bit of code that scrambles up a video game. Levels are done out of order, power-ups don’t do what they used to, and sometimes there are text or art changes made to the game. It turns something everything has played into something no-one has, and makes for much more compelling viewing.

As I type this, the legal GoFundMe sits at $79,650. Increase that above to $79,950, and I’ll switch from playing Super Mario World to Super Mario World Randomizer at the “Way Cool” difficulty on the 26th. If you insist on donating only to FtB, instead of everyone impacted by Richard Carrier’s lawsuit, I’ll still count your donation towards that $79,950. I can’t guarantee I’ll complete the randomized version, as it’s possible to encounter a version that’s impossible to pass (or beyond my skill level!) and nobody’s verified this version is passable, but I’ll give it a go for at least four hours.

So click the link and donate to the legal GoFundMe. And don’t worry about blowing past that goal, I have plenty more up my sleeve. If you want to see what else is happening this fundraiser, we’ve also got you covered.

[HJH 2020-09-14: Whoops, forgot to complete a sentence. I also added a link to the fundraising page proper, now that it’s live.]

Trying an Experiment

Usually when I get addicted to something, obsessing over it for a week or two is enough to get it out of my system. My Minecraft addiction has been going strong for a month and a half, though, with [no] signs of abating. The time I used to use to daydream posts has instead been turned to testing the best design for a lava blade or roughing up a flexible design for a mountain home in Blender.

When pondering what to do about the situation, though, I realized something: I’m a noob, but not a complete one. Yes, I can say “mob grinder” with a straight face, but I’ve also died tonnes of times in order to establish my base. I still don’t have any enchantments, I can count the number of iron bars I own on one hand, but much of that tardiness is because I’ve been more focused on finding the right location for a home. I’m not bragging about my amazing builds or redstone skillz, but I’m also not flailing around either, and I haven’t seen a lot of Minecraft players with that intermediate skill level.

I figure it might be watchable. And if I’m putting in the hours, I might as well try my hand at streaming the game. And so, I have. I’ve currently got five videos up, in fact, though I hear Twitch deletes old videos unless you fork over cash. If videos start disappearing, I’ll archive them on YouTube.

Until that point, here’s a quick overview of what I’m hoping to build and the constraints I’ve placed on myself, then four more videos where I start executing on it. Looking back on the series, I’m still a bit amazed at the pace of improvement, on a technical level. There’s no way I can keep that pace up, but at least I can always craft more mines.

Fundraising Update 1

TL;DR: We’re pretty much on track, though we also haven’t hit the goal of pushing the fund past $78,890.69. Donate and help put the fund over the line!

With the short version out of the way, let’s dive into the details. What’s changed in the past week and change?

import datetime as dt

import matplotlib.pyplot as pl

import pandas as pd
import pandas.tseries.offsets as pdto


cutoff_day = dt.datetime( 2020, 5, 27, tzinfo=dt.timezone(dt.timedelta(hours=-6)) )

donations = pd.read_csv('donations.cleaned.tsv',sep='\t')

donations['epoch'] = pd.to_datetime(donations['created_at'])
donations['delta_epoch'] = donations['epoch'] - cutoff_day
donations['delta_epoch_days'] = donations['delta_epoch'].apply(lambda x: x.days)

# some adjustment is necessary to line up with the current total
donations['culm'] = donations['amount'].cumsum() + 14723

new_donations_mask = donations['delta_epoch_days'] > 0
print( f"There have been {sum(new_donations_mask)} donations since {cutoff_day}." )
There have been 8 donations since 2020-05-27 00:00:00-06:00.

There’s been a reasonable number of donations after I published that original post. What does that look like, relative to the previous graph?

pl.figure(num=None, figsize=(8, 4), dpi=150, facecolor='w', edgecolor='k')

pl.plot( donations['delta_epoch_days'], donations['culm'], '-',c='#aaaaaa')
pl.plot( donations['delta_epoch_days'][new_donations_mask], \
        donations['culm'][new_donations_mask], '-',c='#0099ff')

pl.title("Defense against Carrier SLAPP Suit")

pl.xlabel("days since cutoff")
pl.ylabel("dollars")
pl.xlim( [-365.26,donations['delta_epoch_days'].max()] )
pl.ylim( [55000,82500] )
pl.show()

An updated chart from the past year. New donations are in blue.

That’s certainly an improvement in the short term, though the graph is much too zoomed out to say more. Let’s zoom in, and overlay the posterior.

# load the previously-fitted posterior
flat_chain = np.loadtxt('starting_posterior.csv')


pl.figure(num=None, figsize=(8, 4), dpi=150, facecolor='w', edgecolor='k')

x = np.array([0, donations['delta_epoch_days'].max()])
for m,_,_ in flat_chain:
    pl.plot( x, m*x + 78039, '-r', alpha=0.05 )
    
pl.plot( donations['delta_epoch_days'], donations['culm'], '-', c='#aaaaaa')
pl.plot( donations['delta_epoch_days'][new_donations_mask], \
        donations['culm'][new_donations_mask], '-', c='#0099ff')

pl.title("Defense against Carrier SLAPP Suit")

pl.xlabel("days since cutoff")
pl.ylabel("dollars")
pl.xlim( [-3,x[1]+1] )
pl.ylim( [77800,79000] )

pl.show()

A zoomed-in view of the new donations, with posteriors overlaid.

Hmm, looks like we’re right where the posterior predicted we’d be. My targets were pretty modest, though, consisting of an increase of 3% and 10%, so this doesn’t mean they’ve been missed. Let’s extend the chart to day 16, and explicitly overlay the two targets I set out.

low_target = 78890.69
high_target = 78948.57
target_day = dt.datetime( 2020, 6, 12, 23, 59, tzinfo=dt.timezone(dt.timedelta(hours=-6)) )
target_since_cutoff = (target_day - cutoff_day).days

pl.figure(num=None, figsize=(8, 4), dpi=150, facecolor='w', edgecolor='k')

x = np.array([0, target_since_cutoff])
pl.fill_between( x, [78039, low_target], [78039, high_target], color='#ccbbbb', label='blog post')
pl.fill_between( x, [78039, high_target], [high_target, high_target], color='#ffeeee', label='video')

pl.plot( donations['delta_epoch_days'], donations['culm'], '-',c='#aaaaaa')
pl.plot( donations['delta_epoch_days'][new_donations_mask], \
        donations['culm'][new_donations_mask], '-',c='#0099ff')

pl.title("Defense against Carrier SLAPP Suit")

pl.xlabel("days since cutoff")
pl.ylabel("dollars")
pl.xlim( [-3, target_since_cutoff] )
pl.ylim( [77800,high_target] )

pl.legend(loc='lower right')
pl.show()

The previous graph, this time with targets overlaid.

To earn a blog post and video on Bayes from me, we need the line to be in the pink zone by the time it reaches the end of the graph. For just the blog post, it need only be in the grayish- area. As you can see, it’s painfully close to being in line with the lower of two goals, though if nobody donates between now and Friday it’ll obviously fall quite short.

So if you want to see that blog post, get donating!

Fundraising Target Number 1

If our goal is to raise funds for a good cause, we should at least have an idea of where the funds are at.

(Click here to show the code)
created_at amount epoch delta_epoch culm
0 2017-01-24T07:27:51-06:00 10.0 2017-01-24 07:27:51-06:00 -1218 days +19:51:12 14733.0
1 2017-01-24T07:31:09-06:00 50.0 2017-01-24 07:31:09-06:00 -1218 days +19:54:30 14783.0
2 2017-01-24T07:41:20-06:00 100.0 2017-01-24 07:41:20-06:00 -1218 days +20:04:41 14883.0
3 2017-01-24T07:50:20-06:00 10.0 2017-01-24 07:50:20-06:00 -1218 days +20:13:41 14893.0
4 2017-01-24T08:03:26-06:00 25.0 2017-01-24 08:03:26-06:00 -1218 days +20:26:47 14918.0

Changing the dataset so the last donation happens at time zero makes it both easier to fit the data and easier to understand what’s happening. The first day after the last donation is now day one.

Donations from 2017 don’t tell us much about the current state of the fund, though, so let’s focus on just the last year.

(Click here to show the code)

The last year of donations, for the lawsuit fundraiser.

The donations seem to arrive in bursts, but there have been two quiet portions. One is thanks to the current pandemic, and the other was during last year’s late spring/early summer. It’s hard to tell what the donation rate is just by eye-ball, though. We need to smooth this out via a model.
The simplest such model is linear regression, aka. fitting a line. We want to incorporate uncertainty into the mix, which means a Bayesian fit. Now, what MCMC engine to use, hmmm…. emcee is my overall favourite, but I’m much too reliant on it. I’ve used PyMC3 a few times with success, but recently it’s been acting flaky. Time to pull out the big guns: Stan. I’ve been avoiding it because pystan‘s compilation times drove me nuts, but all the cool kids have switched to cmdstanpy when I looked away. Let’s give that a whirl.

(Click here to show the code)
CPU times: user 5.33 ms, sys: 7.33 ms, total: 12.7 ms
Wall time: 421 ms
CmdStan installed.

We can’t fit to the entire three-year time sequence, that just wouldn’t be fair given the recent slump in donations. How about the last six months? That covers both a few donation burts and a flat period, so it’s more in line with what we’d expect in future.

(Click here to show the code)
There were 117 donations over the last six months.

With the data prepped, we can shift to building the linear model.

(Click here to show the code)

I could have just gone with Stan’s basic model, but flat priors aren’t my style. My preferred prior for the slope is the inverse tangent, as it compensates for the tendency of large slope values to “bunch up” on one another. Stan doesn’t offer it by default, but the Cauchy distribution isn’t too far off.

We’d like the standard deviation to skew towards smaller values. It naturally tends to minimize itself when maximizing the likelihood, but an explicit skew will encourage this process along. Gelman and the Stan crew are drifting towards normal priors, but I still like a Cauchy prior for its weird properties.

Normally I’d plunk the Gaussian distribution in to handle divergence from the deterministic model, but I hear using Student’s T instead will cut down the influence of outliers. Thomas Wiecki recommends one degree of freedom, but Gelman and co. find that it leads to poor convergence in some cases. They recommend somewhere between three and seven degrees of freedom, but skew towards three, so I’ll go with the flow here.

The y-intercept could land pretty much anywhere, making its prior difficult to figure out. Yes, I’ve adjusted the time axis so that the last donation is at time zero, but the recent flat portion pretty much guarantees the y-intercept will be higher than the current amount of funds. The traditional approach is to use a flat prior for the intercept, and I can’t think of a good reason to ditch that.

Not convinced I picked good priors? That’s cool, there should be enough data here that the priors have minimal influence anyway. Moving on, let’s see how long compilation takes.

(Click here to show the code)
CPU times: user 4.91 ms, sys: 5.3 ms, total: 10.2 ms
Wall time: 20.2 s

This is one area where emcee really shines: as a pure python library, it has zero compilation time. Both PyMC3 and Stan need some time to fire up an external compiler, which adds overhead. Twenty seconds isn’t too bad, though, especially if it leads to quick sampling times.

(Click here to show the code)
CPU times: user 14.7 ms, sys: 24.7 ms, total: 39.4 ms
Wall time: 829 ms

And it does! emcee can be pretty zippy for a simple linear regression, but Stan is in another class altogether. PyMC3 floats somewhere between the two, in my experience.

Another great feature of Stan are the built-in diagnostics. These are really handy for confirming the posterior converged, and if not it can give you tips on what’s wrong with the model.

(Click here to show the code)
Processing csv files: /tmp/tmpyfx91ua9/linear_regression-202005262238-1-e393mc6t.csv, /tmp/tmpyfx91ua9/linear_regression-202005262238-2-8u_r8umk.csv, /tmp/tmpyfx91ua9/linear_regression-202005262238-3-m36dbylo.csv, /tmp/tmpyfx91ua9/linear_regression-202005262238-4-hxjnszfe.csv

Checking sampler transitions treedepth.
Treedepth satisfactory for all transitions.

Checking sampler transitions for divergences.
No divergent transitions found.

Checking E-BFMI - sampler transitions HMC potential energy.
E-BFMI satisfactory for all transitions.

Effective sample size satisfactory.

Split R-hat values satisfactory all parameters.

Processing complete, no problems detected.

The odds of a simple model with plenty of datapoints going sideways are pretty small, so this is another non-surprise. Enough waiting, though, let’s see the fit in action. First, we need to extract the posterior from the stored variables …

(Click here to show the code)
There are 256 samples in the posterior.

… and now free of its prison, we can plot the posterior against the original data. I’ll narrow the time window slightly, to make it easier to focus on the fit.

(Click here to show the code)

The same graph as before, but now slightly zoomed in on and with trendlines visible.

Looks like a decent fit to me, so we can start using it to answer a few questions. How much money is flowing into the fund each day, on average? How many years will it be until all those legal bills are paid off? Since humans aren’t good at counting in years, let’s also translate that number into a specific date.

(Click here to show the code)
mean/std/median slope = $51.62/1.65/51.76 per day

mean/std/median years to pay off the legal fees, relative to 2020-05-25 12:36:39-05:00 =
	1.962/0.063/1.955

mean/median estimate for paying off debt =
	2022-05-12 07:49:55.274942-05:00 / 2022-05-09 13:57:13.461426-05:00

Mid-May 2022, eh? That’s… not ideal. How much time can we shave off, if we increase the donation rate? Let’s play out a few scenarios.

(Click here to show the code)
median estimate for paying off debt, increasing rate by   1% = 2022-05-02 17:16:37.476652800
median estimate for paying off debt, increasing rate by   3% = 2022-04-18 23:48:28.185868800
median estimate for paying off debt, increasing rate by  10% = 2022-03-05 21:00:48.510403200
median estimate for paying off debt, increasing rate by  30% = 2021-11-26 00:10:56.277984
median estimate for paying off debt, increasing rate by 100% = 2021-05-17 18:16:56.230752

Bumping up the donation rate by one percent is pitiful. A three percent increase will almost shave off a month, which is just barely worthwhile, and a ten percent increase will roll the date forward by two. Those sound like good starting points, so let’s make them official: increase the current donation rate by three percent, and I’ll start pumping out the aforementioned blog posts on Bayesian statistics. Manage to increase it by 10%, and I’ll also record them as videos.

As implied, I don’t intend to keep the same rate throughout this entire process. If you surprise me with your generosity, I’ll bump up the rate. By the same token, though, if we go through a dry spell I’ll decrease the rate so the targets are easier to hit. My goal is to have at least a 50% success rate on that lower bar. Wouldn’t that make it impossible to hit the video target? Remember, though, it’ll take some time to determine the success rate. That lag should make it possible to blow past the target, and by the time this becomes an issue I’ll have thought of a better fix.

Ah, but over what timeframe should this rate increase? We could easily blow past the three percent target if someone donates a hundred bucks tomorrow, after all, and it’s no fair to announce this and hope your wallets are ready to go in an instant. How about… sixteen days. You’ve got sixteen days to hit one of those rate targets. That’s a nice round number, for a computer scientist, and it should (hopefully!) give me just enough time to whip up the first post. What does that goal translate to, in absolute numbers?

(Click here to show the code)
a   3% increase over 16 days translates to $851.69 + $78039.00 = $78890.69

Right, if you want those blog posts to start flowing you’ve got to get that fundraiser total to $78,890.69 before June 12th. As for the video…

(Click here to show the code)
a  10% increase over 16 days translates to $909.57 + $78039.00 = $78948.57

… you’ve got to hit $78,948.57 by the same date.

Ready? Set? Get donating!

It’s Payback Time

I’m back! Yay! Sorry about all that, but my workload was just ridiculous. Things should be a lot more slack for the next few months, so it’s time I got back blogging. This also means I can finally put into action something I’ve been sitting on for months.

Richard Carrier has been a sore spot for me. He was one of the reasons I got interested in Bayesian statistics, and for a while there I thought he was a cool progressive. Alas, when it was revealed he was instead a vindictive creepy asshole, it shook me a bit. I promised myself I’d help out somehow, but I’d already done the obsessive analysis thing and in hindsight I’m not convinced it did more good than harm. I was at a loss for what I could do, beyond sharing links to the fundraiser.

Now, I think I know. The lawsuits may be long over, thanks to Carrier coincidentally dropping them at roughly the same time he came under threat of a counter-suit, but the legal bill are still there and not going away anytime soon. Worse, with the removal of the threat people are starting to forget about those debts. There have been only five donations this month, and four in April. It’s time to bring a little attention back that way.

One nasty side-effect of Carrier’s lawsuits is that Bayesian statistics has become a punchline in the atheist/skeptic community. The reasoning is understandable, if flawed: Carrier is a crank, he promotes Bayesian statistics, ergo Bayesian statistics must be the tool of crackpots. This has been surreal for me to witness, as Bayes has become a critical tool in my kit over the last three years. I suppose I could survive without it, if I had to, but every alternative I’m aware of is worse. I’m not the only one in this camp, either.

Following the emergence of a novel coronavirus (SARS-CoV-2) and its spread outside of China, Europe is now experiencing large epidemics. In response, many European countries have implemented unprecedented non-pharmaceutical interventions including case isolation, the closure of schools and universities, banning of mass gatherings and/or public events, and most recently, widescale social distancing including local and national lockdowns. In this report, we use a semi-mechanistic Bayesian hierarchical model to attempt to infer the impact of these interventions across 11 European countries.

Flaxman, Seth, Swapnil Mishra, Axel Gandy, H Juliette T Unwin, Helen Coupland, Thomas A Mellan, Tresnia Berah, et al. “Estimating the Number of Infections and the Impact of Non- Pharmaceutical Interventions on COVID-19 in 11 European Countries,” 2020, 35.

In estimating time intervals between symptom onset and outcome, it was necessary to account for the fact that, during a growing epidemic, a higher proportion of the cases will have been infected recently (…). Therefore, we re-parameterised a gamma model to account for exponential growth using a growth rate of 0·14 per day, obtained from the early case onset data (…). Using Bayesian methods, we fitted gamma distributions to the data on time from onset to death and onset to recovery, conditional on having observed the final outcome.

Verity, Robert, Lucy C. Okell, Ilaria Dorigatti, Peter Winskill, Charles Whittaker, Natsuko Imai, Gina Cuomo-Dannenburg, et al. “Estimates of the Severity of Coronavirus Disease 2019: A Model-Based Analysis.” The Lancet Infectious Diseases 0, no. 0 (March 30, 2020). https://doi.org/10.1016/S1473-3099(20)30243-7.

we used Bayesian methods to infer parameter estimates and obtain credible intervals.

Linton, Natalie M., Tetsuro Kobayashi, Yichi Yang, Katsuma Hayashi, Andrei R. Akhmetzhanov, Sung-mok Jung, Baoyin Yuan, Ryo Kinoshita, and Hiroshi Nishiura. “Incubation Period and Other Epidemiological Characteristics of 2019 Novel Coronavirus Infections with Right Truncation: A Statistical Analysis of Publicly Available Case Data.” Journal of Clinical Medicine 9, no. 2 (February 2020): 538. https://doi.org/10.3390/jcm9020538.

A significant chunk of our understanding of COVID-19 depends on Bayesian statistics. I’ll go further and argue that you cannot fully understand this pandemic without it. And yet thanks to Richard Carrier, the atheist/skeptic community is primed to dismiss Bayesian statistics.

So let’s catch two stones with one bird. If enough people donate to this fundraiser, I’ll start blogging a course on Bayesian statistics. I think I’ve got a novel angle on the subject, one that’s easier to slip into than my 201-level stuff and yet more rigorous. If y’all really start tossing in the funds, I’ll make it a video series. Yes yes, there’s a pandemic and potential global depression going on, but that just means I’ll work for cheap! I’ll release the milestones and course outline over the next few days, but there’s no harm in an early start.

Help me help the people Richard Carrier hurt. I’ll try to make it worth your while.

Timeline: Rachel Oates and EssenceOfThought

I’ve already covered some of this material, as has EoT, so you might be wondering why I’m repeating myself months after the events in question.

The old stuff hasn’t been well-organized nor placed in chronological order. My own efforts, for instance, were at the end of the second-half of a long blog post where I was pretty harsh on Rationality Rules. There’s room for a more dispassionate summary of the full context of what happened, especially if allegations about this “will be amplified by social media and echo for weeks, months, maybe years.” I’m pretty firmly on EoT’s side, but by minimizing my commentary in favour of direct quotes I can create a summary that Rachel Oates’ supporters will also find useful. The primary bias of this post will thus be via lies of omission, so I’ll try to be as comprehensive as possible. There’s also material that neither EoT nor I have mentioned, most of it focused on Rachel Oates’ side of the equation, so her point of view is better represented.

With that intro out of the way, let’s begin at the beginning. All dates and times are based on Twitter’s timestamp, which I think uses my timezone of Mountain Daylight Time, though it’ll be helpful to know about India Standard Time. Oh, and CONTENT WARNING for transphobia, plus mention of suicide and self-harm. [Read more…]