The problem with being a perpetual learner in a rapidly changing world is that you are always keenly aware of how much you don't know. You don't know exactly what you're missing, but you know that there is a world of knowledge out there, and that you have only scratched its surface.

Four years ago, on my first day in a brand new web development program, we were told to deploy a new instance of "Ghost" on a Digital Ocean droplet, register a new domain, set up the DNS configuration, and make a blog post and presentation about the whole thing.

It took us a week.

I just did the same thing in two hours... and a non-zero amount of that time that was waiting for the DNS settings to fully propagate so that the SSL certificate from LetsEncrypt could automatically confirm the settings, which I had to do manually as recently as four months ago.

Nevertheless, I am frustrated.

Because what I actually wanted to do was deploy to a CloudFactory, but I got stuck on the (probably) final step, and the server kept crashing during the application health check. Something to do with the PORT environment variable, I suspect, but I was unable to fix it before other things arrived needing my attention.

This is my back up plan: deploy Ghost on DigitalOcean (which is now super-easy) and take a bit longer figuring out the problem with the CloudFactory configuration. Document the learning process.

Everything keeps getting easier, and more powerful, and more amazing. Which is great.

But/and: the instructions for rapidly developing technologies are almost always already out of date. This means that when you try to do something for the first time, you are cobbling it all together with a combination of instructions, dense-text documentation, and (let's face it) Google/Stack Overflow for every error that comes up while you are following said instructions. (1)


Ira Glass named a problem with starting out down a new path: our taste exceeds our capacity. He calls it, "The Gap." He was talking about creatives, but in my experience it can apply to anything.

We know what good work looks like, but we lack the skills to replicate it.

For those of us who insist upon continuous exploration of new ideas and fields, it can be a nearly constant state of being, this sense that true competence is ever just beyond our fingertips.


The problem with being a perpetual learner in a rapidly changing world is that you are always keenly aware of how much you don't know. You don't know exactly what you're missing, but you know that there is a world of knowledge out there, and that you have only scratched its surface.

Deciding what not to know is the hardest thing for the likes of me. The deliberate choice to ignore glimmering, enticing knowledge is painful. What if the answer is right over there?

Yet at the point that you discover that something exists, somebody else is already in the process of building its replacement. Choosing to be at the cutting edge of anything necessarily demands leaving something else out. This is precisely what I have to do to become skilled with creating truly elegant solutions.

In some other part of the world that I also care about, I will have to remain a member of the audience. Getting good at security requires me to give up typography, even though I once spent several months learning hand lettering. I have to choose a programming language (and framework) and stick with it, even if something sexier comes along. (2)

Competence at learning, integration, and confronting the limits of our knowledge is a skill in itself, and perhaps it is one that we would do well to recognize in ourselves. Yet it is a meta-skill... useful, perhaps an essential underpinning, but a pretty weird way to answer that ubiquitous question: "And what do you do?" (3) And so, I am closing gaps. The taste gap, the skills gap, the confidence gap, and above all, the gap between where I am and where I would like to be.

THE GAP by Ira Glass from Daniel Sax on Vimeo

  1. 137, by the way, is an unidentified server error. Those are the worst.
  2. Something sexier always comes along. Which is good, because who wants to be stuck programming in COBOL in 2018?
  3. For the record, "I teach physics," also doesn't work well in most social settings.