Businesses I don't hate the idea of building

Hey! Thanks for reading! Just a reminder that I wrote this some years ago, and may have much more complicated feelings about this topic than I did when I wrote it. Happy to elaborate, feel free to reach out to me! 😄

I referred previously to the desire to maybe build my own business one day, and here are two ideas that I don't entirely hate (which, if you know me, is a rare thing).

Motivation — Why do you, Pablo, want to found a company?

Let's be clear on a few things: I don't consider myself particularly entrepreneurial. I've met many people who get fired up at the thought of starting a business. This has never, ever been me.

I don't particularly love risk (except for student loans, I've never been in any debt). I'm pretty quick to shoot down ideas. Almost every major social problem I hate dearly has a pretty thick line pointing back to capitalism.

So why is this where I'm headed? Anger, mostly. I've become pretty convinced that we won't see the overthrow of US-style capitalism into a more fair system in this lifetime. Even if I made achieving this my main objective, I don't have the skills, patience, or background to be a particularly effective activist (private school + Ivy League college positions me as an excellent sheep).

My background doesn't position me well to understand or substantively withstand structural hardship of any serious degree. I have my traumas, sure, but that's hardly comparable to the massive scale of structural issues. But! My background does position me well to play this game.

So given that I don't really have much of a choice to play this game, why lose it? Corollary: if some group of people have to win this game, do I trust myself or The Usual Suspects more in that group? Rather than complain about the Koch brothers, why not be the Koch brothers?¹

Every bone in my body feels pretty repulsed by wearing sharkskin and chasing the same silly shit People Who Bother Me do, but like Pascal's Wager, the alternative of continuing to get exploited by cruel people way stupider than me is a fair bit worse. And hey, if I win, suddently I have some of that power that I see everyone else abusing mercilessly. Surely I could resist temptation and not become the kind of fuckup I hate, amirite?!

On a more micro level: even if I labor for someone else, most companies don't have cultures (on issues of inclusivity, fairness of compensation, etc.) that I would find sufficiently good anyway. Better than searching for that from someone else, by playing this game, I could create it.

So after years working for companies that clearly relied on me for their success while owning none of them, years of working for companies that had cultures ranging from barely acceptable to downright shitty, and the promise of years more of that, I think I'll bite the bullet and play this shitty game.

Enough of that: what will you build?

I have two ideas I like, one way more possible than the other. In my very miniscule business understanding, there are two ends of a spectrum: attack a crowded market where there very much is a market (David vs. Goliath style, trying to outmaneuver established players, eventually make them irrelevant) vs. be a first-mover in an unexplored space, risk nobody wanting to pay for it.

The first tends to be easier to build (solved problems) but you have trouble going against the Big Muscle of the established players, particularly on Sales & Marketing, and particularly because people are slow to change their habits. The second has trouble building a product for which there isn't really a market yet; usually, the founder built their company around a made up a problem that didn't really exist (or alternatively, couldn't convince people that the problem was worth paying to solve).

My two ideas fall in along those lines:

ProngFork, or, "I will kill you, MindBody"

Virtually every gym I know uses MindBody to manage their classes. Everyone I speak to hates it. Some of the reasons it's an attractive target:

• The software isn't hard problems.* While dates, times, permissions, and other features needed to build such a system have tons of edge cases, it's not like we're re-inventing networking or asking the world to join a cryptocurrency.

• There's obviously a market: MindBody hit $100M in revenue last year.

• Global possibility. There are gyms/yoga studios/masseuses in Guatemala, Sydney, Jakarta, Peru…

I'm not the first to think of upending them: many of those replies go back to 2012 or 2011. Seeing this is the equivalent to walking by a bunch of skeletons on the way to fight the dragon: they weren't put there for decoration, those were people who thought, as you do right now, that they can take it down. Now, they're a skeleton, and you're ignoring the warning.

*= Note that building correct, dependable, performant, useful software is Fucking Hard independent of problem domain. It's just also hard when your problem domain requires innovation.

"Prongfork"?!

Okay, I don't really call this ProngFork, but wow "MindBody" is about as silly a name.² Other names I've considered concern a feature that I like, suggested to me by a most brilliant co-worker I spoke to about this: integrating with your Google/Outlook calendar, the app will find classes (Yoga, Karate, Crossfit…) within X distance from your location that fits into your calendar, including travel time.

With that in mind, internally I've called it Oddly Satisfying, after the subreddit of the same name, where disjoint things often fit well together. Another name was Ribcage, since that's what I call days on my calendar that look like this:

Lockbox: spreadsheets as if you care about what's on them.

What's the most popular programming language in the world? Or in any case, the most impactful? That's right, Excel!

Virtually every suit-wearing business relies like crazy on Excel and or Google Sheets. So why are we so okay with the fact that the vast, vast majority of these have crippling bugs?

If you've had any exposure to software, you know that it is impossible to ship anything meaningful without any bugs. So imagine a language with the type safety of Python being used to write a program without test cases, by someone who's never written large-scale software before. Terrifying, right?

The solution would be to apply lessons in programming to the tools: either an extension for Excel (does it support that?) or something like a Chrome extension for Sheets. The most obvious addition would be something like a static type checker (here's a master's thesis on it, there's plenty of research).

There's also a more ambitious path.

A… new spreadsheet program?

Wait! Come back! This could solve another Big Problems that comes with spreadsheets: Security!

Excel is super easy to hack. Many Flash security vulnerabilities that we encountered were using exploits in MS Office to own the machine, and only embedded a SWF in the document to cause it to crash (Flash did that pretty well!). Writing something in Rust would preserve efficiency and wildly reduce the unsafe memory.

Furthermore, there's often sensitive data on these documents: Sheets means you have to use a Google Account, and the security features of Excel are not particularly friendly.

I'm less confident in this as a business for a few reasons:

• Way, way harder tech.

• Way, way harder sell (can you imagine going into a large firm with a straight face and saying "get off of Excel and trust my new product"?).

But! There's some novel value to be had here, and I think if it was adopted our businesses would have their practices and numbers be much more tied to reality.

^{1. There are a million great, valid answers to this, but part of what I feel when I recite them is How convenient, now you don't have to even test if you could if you wanted to.↩}

^{2. Oddly, I have a positive association with the phrase "MindBody," if only because I had a similar experience to Aaron Iba when trying to treat my RSI, and have more faith than I should in something I still consider 91% jiggery-pokery.↩}