October 11, 2021

New Essay and Thoughts on Clickfarm

New Essay: How to Solve the Sudoku Puzzle with programming

New essay up! How to Solve the Sudoku Puzzle with programming. This one is a little conceptual, so you might want to read it before the rest of this newsletter. Spoilers below!

...

...

...

Spoilerz

My goal for this piece was "Clickfarm content that gradually goes off the rails." As some of you know, I really into clickbait and content farm-style writing. I wanted to capture that style while also subverting it and writing something genuinely interesting.

"By why are you so into content farms, Hillel?" Good question! For one, I am a madman. Two, while any individual clickfarm piece is boring, as a whole they're incredibly fascinating. Clickfarms are written for SEO, not humans. They're written by harried freelancers at a rate of 5-10 dollars per thousand words in industries they know barely anything about. You think tech has a lot of it? I just got an email by an SEO expert asking me to share an article on cockroach extermination.

To make mininum wage off of content farming, you need to write tens of thousands of words a day. That kind of volume is impossible under conventional writing, so the constraint leads to a very different and unnatural style. And that's what's so amazing: there is a distinct clickfarm writing style. That's why you see things like

• Lots of buzzwords dropped in for SEO purposes, even if they don't fit right
• No depth: the piece is nearly-indistinguishable from any other content farm piece on the same topic. Outside sources are only cited when they're easy to parse by an outsider (because oftentimes it's an outsider writing the piece).
• No authorial personality. Personality doesn't scale. There might be "personal" touches, but only the kind of personality that can be easily like systemized, memes or emoji.¹
• Poor copy: low level mistakes, copypastes, factual errors about the content, even plagiarism. The piece follows a cookie cutter narrative, either internally (listicles) or externally (the template can be reused for other articles.)
• No intra-community jargon (unless it's known at the corporate level, like "Agile"). Like they'll never call it "rails" or "javascript", it'll always write "Ruby on Rails" and "JavaScript".

This is my favorite example. Notice how the piece is about being a "Full Stack Java Developer" but not once says anything about Java or its ecosystem! And while that's a little unusual for a clickfarm, it's not at all surprising! You are familiar with the genre's tropes.

Clickfarm is a writing genre that emerges naturally from the weird constraints, and I love that.

But genres are meant to played with, and nobody's doing that; who the hell chooses to write clickfarm if they have other options?

Me.

So, "How to solve the sudoku puzzle with programming". Not a piece I expect to be popular, but one I wanted to write, dangit. But I didn't want to just throw garbage out there, it had to be good clickfarm. Which is a contradiction, since clickfarm is always, always about prioritizing quantity over quality. Instead I tried to subverted it by having the boring clickfarm whiplash into something genuinely interesting. Sudoku solving makes that really easy: there's a couple of clickfarmy ways to solve it, and a ton of articles on those ways. But there's also the DIMACS approach, which looks ridiculous and unhelpful before suddenly whipping around and trivializing the problem.

(Also it one-ups all those people who solve Sudoku with a constraint solver. I'm solving Sudoku with constraint ASSEMBLY. Suck it, nerds!)

My first draft captured the clickfarm style a little too well. None of my test readers couldn't finish it. Even though they knew the twist was coming, they couldn't actually get there. I needed to make the intro more entertaining without breaking the illusion that this was "actually" clickfarm. I wanted the reader to never be 100% sure this was a parody. Instead of making the intro something they laughed with, it needed to be something they laughed at. Something so bad it's good, which any fan of the Lyttle Lytton know is hard to do deliberately.

Oh well, draft two! It leaned a lot more into the badness of content farm, and people were actually able to reach the twist. A sample of that draft:

Remember: a boolean is a data structure that's true or false! 2 is not a boolean!!!

It took me a couple of days to realize the problem. Instead of sounding like a clickfarm, it now sounded like an overenthusiastic bootcamp grad. Turns out there's a really fine line between the two: newbies also write low-quality pieces... because they're newbies! While I was comfortable making fun of clickfarms, I didn't want to "punch down" at beginners. That just felt mean-spirited.

My fix was to make it more soulless. Beginners make mistakes out of inexperience, clickfarms make mistakes out of apathy. I went through and stripped out all of the personality, which was a problem because that's where I had all of the jokes. Draft three was about making the soullessness entertaining to read: inopportune copy-pastes, painfully obvious content fillers, dumb mistakes about technology, etc. I think I managed? It's hard to tell.

I'm satisfied with the piece now, but I don't think I stuck the landing. There's just too many forces here: I wanted something obviously content farm, obviously not bootcamp, entertaining to read, but entirely in a "laughing-at" way, without the joke being too obvious. Ultimately I think I made it too self-aware, so most people will quickly see through the joke. But I also know the punchline, I have no idea how people will take it.

To be honest I don't expect much of the wider internet to read it. The title is real, real bad. Authentic, but bad. Most people, quite reasonably, will skip right past it. But any sort of clever title would prime the reader to expect cleverness, which would ruin the clickfarm style. Oh well, sometimes you gotta make sacrifices for your craft!

(Since this is a tech newsletter and not a writing newsletter, I'm gonna do another, more tech-oriented piece later this week.)

If you're reading this on the web, you can subscribe here. Updates are once a week. My main website is here.

My new book, Logic for Programmers, is now in early access! Get it here.