March Links

MIT is taking some actions to deal with credentialism in higher education, though we'll have to wait and see how it plays out.

Megan McArdle discusses the ways that adults exploit youth activism to support favored causes while ignoring kids' needs when it's no longer convenient.

An old article from PEW Journalism on how the media covers topics disproportionately to their real-life implications and importance. Related: who fact checks the fact checkers?

Rationalist bloggers review Twelve Rules for Life.

The European Space Agency is testing an air-breathing electric engine. It's not quite a jet engine that works in space, but it's surprisingly close. NASA, meanwhile, is working on 3D-printed rocket nozzles.

Planetary Resources is struggling after funding setbacks. It's unclear what their new strategy will be but the soundbytes don't inspire optimism.

People had some weird ideas about averages in the first half of the 20th century, with implications ranging from airplane design to fashion. Unsurprisingly, the fields where mistakes had measurable consequences learned the lesson faster. We're still beholded to those lessons today, with most aircraft dimensions sized to accommodate 95th percentile males and 5th percentile females.

Ethereal art, made with drones.

February Links

One of the things I missed during senior design: Scott Alexander continuing to hack at a Bayesian theory of mind. Worth comparing to Objectivist epistemology and related, realist theories.

Jacob over at Put A Number On It has some interesting advice for being lucky: instead of trying to improve the base rate, increase your variance. The post includes examples from both sports and politics, and I'm sure you can generate a few more yourself.

Slate Star Codex celebrates five years of blogging.

The Mars 2020 rover will be taking at least one Martian meteorite sample for instrument calibration. This isn't the first piece of the Red Planet to head back, however: Mars Global Surveyor still has a small piece in orbit. We could still get that one back if we rush things.

I'm not sure how you could miss it, but just in case: Falcon Heavy test flight went mostly successfully. I'm counting this toward my 2018 predictions, despite a hard landing on the central core, because the basic architecture functioned properly.

Speaking of space, everyone finally noticed the NASA plutonium shortage, because the Department of Energy recently restarted production. The KiloPower Reactor project is still progressing, with reactor tests now underway in Nevada.

Scott Sumner on Overton Window shifts in mainstream economics and the continued question of housing prices.

NVIDIA releases a "fast photorealistic style transfer algorithm" that will apply the style of one image to another. According to the paper, it purports to be significantly better than previous approaches. The code is available under Creative Commons with some honestly impressive examples.

January Links

2017 was the safest year for aviation in decades. A lot of people have tried to politicize this, but I'm pretty sure aircraft engineers and operators get 99% of the credit.

Arguably related: Why did everything take so long?

Visualization of the music from Interstellar.

Astronaut John Young passes away at 87. On a related note, a longevity research FAQ.

Falcon Heavy gets a launch date: February 6th. It'll be exciting to see whether my prediction proves correct.

An older article on dolphin intelligence.

Scott Alexander on bitcoin and bias.

Speaking of bias, NY Times opinion piece on how social media is making us dumber. Specifically, how soundbytes and decontextualized quotation lead people to believe that public figures advocate the polar opposite of what those public figures are actually advocating.

Shifting Registers

Epistemic Status: Probably re-inventing the wheel

Epistemic Effort: Regular ruminations over the span of several months

In linguistics, a register describes the differences in word choice, pronunciation, sentence construction, etc. that we use in various social settings. For a simple example, you would use a totally different register talking to your boss in a meeting than your spouse after dinner. 

Probably. It wouldn't surprise me if some people use the same register everywhere. I definitely have trouble shifting to a friendly register, even with people I'm sorta comfortable with. But that's not the point of this post. Rather, I want to steal the term to describe a more psychological phenomena.

First, though, let me tell you a story.

Back during my last semester at Purdue, my mom had some inkling of how serious the situation was, and decided to sign me up for a session with her therapist when I was around for spring break. The session was largely forgettable, in no small part because you don't really need a hypnotist to put you to sleep when you're as tired as I was, but one thing I remember saying stuck out.

I was trying to describe how my stress response worked, and ran into an issue explaining it. You see, how I operated when I was in West Lafayette was almost entirely divorced with how I operated in Kansas City.

This makes perfect sense when you think about it. When I was at home, I was sleeping all day and didn't have class. When I was at school, I had classes and homework and exams and so on. At school, I lived in a dorm, with a roommate, showered and brushed my teeth in a communal bathroom, etc. At home, I had dramatically more privacy. I rode in cars instead of walking or taking the bus.

I tried to explain that my behavior, mindset, whatever you want to call it was very different when I was at home for a break, even a mid-semester break, than at school. I don't think I did a very good job. But that realization has stuck with me.

This is what I'm describing as a register. Even now, I'm in a very different register when I'm at home or at school. Yesterday I was in Kansas City, today I'm in Lawrence. In KC, I'm mostly killing time because I literally never know when I'll be interrupted for whatever activity my parents want me for. My most productive hours were monopolized by manual labor in the continuing move process, and then, exhausted, I couldn't exert decent discipline in a non-optimized environment.

Today, I am the only person ultimately controlling my daily schedule. When I finish writing this blog post, I'll finish my grocery list and go to the store, put in laundry after I come back, and start reading up for this semester. I still don't have great self-discipline, but my routine is designed around my needs and apartment layout.

Concrete examples don't really get at the underlying difference, unfortunately. I'm not trying to describe the way my routine varies, but rather the way my mindset varies. At school, I'm focused on more optimal forms of recreation between churning out assignments. That's the ideal, at least; I frequently still fall short. But that's a hell of a lot better than waiting for the largely-unpredictable parental help requests. It's difficult to get anything done, so after all these years I've basically stopped trying. It's not a mindset I like, so naturally I try to minimize my time there.

Shifting registers isn't quick or easy. Part of the reason I went back to school early this semester was to begin the process early. All too often, during what non-engineers call "syllabus week", I'm still operating primarily in the home register than the school register. I slack off out of habit instead of clearing out easy assignments quickly and efficiently. I don't like that but there's only so much that BRUTE STRENGTH can do for it. Putting myself back in the correct habitspace is a much smarter tactic.

If I didn't get the idea across well-enough, well, maybe I just need to switch registers.

Predictions for 2018

The New Year is a time for reflection and self-improvement, and in that spirit I'm putting out my predictions for 2018. The rules are the same as they were in 2017. At the end of the year, I'll come back and see how accurate I was.

PERSONAL/ACADEMIC

I will read 15 books this year: 95%
....20 books: 80%
....30 books: 60%
I will write an average of two blog posts per month: 80%
....three blog posts per month: 70%
....four blog posts per month: 60%
I will not experience a major political/religious/philosophical conversion: 95%
I will write more than 10,000 words of fiction this year: 70%
I will still be using Facebook, Twitter, and Tumblr: 90%
I will remain single this year: 60%
I will vote in the 2018 midterm elections: 95%
I will not be hospitalized: 95%
I will not have my wisdom teeth removed: 90%
My French will improve this year: 60%
Jayhawk Rocket Design will begin constructing the hybrid rocket engine: 70%
I will graduate with my degree in aerospace engineering: 95%
....in May: 90%
....with a GPA 3.2 or above: 80%
KU Aerospace will place in at least one AIAA Design Competition: 90%
....two AIAA Design Competitions: 80%
I will no longer be living in Kansas: 60%
I will either find an engineering job or enroll in grad school by December: 95%

SCIENCE/SPACE

No successful human clones announced: 80%
No major progress on figuring out the "EM Drive": 70%
Falcon Heavy successfully launches this year: 80%
SpaceX Dragon 2 test-flown this year: 90%
....successfully: 80%
Boeing Starliner test-flown this year: 90%
....successfully: 80%
Astronauts will launch from American soil: 70%
Virgin Galactic tourist flights won't begin this year: 60%
Blue Origin begins crewed tests this year: 60%
No crewed flights beyond Low Earth Orbit: 90%
NASA Orion will remain on-schedule: 70%
No deaths in space this year: 90%
No country leaves the ISS Treaty: 90%

POLITICS/WORLD

Donald Trump will still be President at the end of 2018: 95%
Mike Pence will still be Vice President: 95%
Republicans will lose the House and/or Senate at midterms: 90%
Theresa May will still be Prime Minister of the United Kingdom: 90%
UK will remain on track to leave the European Union: 80%
No additional countries will leave the European Union: 70%
US Nominal GDP will finish ahead of the EU and China: 80%
Bitcoin will end the year valued $500 or above: 80%
....$1000 or above: 70%
....$10,000 or above: 60%
Fewer than 10 cases of wild polio will be contracted: 70%

META

I will remember to score these predictions: 95%
I will have been overconfident in these predictions: 80%
I will have been less overconfident than in 2017: 60%

This year I am making a total of fifty predictions, and attempting to be more conservative than last year. That said, I still expect to be overconfident. See you back in twelve months to see whether I was successful.

2017 Prediction Results

At the start of the year, I made sixty predictions about how 2017 would go. Now it's time to see just how badly I did. The predictions left unmarked were correct, those struck-through were incorrect. I'm only throwing out the Obamacare prediction, because no one can agree on what constitutes a repeal. It is marked in italics.

PERSONAL

I will still be living in Kansas: 95%
My parents will have moved into their new house: 90%
My parents will have sold our current house: 80%
I will still be annoyed about it: 70%
I will read 10 books this year: 90%
....12 books: 85%
....15 books: 75%
....20 books: 60%
I will write an average of one blog post per month: 80%
....two blog posts per month: 60%
I will not succeed in a longform fiction writing project this year: 85%
I will not get back together with my ex-girlfriend: 85%
I will not get a new girlfriend: 65%
I will not get a boyfriend: 95%
I will not experience a major political conversion this year: 80%
I will not experience a major religious conversion this year: 95%
I will be able to perform 15 push-ups with ease: 90%
....20 push-ups: 70%
....30 push-ups: 50%
I will not be hospitalized: 95%
I will not begin drinking regularly: 90%
I will not attend a solstice event: 80%
I will participate in the LessWrong community: 70%


ACADEMIC/PROFESSIONAL
Jayhawk Rocket Design will fly the hybrid engine on a rocket: 65%
Jayhawk Rocket Design begin work on a two-stage rocket: 70%
I will have had an engineering internship: 70%
I will have a job lined up post-graduation: 60%
I will still be studying aerospace engineering: 95%
....at the University of Kansas: 93%
....on track to graduate in May 2018: 90%
I will end the year with GPA 3.0 or above: 80%
Jayhawk Rocket Design will test fire our hybrid rocket engine: 80%
I will not be planning to attend grad school: 90%
I will learn Javascript this year: 80%
I will not learn Haskell this year: 80%
I will not begin learning Russian this year: 80%
My French will not improve this year: 75%

SCIENCE/SPACE

Human trials of CRISPR announced: 80%
No successful human clones announced: 65%
Self-driving cars licensed in the US or Canada: 50%
"EM Drive" will be explained by existing physics: 75%
KIC 8462852 will not be satisfactorily explained: 80%
No strong evidence of extra-terrestrial life will be found: 95%
SpaceX launches a reused rocket: 80%
Reused rocket is successful: 70%
Falcon Heavy will fly this year: 65%
No deaths in space this year: 90%
No country leaves the ISS Treaty: 85%

POLITICS/WORLD

Donald Trump will still be President at the end of 2017: 95%
Mike Pence will still be Vice President: 95%
Obamacare will not have been repealed: 60%
United States will not go to war with a nuclear power: 85%
United States will not enter a major new war (>100 US casualties): 70%
North Korean government will not be overthrown or displaced: 95%
North Korean government will not begin liberal reforms: 93%
Theresa May will still be Prime Minister of the United Kingdom: 90%
United Kingdom will not have triggered Article 50: 70%
Ukraine Conflict will not be resolved in 2017: 85%

META

I will remember to score these predictions: 95%
I will have been overconfident in these predictions: 80%

This last one sort of requires me to do the math first, so let's get to it. Spoiler: it won't be the tie-breaker question.

50% predictions: 0 right and 2 wrong, for a score of 0%
60% predictions: 2 right and 1 wrong, for a score of 67%
65% predictions: 2 right and 2 wrong, for a score of 50%
70% predictions: 4 right and 4 wrong, for a score of 50%
75% predictions: 2 right and 1 wrong, for a score of 67%
80% predictions: 10 right and 3 wrong, for a score of 77%
85% predictions: 6 right and 0 wrong, for a score of 100%
90% predictions: 8 right and 0 wrong, for a score of 100%
93% predictions: 2 right and 0 wrong, for a score of 100%
95% predictions: 10 right and 0 wrong, for a score of 100%

Graphed, my results look like this: 

What does this chart mean? Bear with me here, because this may get confusing.

The orange line represents perfect calibration i.e. 90% of the predictions made with 90% confidence are accurate. The closer to the line, the better calibrated you are as a predictor. Results above the line are underconfident, while results below the line are overconfident. My predictions made with high confidence were underconfident, while those made with somewhat lower confidence nevertheless proved to be overconfident.

As a general rule, underconfidence is preferable to overconfidence. In colloquial terms, it's better to be safe than sorry.

If we look at individual categories, it's clear that my worst performance was in Academic/Professional predictions, while my best was either Personal or Politics/World, depending on how you choose to count. Technically Meta was 100% correct, but that wasn't really so much of a prediction as an acknowledgement that January!Me had no idea what I was doing.

There's several issues I want to rectify before my 2018 predictions. Firstly, several of my sample sizes were too small. For example, there's only two predictions with 50% calibration. I got both of these wrong, but it's hard to say whether I was dramatically overconfident or just a bit unlucky. Meanwhile, most of my predictions were made with 85%, 90%, 93%, or 95% confidence. Arguably, I was essentially shooting fish in a barrel on these. It would be better to distribute my predictions more evenly.

Furthermore, it may not be a good idea to use even two significant figures for this kind of project. Throwing out the 50% predictions and lumping everything into decades (60% and 65% predictions together, 70% and 75% predictions together....) we get results looking more like this:

with the breakdown as follows:

60–69% predictions: 4 right and 3 wrong, for a score of 57%

70–79% predictions: 6 right and 5 wrong, for a score of 55%

80–89% predictions: 16 right and 2 wrong, for a score of 84%

90–100% predictions: 20 right and 0 wrong, for a score of 100%

This more seriously reflects the confidence gap between high probability outcomes (things more than 80% likely to happen) and possibilities with lower odds. Basically, I scored a lot of probably things too low, and a bunch of improbable things too high. It's an embarrassing sort of mistake to make, but figuring out where I'm under and overconfident is the entirely purpose of this exercise, so mission accomplished.

2018 predictions, hopefully reflecting these insights, coming soon.

November/December Links

A lot of people reacted poorly to a Washington Post story implying that the Trump administration banned the Center for Disease Control from using certain words in their documents. According to National Review and the New York Times, this misrepresents the actual, more complicated situation.

Speaking of healthcare, HIV/AIDS is looking to get a lot nastier in the near future. It's not really clear what we can do about it, beyond more of the same.

Something to do once I'm out of school: print my blog. Matt Shapiro argues that we really can't depend on platforms to continue providing hosting, forever, and having off-site backups is increasingly necessary if you want your works to survive.

In 1849, a Virginia slave named Henry Brown mailed himself to freedom in a wooden box. His escape was enabled by northern abolitionists, but also by confidential parcel services. To me, this underscores the importance of allowing free, private competitors to government agencies wherever and whenever possible.

Speaking of federal services: the origin of NORAD's Santa tracker.

Astronaut Bruce McCandless II dies at age 80.

AIAA reports of the progress of advanced spacecraft propulsion in 2017.

Elon Musk tweets photos of the assembled Falcon Heavy first stage.

Going Off On A Tangent

I just remembered an incident from middle school. As a middle school memory, it's unremarkable in that it shouldn't have happened in the first place, but is unusual in that some small bit of wisdom can be gleaned from it.

Back then I was what passed for a gifted child, and consequently enrolled in what passed for a gifted education class. During one of my projects—I don't remember which one—one of the teachers said something to the effect that it seemed like I was going off on a tangent. I didn't know what a tangent was, so I tried to explain myself using the entirely incorrect geometric metaphor. Something about legs of a triangle. That's what you get for teaching algebra before trig before calculus.

With the benefit of hindsight, I can explain myself better.

Whatever line of research I was trying to pursue only looked like a tangent because outside observers weren't looking at the whole picture. My plans were outrageously overambitious as a rule; teaching a preteen the concept of planning fallacy isn't necessarily easy. Nevertheless, I can't shake the feeling that observing the size of the plan is not the hard part of that task.

If we're going to make it through the 21st century, we've got to fix a lot of different forms of irrationality. Giving our "best" and "brightest" such shoddy training does not bode well. 

And Two Other Things

Further meditating on the problem of progress, I noticed a few other things which had been going right in the world and my personal life that contributed to the perception that things were moving a specified direction.

One of the most memorable things happened on the winter solstice of 2015: SpaceX, after many failures, successfully landed a rocket on the ground. It's worth noting that this technical triumph was underscored by multiple sorts of suffering in my surroundings. I'd literally fallen and hurt myself pretty badly that evening, and then been dragged to a very uninteresting Longest Night church service that ended mere moments before the landing. My attention was entirely directed elsewhere than the mundanity around me.

At this point, landing rockets has become essentially routine. I don't even watch launches anymore, though part of that is because I don't know when they're happening without Twitter. Maybe there's an email alert service for that.

Another thing which improved in the first half of 2016 began a week after the SpaceX landing. After an entire semester apart, I drove up to St. Paul to enter the new year with my then-girlfriend. It'd taken a long time to finally get together, and then we hadn't arranged to meet up again for the entire fall. But after that, we saw each other at least every three months until breaking up. This felt like emotional progress, getting to spend time with someone I loved on a regular if infrequent basis.

...there was going to be at least one other thing, but it's not coming back to me now. Hopefully this gets the point across: largely by accident, a lot of factors were lining up in what seemed like an organized pattern, only to diverge at a later date. Homo sapiens' overactive pattern-matching instincts strike again.

Try not to make that mistake in your own future. When things feel like they're going well, don't assume that they'll stay that way. Stuff can always go wrong, and—all other things being equal—probably will.

Hyperbolic Discounting Revisited

Related Reading: Picoprocrastination, Hyperbolic Discounting in Education

My, what a difference a year makes! As the months till graduation tick down, my time horizon has seemed to grow. It's not clear whether that's a product of the hard limit of my knowledge approaching, or my lived experience increasing.

Either way, my dissatisfaction with past selves has risen dramatically as 2017 progressed. At this point, I can even admit that my behavior—while not necessarily unpredictable given those initial conditions—was the primary reason I failed so dramatically during my years at Purdue. Four years doesn't seem like a particularly long time, not nearly as long as it felt five years ago, during my first semester of university.

Motivating myself with the prospect of being done with it (no, not like that) is now a viable strategy. It never worked before, and I found those kinds of pep talks somewhere between frustrating and completely incomprehensible. When I wrote about hyperbolic discounting in education, I was just barely starting to react to that line of reasoning, after a semester of struggling to make myself work through a workload that, in retrospect, was pretty good. I probably wouldn't have believed it then, but I'm actually nostalgic for Dr. D's thermodynamics class. It's unclear, however, how much of that perception stems from the broader situation.

Noticing progress can be a subtle art, but pleasurable nonetheless.