So here we are, the post-mortem post for our first project. Well, I say “our”, but in reality Stormcall Software is pretty much a solo dev studio (at least for now.) I wanted to wait a couple of months to post this, but given that this has been a small project that, admittedly, didn’t see any major success I decided to get it done and out of the way today since this marks a month since the game’s most recent and likely final update.

When I decided to return to game development, I wanted to come out of the gates strong. I think it’s safe to say Double Antonyms failed to do that. While I’m not ashamed of what I’ve put together, I’m not exactly proud of it and I feel as though things could have gone much smoother.

Different Platforms

The original plan for Double Antonyms was to release three different versions of the game: a free web version that was playable in the web browser but with limited features, a paid version available for three main PC operating systems (Windows, Mac, Linux), and an ad-supported version available on mobile. This obviously did not come to fruition due to a number of difficulties along the way which could be resolved within the time constraints that I had set.

The Mac OS version was dropped mostly due to Apple’s developer policies, which are ludicrous, quite frankly. While most platforms seem to be doing everything in their power to make their platform attractive to developers and encourage development on their platform, Apple has taken a rather snobbish approach that conveys the message that we should be the ones grovelling to them for the privilege of developing for their platform. Linux development is obviously free, and developing on Windows is free too (unless you want to publish your game/app onto the Windows Store, for which Microsoft charges a modest, one-time fee.) Android charges a small fee to publish apps on Google Play.

And then there’s Apple.

For the privilege of developing on Mac or iOS, their royal excellency over at Apple demands us members of the peasantry must pay a sum of $99 USD per year.

And to that I have two things to say: Fuck off, and fuck you.

So obviously the Mac OS and iOS versions were off the agenda. Just four platforms now. The next hurdle was Android, and while I did successfully put together a working Android version, the original premise behind this version was to have it available for free but supported by ads. This meant having to implement AdMob, which did not go well. The truth is, the further I dived into Double Antonyms’s development post-release, the more I felt like it was an obstacle to future projects that I wanted to work on. I decided to just cut my losses on this one.

AI-Generated Content

This is probably the big elephant in the room. It’s not something I wanted to do, but it was done. Double Antonyms makes use of a large word list. I don’t remember how large the word list ended up being, but I believe it was around 20,000 words or so. Quite large. The thing is, when you deal with those kinds of word lists, you can’t just type out each individual word in a JSON file along with their definitions, synonyms, and antonyms - such a ridiculous task would take months.

My first approach to this problem was to use the WordNet word database and use a variety of hand-coded Node.js scripts to produce a JSON file containing each word’s definition, antonyms, a handful of example synonyms, and a bunch of other data that the game made use of behind the scenes. This word list was fine for testing, but for the final product it was not fit for purpose. Many words only had a very small number of antonyms listed, and it was just a very limited word list in general.

So, this led to me to go down the AI route. The thing is, most AI chat applications were either unable to produce the JSON file as requested, or simply flat-out refused to. The only two that would actually work were Grok and Google Gemini, and conversations in Grok could only go on for so long before it reaches the token limit. Google Gemini was the only option and it was… better. Far from perfect, but better. Hand-coded Node.js scripts and manual intervention was used in order to refine the word list a bit, but the result was still sub-optimal.

Final Thoughts

Overall, the development of Double Antonyms could have gone better. While the development process as a whole didn’t end up being too arduous, I found myself falling into certain pitfalls. The idea behind Double Antonyms, as mentioned previously, was to have a free version of the game available on mobile with ads, and the money from these ads would be used not only to support myself, but also to fund the next project. This obviously did not go as planned.

The v1.1.0 update will be the final update for this game, as I’m devoting all of my available project to my next project, which I’m hoping to release by early 2026.

In summary, Double Antonyms fell short of expectations. Development could have gone better, but it could have gone a lot worse too. Better than nothing.