Finally shipping

A funny thing happens when you finally ship software. Recently I’ve been writing about the process of getting Trailblazers Trivia off the ground. My friend Nathan and I have been tweaking designs, trying to get the pieces together, and working to move the app across the finish line.

Well now that it’s out—and you can download it here, would love some feedback!—it feels like there’s a weight off our shoulders. 

That thing that was hovering in the back of my mind, that idea that we were so near the finish line—but never crossed over, has finally lifted. 

All the early decisions to cut features and ruthlessly identify an MVP have paid off. The app is in the wild and available. It’s shipped. 

Is it perfect? No. But is it something to be proud of? Well yes, for me it is. It represents taking an idea and pushing it to the finish line. Also, it represents realizing my limitations and knowing when to look for help. 

There’s a saying in software, that by the time you ship something you should already be slightly embarrassed by it, because of your improved plans and ideas and progress on the next iteration. 

That’s very much true here. We released a tiny trivia app that is playable, and in my mind a lot of fun. But it’s limited in what it does. 

The main use case right now is for a teacher to lead a trivia game with a group of kids in two teams, with the teacher facilitating whether each team in turn answered correctly. 

That’s a little confusing when someone first steps into the app. There's not a lot of context on that intended use case. Also, I’ve removed all the explainer information in order to focus on just getting the idea out—so all the onboarding concepts I’ve thought through are stripped away for now. 

But that’s ok.

At this stage it was important for us to release the idea, prove that we could do it, then iterate and improve on it moving forward. 

And here's the fun part. Now that it's out, we can take a deep breath and continue building. There's no longer the pressing concern of its existence, now there's questions about improvement and adding in delight.

So over the weekend I worked on the next step—a single player mode.¹ In Friday’s post I sketched out the general idea for it. Then, with the help of Claude I created the feature one step at a time.²

It was slow going.

I had a lot of setbacks, not knowing for sure if the changes I made were the right ones. But, with each step I tested the app, making sure that the functionality I needed was working, looking for issues in the interface. Slowly I was able to build out the single player functionality. The trick was to only change one or two small things at a time, asking Claude to move an element—update the logic behind a button, or modify the size of a headline. Bit by bit the app shifted. 

This could probably have gone much faster if I understood all the code. But I really wanted to see if it’d be possible to add on a whole new feature. Now, the real test will come when Nathan reviews my code. I’m a bit worried about that part. At one point I did a cleanup and removed a bunch of—what seemed to be—extraneous code that had been added throughout the session.  

So why didn’t I just leave the development for my partner to finish out the piece? I definitely could have, and I know he’d have completed it in less absolute time than it took me. 

But part of this whole experiment of building this app is seeing how much each of us can do, how far we can go, and how quickly we can iterate. With the foundation in place I’m now looking to see what else can be added, how we can improve for delight, what it will take to make the product truly great. 

Each of us can do so much across the surface of the app, and instead of just resigning myself to the design part, I’m trying to see if I can dig into the experience of building. An app like this only works if it’s fun, if it feels right, if it’s a delight to experience. I can’t quite get a sense of that through my sketches and Figma designs. I need to test it, play with it, move things around. And it really seems like we’re approaching a time where product minded folks can do so much more, and that’s got me excited. 

So if this process works, we'll be able to divide and conquer. There's a dozen more features I want to build, and being able to design out part of them, while designing and coding other parts, will be the real unlock. I'll keep sharing as we progress, this is entirely new ground for us.

1. And that's so exciting. We released a two player mode. But now we can do more, we can add on features, we can tweak and play with ideas. It's a whole playground that now exists to be built upon. Also, in order to get single player mode ready we needed wrong answers alongside the correct ones. Well, let me tell you. Coming up with ~1,000 wrong answers alongside the right ones is harder than it sounds. There's only so many mountains in ancient Israel that I can remember.
2. Funny thing about Xcode. First, I love that you can now use Claude and ChatGPT directly in the app. You can use natural language to get help and directly change code. That’s amazing. But, for some reason it keeps reverting to ChatGPT, even though I prefer Claude. At one point I was wrestling back and forth with the LLM, arguing a point for several minutes, only to realize it had again defaulted to ChatGPT. I switched it to Claude and on the next try the problem was resolved. Both seem sufficient for coding, but Claude is just better at this stage for what I’m needing to do.