My Superpower

In high school, I was an in­sa­tiable con­sumer of sci­ence fic­tion. In hind­sight, it is un­be­liev­able how much time I spent read­ing in­stead of do­ing the other things usu­ally con­sid­ered to be typ­i­cal for teenagers (going out, do­ing home­work, etc.). I was the clas­sic sci­ence fic­tion nerd. My live re­volved around the words of Asimov, Weir, Clarke, and prob­a­bly every other sci­ence fic­tion au­thor you could name.

It was not un­til rel­a­tively late in my aca­d­e­mic ca­reer when I re­al­ized my thirst for the writ­ten word had given me a su­per­power. All of those years slurp­ing up sci­ence fic­tion had ac­ci­den­tally given me an un­usual abil­ity (in this day and age). An abil­ity that, once upon a time, had been com­mon­place.

Fortunately, this abil­ity can be trained quite quickly. What had taken me years to learn by ac­ci­dent can be de­vel­oped in just a few weeks (or days, if you’re com­mit­ted).

This blog post will serve as doc­u­men­ta­tion of:

1. What this abil­ity is.
2. How you can de­velop it.
3. How it can be ap­plied to the day-to-day work of pro­fes­sional soft­ware en­gi­neers.

The Ability to Read

In the in­ter­est of avoid­ing click­bait, I’ll just come out and say it plainly: this “superpower” is the abil­ity to read full books from start to fin­ish.

Sounds bor­ing, right?

I call it a su­per­power be­cause it is be­com­ing rarer by the day. Fewer pro­fes­sion­als read any­thing longer than a short blog post or ar­ti­cle to­day than at any point in the last cen­tury. I won’t get into why, since there is al­ready an abun­dance of doc­u­men­ta­tion ex­plain­ing why peo­ple have been read­ing less. But in or­der for some­thing to be a su­per­power, it needs to be more than just rare. Why should read­ing count as one?

Reading qual­i­fies be­cause it al­lows one to do some­thing that is oth­er­wise ex­tremely hard: un­der­stand a topic or idea at a level ap­proach­ing an ex­pert.

When I was in school, I re­mem­ber sit­ting in lec­tures sur­rounded by stu­dents who hated our cours­es’ text­books. My col­leagues called the in­fer­nal blocks of pa­per “boring”, “repetitive”, and “heavy”. Indeed, they were, but they also con­tained most of the in­for­ma­tion the pro­fes­sor would lec­ture about, and pre­sented it in a for­mat I could con­sume in a frac­tion of the time.

I re­al­ized that I could save my­self hours of te­dious home­work and lec­ture time by sim­ply read­ing ahead in a course’s text­book.

Now that I’ve spent a de­cent chunk of my life re­moved from my stud­ies, I’ve come to re­al­ize that this same phi­los­o­phy can be ap­plied pro­fes­sion­ally in our work as en­gi­neers. Hun­dreds of hours of in­tel­lec­tual floun­der­ing can be saved with just a lit­tle bit of read­ing.

Depth Provides Certainty

To il­lus­trate why the kind of deep un­der­stand­ing that can be ob­tained through lit­er­a­ture is rel­e­vant, I’ll bring an ex­am­ple from my own re­cent work.

For con­text, I work on Harper, a gram­mar checker for those who care for their pri­vacy. We’ve re­cently been work­ing to ex­pand our of­fer­ing to com­pete with Grammarly Premium. This is a com­plex, open-ended prob­lem with no “right” an­swer. I can imag­ine two ap­proaches to it, one good and an­other bad.

I was pretty well versed in the kinds of of­fer­ings that Grammarly pro­vides, and it was pretty ob­vi­ous how many of them work. Gram­marly takes some LLM provider with an API (like OpenAI or Anthropic), pro­vides a bunch of style guides as con­text and wraps it all in a pretty UI and sells that at a pre­mium. We could do that same thing, right?

In the bad case, I could just copy what Grammarly is al­ready do­ing and call it day. Good idea, right?

I did­n’t think so, so I sat down and read through some of the prime-time lit­er­a­ture com­ing out of labs writ­ten by ex­perts in nat­ural lan­guage and the arts. They pointed me in a com­pletely dif­fer­ent di­rec­tion.

By read­ing deeply, I was able to come to the re­al­iza­tion that many of the fea­tures that Grammarly hides be­hind their pay­wall are rel­a­tively sim­ple to im­ple­ment, if you know what you’re do­ing. Not only that, but com­bined with the ex­ist­ing in­fra­struc­ture Harper al­ready owns, we could do it cheaply with­out dis­re­spect­ing our cus­tomer’s pri­vacy or copy­right. In other words, we could dif­fer­en­ti­ate our­selves from the com­pe­ti­tion by tack­ling the prob­lem deeply and ap­proach­ing it from a unique an­gle.

This story has hap­pened over and over again, and I can imag­ine that it ap­plies to most of the peo­ple work­ing as en­gi­neers to­day. A big open-ended prob­lem needs to be solved. Rather than ap­proach it in the ob­vi­ous man­ner, you can learn from the ex­perts your­self and pro­duce some­thing bet­ter as a re­sult of their in­sight.

How to Develop the Skill of Reading

On oc­ca­sion, I will find my­self in a dry spell. I will go sev­eral weeks or even months with­out read­ing any­thing of much sig­nif­i­cance. In do­ing so, I will grad­u­ally lose my abil­ity to read with depth for ex­tended pe­ri­ods of time. My mind’s mus­cle for grap­pling the tough ideas pro­posed in the words of ex­perts will weaken.

When that hap­pens, here are two things that I do to beef it back up.

Schedule a Block of Time for Depth

Often, the biggest bar­ri­ers to depth are the sim­ple dis­trac­tions of life. You have to pick up the kids, cook din­ner, or go shop­ping. There is al­ways some­thing on the docket that gets in the way.

My so­lu­tion: sched­ule a ses­sion of deep read­ing so that it looks like any other chore on your cal­en­dar. If I don’t do this and, in­stead, try to fit it in­side the gaps, I find that my mind will come up with ex­cuses for do­ing any­thing other than read.

My mind says, “But Elijah, you need to do your taxes right now. You should do that in­stead, be­cause read­ing is bor­ing.” Your mind might say some­thing else. I hap­pen to en­joy do­ing my taxes.

Schedule time for deep read­ing and you’ll be able to say “no” to that pro­cras­ti­na­tion mon­key in your head.

Find a Topic That Is of Immediate Relevance

I love my job, and I get a huge thrill when I find out a way that I can do it bet­ter. So if I need to find some­thing to read, I of­ten start by ask­ing my­self, “what is the hard­est part of my job right now?”

The an­swer can be:

• A spe­cific prob­lem I’m try­ing to solve (see the sec­tion above ti­tled “Depth Provides Certainty”).
• A repet­i­tive process that I do reg­u­larly.
• Something I am un­sure about which is dis­cour­ag­ing me from work­ing with gusto.

From there, I plainly ask ChatGPT: “What is a good book that cov­ers X in depth for a per­son with Y level of pre­ex­ist­ing knowl­edge of the sub­ject.”

To de­velop the skill of read­ing deeply, your book does­n’t need to be about a sub­ject re­lated to your job. That just hap­pens to work well for me.

Just Start

A jour­ney of a thou­sand miles be­gins with a sin­gle step. A book of a thou­sand pages be­gins with a sin­gle para­graph. The best thing you can do is start.

So, I im­plore you. If you’re stuck on a prob­lem or an idea, try read­ing a book about it. I have per­son­ally found an im­mea­sur­able amount of mean­ing through lit­er­a­ture, and it’s un­blocked me on a num­ber of prob­lems in my pro­fes­sional and per­sonal life. I sus­pect it can do the same for you.