Against the Mathematicians

twisting and turning to rid myself of human language

[epistemic-horror]

We replicate the entire anomalies literature in finance and accounting by compiling a largest-to-date data library that contains 447 anomaly variables. With microcaps alleviated via New York Stock Exchange breakpoints and value-weighted returns, 286 anomalies (64%) including 95 out of 102 liquidity variables (93%) are insignificant at the conventional 5% level. Imposing the cutoff t-value of three raises the number of insignificance to 380 (85%). Even for the 161 significant anomalies, their magnitudes are often much lower than originally reported. Out of the 161, the q-factor model leaves 115 alphas insignificant (150 with t < 3).

the field of academic finance is fake news

[fuck-customers]

Saw this at a local restaurant. I think it should be hung up at every restaurant…

[dronegoddess-deactivated2017053]

I thought this was a 404 error at first.

[epistemic-horror]

“wait, but servers aren’t human at all!”

[epistemic-horror]

CNN on Twitter

“The Trump administration is ending Michelle Obama's "Let Girls Learn" education program https://t.co/EeOS19CbyY”

P O W E R F U L

[epistemic-horror]

It’d been years that I had been saying to myself “There’s gotta be a better way. What is the better way? I’m gonna keep looking around, find the better way.” And so I stumbled upon [redacted], and I was kind of like yeah, okay whatever, and eventually I kept seeing it enough times that I was like okay, I’ll give this a shot, whatever, and I looked, and I was able to start taking my problem list of “How are you gonna handle this? How are you gonna handle this?” and every single thing I threw at it there was a good solution. And I was like, Hmmm. Maybe there’s something here.

[epistemic-horror]

(the joke here which i forgot to put in the tags is that what i redacted is “functional programming”)

[epistemic-horror]

It’d been years that I had been saying to myself “There’s gotta be a better way. What is the better way? I’m gonna keep looking around, find the better way.” And so I stumbled upon [redacted], and I was kind of like yeah, okay whatever, and eventually I kept seeing it enough times that I was like okay, I’ll give this a shot, whatever, and I looked, and I was able to start taking my problem list of “How are you gonna handle this? How are you gonna handle this?” and every single thing I threw at it there was a good solution. And I was like, Hmmm. Maybe there’s something here.

[nostalgebraist]

One strange thing about the Homestuck community was that people were always claiming Homestuck had gotten worse, relative to some point of comparison that was constantly shifting – the glory days were in early or late 2010, they were in 2011, they were in 2012 – and although these looked similar on the surface to the standard hipster gripes found in every community (”before they sold out” etc.), there was a strange sense of fundamental truth to them.

I first read it in the fall of 2011, and when Act 6 started, I was one of the first people to start complaining that something felt off about it.  People told me to give it time, and I did, and then I felt vindicated when other people finally decided “yeah, Act 6 has really gone to shit” … even though people were deciding that years and thousands of pages after I did, at very different points in the story.  And even when I first joined the conversation, there were people who thought I was uncritically eating up low-quality content after a point, that it had stopped being good somewhere in 2011 as the schedule slowed down, or in late 2010 when everything started being about trolls.

People got off the train at drastically different points, but the fact that you eventually would get off the train started to seem like an inherent property of the work itself: it is very widely agreed of Homestuck that it “stops being good at some point,” even though there is little agreement on where that point is.  (As if to prove the rule, many/most of the relatively few people who were still on board in the last days were put off by the ending itself.)

I’m saying this after reading a post talking about how Homestuck (as in, the continued output of the media franchise now that the comic is over) has gone downhill, as illustrated by stuff that happened today, in 2017.  It sure sounds bad, as one would expect it to be, given how much time it’s had to decline!  (By 2030, “Homestuck” will be listed in some future president’s equivalent of the Axis of Evil.)

The explanation for this is probably something boring having to do with people valuing many distinct things in an expansive work, but a magical, irrational part of my brain wants to believe that a trend this steady is the sign of something deep and natural, like some geological cycle, and that one day, as with the passing of seasons and the rise and fall of empires, we (or our descendants??) will see Homestuck soar back into greatness.

[epistemic-horror]

i’ve had some free time on my hands lately and have spent it rewatching classic simpsons episodes, as one does

and sometimes i go to the simpsons archive for a really great episode and it’s incredible how many Bs and C+s and D-s there are in the contemporary reviews from usenet fans

and it’s exactly the same, in seasons 3 and 4 the internet was complaining that it wasn’t enough like seasons 1 or 2, in seasons 5 and 6 people thought the show peaked in seasons 3 and 4, and so on

this seems like a fairly typical problem for long-running serials

[slatestarscratchpad]

Are hiperstimuli and the rising addictivity of products (snacks, Netflix, social media, erotica, games) a serious concern for mental health and social freedoms? Is it high time to reset our reward systems and seek simple, slow living? Quoting Adam Alter: "Why are the world’s greatest public technocrats also its greatest private technophobes? It seemed as if they were following the cardinal rule of drug dealing: never get high on your own supply."

I wouldn’t frame it in terms of “mental health” - I think it’s on an axis skew to anxiety/depression/etc - but yeah, I find it pretty scary. I’m reading “The Hungry Brain” now which argues that obesity is because modern food is a superstimulus (well, it’s more complicated than that, I’ll review it eventually). In other areas, though, I’m surprised at how well we’re doing. Tumblr is probably a superstimulus for community or something, but not really in a bad way. It just means we’re getting more community. Netflix and erotica and so on seem to be making a lot of people very happy with unclear side effects (I know some people think porn has weird mental health issues, but my barrier for evidence here is high because Puritans gonna Puritan).

I expect at some point someone will come up with something horrible and then we’ll have to really confront this. If nothing else, the Amish will inherit the earth and we can start over from there.

Also, when did “superstimulus” become “hyperstimulus”. Stimulus inflation?

[bpd-dylan-hall-deactivated20190]

Tumblr isn’t really a superstimulus for community, it’s just a regular community

television is a superstimulus for community

it wiped out all the communities that used to exist, which means that the Internet is now basically the only option for a proper community space

it is sort of unfortunate that the only functional community space is social media, which is engineered to be addictive, but this is also a complaint you could make about bars

[epistemic-horror]

go to church lol

[epistemic-horror]

probably going to kill this persona pretty soon, just so everyone knows. it’s not really who i want to be, anymore

[worldoptimization-deactivated20]

Cathy O’Neil is worried about insurers using data to predict patients’ health outcomes. She writes in Bloomberg View that “if we’re not careful, pretty soon it’ll be almost like there’s no insurance at all.”

I feel like this reflects kind of a weird view about the purpose of insurance.

Why do I have, say, flood insurance on my house? Let’s say there’s a 1% chance that my house gets destroyed in a flood, and my house is worth a million dollars. In a world without insurance, I have to save all the money I possibly can in order to be prepared for this potential cost, and even if I do I still probably won’t have saved enough and I’ll end up bankrupt or something. With insurance, I can just pay the $10,000 and not worry about it again.

The insurance is necessary because of our uncertainty: no one can predict exactly when and where floods will happen. It’s not about making everyone share the risks of floods. It would be ridiculous to suggest that it’s only fair that someone in Iowa pay the same for flood insurance as someone in Florida. The person in Florida should internalize the costs of the extra risks they’re taking.

And like, if we somehow got better at predicting floods, that would be great, right? Sure, it would make some people’s insurance more expensive. But it would make others’ cheaper. And it would allow everyone to better account for the cost of floods when thinking about the tradeoffs of different houses.

Substitute cancer or heart disease for floods. What makes that situation different?

I guess one thing people would say is that where you live is your choice, while your health outcomes aren’t. But I don’t think that’s super clear. You can do lots of things (exercise, diet, sleep) to influence your health outcomes. And there are lots of reasons people might be constrained in their choice of where to live–job opportunities, family members who need caregiving, whatever.

I think that what Cathy O'Neil really wants is socialized medicine. She thinks it’s unfair that people with worse health have to pay more for healthcare, and the government should just pay for everyone’s health care out of progressive taxes or whatever. And sure, it is unfair.

But what we have is a healthcare system based on private insurance. And I’m not sure how I feel about trying to make that more like socialized medicine by actively trying as a society to be as bad at making predictions about people’s health as possible.

[epistemic-horror]

 also if insurer big data helps us get better at predicting health outcomes, that might not be unalloyed terribleness?

the story bob gordon tells is: modern life insurance became widespread in america in the early 20th century. since life insurance companies are interested in making sure their policyholders stay alive, 

[B]y the early twentieth century most large [insurance] companies provided information on personal health and hygiene. Several companies distributed millions of pamphlets on topics such as avoidance of tuberculosis, care of children, purity of milk, “flies and filth,” and a host of other potential hazards. Some companies even provided free medical examinations and visits by nurses.

Your health insurer ultimately wants you to not get sick or hurt, because that costs them money, and while dropping people who are likely to get sick or hurt is one possible effect of better prediction, another one is helping people in general minimize their risk of getting sick or hurt. seems to me we should discourage the first and encourage the second.

[resinsculpture-deactivated20221]

For some reason my brain won’t shut up with particular phrases, sometimes. Right now it’s stuck on this one: “We gave each other, at that moment, looks of fiendish disregard”.

[raginrayguns]

How did thomas jefferson figure out what rights we have? Have we found any new ones since then? Was he wrong about any of the ones he found?

Hmm I’m reading the “Natural and Legal rights” Wikipedia article and it looks like there’s lots of ongoing research and controversy. But there’s no list of the settled, agreed upon ones for which indisputable evidence has been discovered

[raginrayguns]

Oh shit guys actually there’s a UN Universal Declaration of Human Rights! http://www.un.org/en/universal-declaration-human-rights/ And one of them is not being enslaved, how did Thomas Jefferson miss that one?

[slatestarscratchpad]

Jefferson’s early studies of rights are now known to have been underpowered. If he had just used a larger sample size, none of this would have happened.

[epistemic-horror]

bentham was right, as usual

[epistemic-horror]

i think i saw at least four or five javascript libraries with emoji all over their github READMEs today

[epistemic-horror]

joke: javascript turns you gay

woke: ecmascript turns you into eurotrash

[epistemic-horror]

tfw all ur tests pass and ur not sure if that means ur code works or ur tests are shitty

[epistemic-horror]

joke: javascript turns you gay

woke: ecmascript turns you into eurotrash

[resinsculpture-deactivated20221]

Is there anything that you would recommend of nydwracu's as a gentle introduction?

I’m the wrong person to ask; I’ve dealt with him on social media and read some of the accompanying rants, but I’m not really familiar with his “serious” work. Maybe @epistemic-horror has some suggestions?

[epistemic-horror]

“stop paying attention to politics and go to church lol” -- nydwracu, basically

[epistemic-horror]

don’t let this excellent story about a trans boy on test being forced to compete in --- and easily winning --- the texas high school girls’ wrestling state championship distract you from the fact that the washington post’s new slogan is ridiculous

[epistemic-horror]

it’s also possible that twitter decided that a tweet openly mocking the cia/’deep state’ from an account with a 9/11 pepe avatar and the unabomber manifesto in the bio showing up on the president’s timeline was a bad look for the company.