Practical Civic Tech

Get stupider, idiot, and other tips and tricks

The problem: almost nothing built at civic hackathons is actually used. This wastes the time and excitement of technologists, and keeps abundance-minded political actors from getting great tools. So why is this, and what can we (you, and me) do about it?

Update: Here’s part 2 of this post, with my list of civic hacks to work on:

Civic Tech Project List

Maximum New York’s practical theory of civic tech

My suggestions boil down to this:

• Build things that don’t require anyone’s permission to create or distribute

• Have a governmental-mechanics guide

• Build for a specific person who has asked for your project

• Get stupider, idiot

I also have these bonus recommendations:

• Don’t forget the legislature

• Ditch the skeptics

• Have a good time

• I’ll help you

Here’s a deeper look at each of these points and more.

Permissionlessness

There are a ton of tools and projects that can move the ball forward in politics. But you can slow down your feedback loops, your deployment schedule, and your enthusiasm, by building the wrong kind of thing.

First of all, build in a permissionless fashion. This means: you don’t need to ask anyone to build your project, and you don’t need proprietary data you can’t access. If you need some special government dataset that you won’t get for months (if at all), you’re exiting the “hacking” realm and entering the “very long project to manage and track” realm.

Second of all, deploy in a permissionless fashion. This means: build something that can be used as soon as it’s done. Do not build something that will need someone else’s permission before rollout; practically, this mostly means “don’t build something that has to go through government procurement.”

But how do you know which things can be built and deployed without permission? How do you know what will need procurement? How do you know what is useful to build?

Get a governmental mechanics guide

You need to have at least one good politics friend, who understands government, to help you scope your project in many ways. They can also introduce you to other people in the political realm who can tell you what tools they need.

If you don’t do this, the odds are much higher that: (1) you’ll build something no one needs, based on faulty assumptions, (2) you’ll avoid building things people actually need, and (3) you’ll have incorrect ideas about what’s possible.

Let me give you a simple example: an acquaintance wanted to build a tool that reviewed all the transcripts from the New York City Council. So they googled “NYC Council transcripts” and got this page as one of the top hits. It says you can buy a transcript for one City Council meeting for $20 from the New York City Clerk. The acquaintance, upon seeing this (and not knowing much about the Council) figured the project was too expensive to do casually.

But you can also get all transcripts for free from Legistar, albeit it’s not obvious how. Suddenly their project is doable again, and free!

Technologists need to marry their technical expertise with someone else’s govmech expertise. But not just to save time and cost, and this leads me to one point I cannot emphasize enough:

Build something for someone specific who has asked for it.

If you don’t do these things, your project is likely headed for the “won’t be used” pile for any number of reasons.

Get stupider, idiot

Most people know zero about anything. Stop assuming people (in or outside of government) know things. No one knows anything. You can drive change and empower political actors by surfacing basic data with great graphics.

For example: everyone wants to discuss education policy, but no one knows what the Department of Education is, or the shape of its budget. Everyone wants to discuss housing policy, but none of them can tell you how many housing units NYC has.

You do not have to build a “world changing” app to help your city. You do not need to discover a heretofore unknown trend in data. You do not need to educate the whole populace. You just need to make a cool graph and write a nice essay to go with it.

“But!” you protest, “my graph is only about a basic thing, and everyone already knows that!”

Look. I teach classes on NYC gov and law. I’m telling you. No one knows anything.¹ We can always use better visualizations on even the most basic things, whatever you think those are.

If you want to build an app of any kind, it can be simple and also effective. Complexity will probably kill it.

Don’t forget the legislature

A lot of civic tech efforts focus on the administrative state—your departments of education, sanitation, environment, health, etc. New York City’s civic tech efforts over-focus on them, in my opinion.

NYC’s City Council gets far too little attention from civic technologists, even though it could desperately use it (as could political operators concerned with the Council). Look at their public interface and weep.

Legistar, baby! Intro.nyc is a great example of civic tech that focuses on the Council.

Ditch the skeptics

Many people within the governmental/political system are jaded, and don’t think much more is possible than the status quo. Get away from these people.

My attitude: the only laws are physics and the ones people enforce with guns. Other than that, you can do whatever you want. Almost nothing has been tried. Everything is technically possible, it’s just a matter of discovering the path to each possibility.

If you bring me a civic tech idea and I don’t personally see an application for it, I will say something like, “Well I wouldn’t use it, and here’s why, but here’s how I would go about finding out who might use it, you never know.”

Two years ago I started telling people I wanted to start a civics school. The near-universal response from people inside the political system was dismissal. They thought the idea was naive—who on earth would actually take classes on local government?

Here’s the deal: they didn’t actually know that. It was a made up constraint about the world built on their own internal pessimism. They wouldn’t take a class they envisioned to be boring, because they think local government is boring. These are all incorrect assumptions, and they were wrong. For the record: it’s not like they tried to build a school and failed. They just armchair-declared it impossible.

Don’t listen to these people, and seek out genuine govmech people who can help you find the path to possibility, even if they can’t get you all the way there.

Have a good time

Building something just because it’s a good time is enough of a reason to do it. The point of this essay is to give you tips on how to build something that will be used productively. But hacking is an act of interest, play, and pleasant exertion. It is a good in and of itself, just like exercise. And these kinds of hacks can often result in something usable that no one expected!

Having a good time is a tactical advantage, and combining efforts with govmech experts can compound that advantage.

I’ll help you

Part 2 of this essay (coming in a few days) will be my massive list of civic hack projects I’d like to see, and would personally use and deploy. If you want to build something that will actually get used, I can give you that guarantee.

I’m also more than happy to discuss whatever civic hacking project you’re already building, or want to build. I might not have all the answers you need, but I also might! And I’ll certainly be able to point you in other good directions.

The Maximum New York alumni pool also has many talented technologists within it, and I know they’d love to work with you too.

Just email (daniel@maximumnewyork.com) or DM me (here on Substack, or Twitter).

Thank you to Andrew, Liam, and Priya for discussing these things at upsetting length with me.

1. I am not pessimistic about this. Why? Three reasons: no one knows anything, and yet we still haven’t fallen into the ocean! Our institutions might be in need of polish, but they are robust. Second: this is a massive opportunity. Whoever can meaningfully solve civic education will radically alter our political ecology by injecting many knowledgeable people into it! Third: this is not literally true, although it is relatively true. Of course there are many people who understand many parts of NYC’s government—but there are far fewer of them than you think. FAR fewer.