There’s no doubt that coding agents can make an individual developer more productive, at least in the initial stage of actually creating the code. However, I’ve got bad news for all the software development organizations out there. AI isn’t going to make you more productive. To do that, you need to improve your process.

Why is that? Let me explain.

I’m a proponent of the #NoEstimates movement. One of the central premises is that developer speed does not determine when you’ll be done. Your organization’s process does. Instead of relying on estimates you should look at the historical data and then use forecasting to determine when a set of features (i.e. a release), will be done. For more information see the No Estimates book.

On of the first programs I ever ran on a computer was a text adventure game, also known as Interactive Fiction. I think the first one I played was Adventureland by Scott Adams, which was based on the first ever text adventure called Adventure by Crowther and Woods. Adventureland was the first text adventure available for personal computers.

Not long after that I discovered Zork I, the first game by Infocom. I loved the Infocom games, I played most of them, spending many hours solving the games.

Since February of 2026, I’ve been publishing a series of videos on implementing Swish, a Clojure-like Lisp in Swift using Claude Code. You can find it here.

Swish Logo

I used Common Lisp for a solid year in college, and really enjoyed it. I always wanted to use it professionally, but never really had the chance.

Featured in Clojure Deref on December 30th, 2025

In my previous article, I showed how to embed a S7 Scheme interpreter in an iOS app. This time, I will show you how to embed a ClojureScript interpreter, or at least a dialect of it.

Clojure itself, runs on the JVM, and there’s not really a way to embed the JVM in an iOS app. Maybe once swift-java gets rolling.

There is GraalVM, which lets you compile java code to native code, but it doesn’t support compiling for iOS. Babashka, which is a native Clojure dialect interpreter uses GraalVM.

I am a fan of Lisp. I was first attracted to it because it was the original language of artificial intelligence because of its ability to manipulate symbols and its own code. I spent a whole year using Lisp in a series of courses on artificial intelligence in college. I became quite adept at using it and quite enjoyed using it as well.

My Obsession

Lately, I’ve become kind of obsessed with being able to use Lisp in my iOS apps. I want to put a Lisp interpreter in my app and program my business logic in Lisp. Every once in an awhile I get the urge to figure it out, try, and fail. This keeps happening.

This is part 6 of my blog series on the history of my app Numerology. See Part 5 here.

When we left off, Numerology was chugging along, slowly making less and less each month. First $300, then $200, sometimes as low as $100. Sometimes it would spurt back up to $300.

This continued until fairly recently. In 2024, I noticed a competitor that always showed up first in the App Store search results when you searched for ‘numerology’. I decided to look into them a little more. According to Sensor Tower, their app was making over $5,000/month on the Apple App Store. They had an Android app as well, and it was making $5,000/month as well.

This is part 5 of my blog series on the history of my app Numerology. See Part 4 here.

I was probably the first numerology app on the App Store. If you searched for numerology, my app would be first in the results. At some point this was no longer true. A competitor came out and my sales started to go down. Today, in 2025, there are possibly over a 100 numerology apps on the store.

This is part 4 of my blog series on the history of my app Numerology. See Part 3 here.

The First App Store Version

In 2008, Apple announced the App Store and I applied for an Apple Developer membership ($99/year), so I could sell apps on the App Store. Everyone else did the same and Apple was swamped. Despite applying as soon as possible, I didn’t get approved for about 6 months.

This is part 3 of my blog series on the history of my app Numerology. See Part 2 here.

The Birth of Cocoa

In 2001, Apple released the first version of Mac OS X 10.0, the first desktop version of their new OS based on NEXTStep. All the great APIs from NeXTStep (hence the NS prefixes to all the class names and such) together were called Cocoa.

Now I finally had my chance to use the wonderful platform I’d been wanting to use for almost a decade. My full-time job was still with Java and enterprise world and it was increasingly not fun. I spent my off-time learning Cocoa. My first app was not Numerology though. It was an app called PhoneWord that found all the words you could make from a phone number. For example, from the phone number 555-5299 you could get 555-JAZZ.

This is part 2 of my blog series on the history of my app Numerology. See Part 1 here.

The First Version

It was the very early 90’s. Steve Jobs was at NeXT. I had some model of Macintosh II running on a 680x0 processor. I also had a compiler for C with object-oriented extensions (basically a subset of C++), called THINK C, with a class library called TCL, or the Think Class Library. For more about THINK C, see this article.