I Made Something

I believe, as I heard a mother say on This American Life this week, in doing the things you are capable of, so I've created a digital essay about how we teach information. Click this thing to see it:

How do we learn what's true?

Let me now use this space to write a bit about the essay's motivation.

Liberalism maintains that once we collect enough data, we will arrive at the correct models. But the increasing importance of narratives—and the consequent decrease in the importance in their underlying facts—has thrown this epistemology into doubt. Much contemporary writing seeks to recover a sense of reality from the soup of competing truths that defines postmodernity. Seeking to disclaim bias and reassert the regime of factuality, writers of all ideologies have begun to speak of argument as the act of staking out territory in an information war, proving that postmodernism is here to stay. As a crude example, here I’ve concatenated a paragraph from a right-wing conspiracy site with an opening passage from a mainstream liberal essayist:

The manipulation of facts and the slow relentless war on reality is being waged on this landscape of the mind. When those who seek to control humanity can convince the world that what they say is true, we will rapidly descend into the most oppressive tyranny ever seen.

Most of us can’t afford the luxury of investigating, because we have more pressing things to do: we have to go to work, take care of the kids, or look after elderly parents. Unfortunately, history does not give discounts. If the future of humanity is decided in your absence, because you are too busy feeding and clothing your kids, you and they will not be exempt from the consequences. This is unfair; but who said history was fair?

(Solution: The first paragraph is from InfoWars. The second is from the introduction to Yuval Noah Harari's 21 Lessons for the 21st Century.)

Media-literate readers can tell the authors’ ideologies apart by the change in shibboleths: the right’s prophecies of battle, chaos, and Armageddon become the left’s boring sympathy for unpaid domestic labor. But both agree on one thing: that only a dark era of epistemological chaos can follow the sunset of objectivity. Whether they tell the leftist story of identity- and class-based oppression or preach brass-tacks Evangelical nationalism, modern ideologies claim veracity by positioning themselves as rocks of intellectual certainty amid a tumult of postmodern confabulation. It’s not the correct ideology that wins the most followers, but the one with highest degree of narrative fluency—the one that knows who to talk to and what tone to take.

I wanted to do something about it.

Things You Can’t Teach Yourself

Remember that moment a couple of years ago when, all of a sudden, everyone was listening to podcasts?

Podcasts weren't a new technology; there was no PR campaign. Rather, it seems like they benefited from a lucky coincidence of forces: The election of a contentious president made everyone care about the news. The release of the Apple Podcasts app put an easy subscription tool in everyone's pocket (I don't miss middle school, when I subscribed to all the tech podcasts via RSS feed and painstakingly downloaded a fresh batch of .mp3s each week). Serial happened. Nowadays, all you have to do is say Quip electric toothbrush in a crowded urban space to hear a concert of groans.

I bet that online courses will be the new podcasts.

My evidence? My mom is now taking online courses, and it was around the time Mom started listening to podcasts that everyone else did.

I, too, am taking online courses. Something I've observed is that open-access, labor-of-love sites tend to surpass paid MOOCs. Paul's Online Math Notes, for example, is more legible and better paced than any of the other online calc courses. I'm in differential equations now, and the only reason I've paid to enroll in an online class is to issue myself quizzes; Paul doesn't have any diff EQ practice problems yet. Likewise, I did a course on data analysis using Python, and its best part was the link to An Introduction to Statistical Learning, a free machine-learning textbook that doesn't shy away from the algebra that Python packages are designed to help you shy away from. I've also been working on HTML and CSS (W3Schools) because I'm the web editor for Fulbright's Infusion litmag, which just released its first issue of the grant year. (This blog might also be due for a facelift … )

A second observation: STEMmy things like math and coding are easy to learn online, but the resources are more paltry if you want to work on the humanities or social sciences. Ironically, this might be due to humanists' high aesthetic standards. People who know Python usually know how to write HTML, too, so they start by tossing their notes online, then beautify them in stages. On the other hand, lettered-arts nerds love the feeling of beautiful typography and layout, but they lack the skills and confidence to code it themselves, so they rely on expensive print journals and paywalled websites to host their research—and then wonder why people call the humanities out of touch.

There's this thing called the digital humanities, which in theory is about getting academics to design websites and compile online databases. But the movement falls short by yielding to humanists' anxieties about code. Rather than take actual web-design courses, typical grad programs encourage students to attend one-shot DH workshops, where they learn WYSIWYG tools like Scalar (sorry, alma mater) and Wix. These are great ways to get your feet wet and explore the possibilities of web design, of course, but truly lit projects in DH usually combine solid underlying research with attentive, material-aware design well beyond Squarespace's capabilities. A favorite is the Slave Voyages Database, although I must admit I preferred its web-1.0 look to last month's AirSpace-informed redesign.

Another mistake of the DH movement is to reduce to the web to a content-delivery system—and miss its potential as a medium unto itself: I won't @ any individuals here, but while uploading PDFs of your journal articles to Scribd or Academia.edu is better than nothing, it probably doesn't count as digital humanities.

Lingua Franca

I’m back from Ulsan, where I spent my days studying at this spaceship of a public library:

In cultivating a resolution to learn Korean, I notice I’ve contracted an anxiety about being that American, the one who crosses the ocean only to socialize with other foreigners. The anxiety had me shun expat restaurants, abstain from the English section of the library, and direct my gaze sternly forward when passing tourists in the street.

I was overdoing it.

In Ulsan, I discovered an Indian halal restaurant hiding around the corner from my guesthouse. Intercultural metaphors abounded: the trilingual menu introduced samosas as deep-fried mandu; on the wall hung a pointillist treatment of Korea’s iconic autumn ginkgos. I asked the server if I should order in Korean or English or what. Korean, please, he said, and we chatted a little about how business was doing.

Lingua franca: an expression in Latin, meaning French, used most often to refer to English.

I originally studied Korean so I could communicate with, you know, Koreans. But I cherish the shared dysphasia that arises when speaking Korean as a bridge language. Each knows the other’s effort, knows that the conversation can only take place because we chose to make it possible.

Shoring Up Certainty

I am thinking about the word overbearing.

My computer’s desktop is an early-winter landscape, decaying .txt files strewn all over. As I sift through the leaves, I notice I often use this hedging device: when a particular behavior or phenomenon is upsetting me, rather than say outright that I am upset by that thing, I say that I’m ruminating on the meaning of the word that describes it.

It’s not the worst habit to have. I’d like to be the sort of reflective person who, before rushing to label someone as nosy or overbearing or whatever, thinks about the true meaning of those words and their appropriateness for the context.

But quibbling over definitions is also a way of privileging theory over practice. It lets me shore up certainty in the literal accuracy of my statements by baking uncertainty into their phrasing. You probably have met someone who talks like this: I’m not sure if annoyed is quite the right word, but I feel a kind of … annoyance about the way he—

Sometimes epistemological honesty comes at the cost of obnoxious phrasing.

My school's break started at the beginning of January and will run until the end of February, with a few odd work days in the middle. By the time this post goes up, I'll be several days into a three-week stay in Ulsan, where I'll continue holing myself up to work on writing and calculus.

I woke up in a cold sweat recently, having dreamed I was a interviewing for some generic marketing job when the manager asked, I see you taught English in Korea for two years—care to tell me some about the measurable outcomes of that work?

The next morning, I signed into LinkedIn and aggressively added new connections to abate my fears of never ever being hired. (The fears have remained at bay for several days—is that a measurable outcome?) Far be it from me to fish for pity, but there's something singularly hard to itemize about the ways I am growing, the things I am learning here. To choose a concrete example, it's not that hard for me to identify progress in the math I've been studying—I can point quite easily at the problems I can solve now that I couldn't in September. But I don't know how to go about "measuring" these outcomes in a way that would satisfy my nightmare interviewer. Ditto with Korean, although at least there's a formal certification test there (the TOPIK) that I'll take in a few months.

I'm not writing out of a sense of futility here. I truly want to know: How can self-educated people demonstrate proficiency in fields where there aren't certification exams or technical interviews? Do people who work in hiring intentionally seek these people, or is it more cost-effective simply to pursue candidates with traditional credentials? Email me your thoughts.


So Max, in English, virgin means an unmarried woman, right? asks my coteacher.

Sure. It means she's, you know, pure.


Well, I say, it's a little more specific than that.

Our office Catholic joins in: It means she's clean.


And free of sin. Well-mannered.

… Let's go with that.

My Snail

Speaking Korean, I want to say, Some people spend their life apologizing for who they are, but the literal translation seems cold and analytical, so I come up with a new symbol: Some people live like garden snails. Of course it's a different expression; it was bound to be different. Because I am myself, I get to approve my translations of myself, to assert that snails are apologetic insofar as they embody the sort of self-effacing person I'm referring to. To be understood—to know that the words passed through is the Korean idiom—comforts me far more than the obscure boast of a word-for-word translation.

But I meant to talk about vector functions. I'm into Calc Ⅲ now, and I keep noticing that the brainfeel I get talking around clunky phrases in Korean is very similar to the one I get when parameterizing an equation. I can't shake the feeling that math is just lyric poetry played out on graph paper, a rigorous way of plugging metaphors into each other. That contrived variable t isn't a part of the curve we are trying to draw; it merely metaphorizes—parameterizes—the stylus tracing out the shape we're interested in. The parameter is a figurative device.

Stoned mathematicians like to ask next: Is the function itself actual? What makes it more actual than its parameter? We might ask the same, in literature, of the breeze or babirusa likened to a human. A little change in perspective, and we realize that the metaphor is working in two directions at once, asserting as much about humans as about the nonhumans it personifies. We say that it's the parameter that animates the curve, but this is just a mind hack; it's equally true to say that the curve beckons its parameterization. Squiggles surely precede line integrals.

To speak pragmatically, the fluidity of metaphor is why I'm comfortable with translation styles that others might call too liberal. We can equate n-dimensional curves and their paramaterizations because we know that they are simply two ways of modeling a mathematical object whose existence was already only hypothetical. Language, too, does not terminate or close upon itself; whether translated or merely set in a different font face, it undergoes constant reinterpretation. Let me have my snail. Let me write my own concordance.

Bus Tickets Cost a Lot

Personal-finance gurus tell you to spreadsheet every single income and expense, tagging them all by category—because more data is better, right?

I, for one, can't be bothered. Instead, I've created a hella-low-maintenance spreadsheet in which I pay myself a weekly allowance of ₩100,000 ($90) and track my spending against my predicted account balance.

I only update my spreadsheet with three events:

  • Every month, I input my pay, which varies slightly because my school deducts my cafeteria lunches.
  • I make a note in another column every time I transfer money to the US.
  • Every half month, I note my account balance.

A function then computes my target account balance—the amount of money that would be in my account if I spent my weekly ₩100,000 allowance and nothing more.

Because my school pays directly for my housing and most food, the only nonnegotiable expenses that have to come out of this allowance are my phone plan (₩46,000 a month) and my transportation (maybe ₩15,000 a week). A true spreadsheet fetishist would log these as well and reduce the weekly allowance accordingly. But such scrupulous data entry would consume time I'd rather spend writing pithy blog posts.

From my semimonthly account balances, I subtract the target balance for the corresponding days to produce the following graph.

Here's how you read it: If I spend exactly ₩100,000 a week, then my account balance will parallel my target balance and the graph will flatline. If I stay under budget, the blue line goes up; if I overspend, it goes down. I'm tracking the difference between my target and actual balances, rather than their absolute values, to mitigate the spending impulse that comes from watching my account balance spike when my monthly pay comes in.

Even if the graph falls below zero, I'm not immediately losing money; I'm just saving less. Conceptually speaking, you can view as the graph of my additional savings on top of what I already save by implementing the allowance in the first place. In other words, it's showing the current balance of my allowance account.

You can see that I remained pretty frugal during my first few weeks here, then started to spend more around mid-October. In fact, I made two weekend trips to Seoul, and bus tickets and hotels cost a lot. But my position is still north of zero.

For what it's worth, the trend line shown, a least-squares regression courtesy of Google Sheets, is somewhat meaningless. This graph isn't predictive in nature, unless you're one of those in a deterministic universe, there's no such thing as free will zealots. Instead, I use the graph as a psychological gambit: For every day that I don't spend money, my target balance still declines by ₩14,000, meaning my margin above zero increases by the same. It makes me feel like I'm constantly getting paid.