Tristan Dunn

Old Game, New Technology

Building my 24 year old game for the modern web.

I was first attracted to web development in 1997 to build a skateboarding website on GeoCities. Using GeoCities ended up being the gateway to my career as a software engineer. Along the way though the real passion that helped me learn to program was building video games, specifically a role-playing game.

Most of high school I had a spare notebook that I’d write code in during class to bring home and transcribe into Visual Basic 6. Figuring out an optimal method for the client and server to communicate, working out how to calculate the experience needed for each level, or how monsters would fight back. All in my notebook.

Learning Visual Basic 6 lead to my first job as a student worker writing ASP. I had a lot of free time there which lead to me learning PHP, JavaScript, and CSS. At each phase I’d adapt and improve the game using the new technology I was learning. In 2009 I used the source code of my Ruby on Rails version of the game to apply at thoughtbot and it’s the reason I created my pusher-fake library almost ten years ago.

Over the years the game became a test bed more and more for me to learn about new technologies. While it’s been great for my career, the fun has faded over the years and it typically meant the internet never even saw the game. I want that to change that now, but want to explore the history a bit first.

The Past

The game has bounced around between different technologies over the years, but that also included different names, stories, and gameplay changes.

The Fallen Dragons

I believe my friend and I first called the game Aurora, but we couldn’t get a good domain for it and settled on The Fallen Dragons instead. The Fallen Dragons was a client-server game written in Visual Basic 6, inspired by Realms of Kaos. I learned all sorts of new concepts while writing a client and server including a lot about networking, since I was running the server on my computer at home.

Probably the most fun of all the games, since everything was new and exciting. I still remember the random person who downloaded our client and played off and on for a week. They said they were from Russia and asked us to make custom in-game items for them. That was a fun week. Other than that no one beyond a couple of close friends ever played the game.

Monster Massacre

Monster Massacre was my first web-based version, written in PHP. A point-and-click game where you had a stamina system that refreshed over time to limit how often you could fight. There were probably hundreds, if not thousands, of similar games at the time.

It had a slightly bigger player base, with friends of friends playing it over the span of a couple of months. This was my first exposure to web security, since my friends would often try to hack it. Such as discovering they could manually run the periodic script to replenish stamina that I left it exposed to the web. That was a fun discovery since they didn’t readily tell me and used it to level up faster.

The source code of the later versions are still sitting around in my backups. I laugh a little every time I go back and look at the horrible code, since they’re from around 2005.


I named the first Ruby on Rails version Miroha, which I have zero recollection of how I came up with the name. I registered the domain nearly 16 years ago and the name certainly sounds more legitimate than Monster Massacre.

I don’t recall if anyone ever played it other than a friend or two, but it’s been my main test bed for new technology since it started. Over the years I’ve used long-polling, experimented with server-sent events, wrote my own WebSockets server in Node.js, briefly used Pusher which spawned an open source library, tried out React, explored new CSS techniques like flexbox and Tailwind CSS, learned how to use Turbo Rails, and so much more.

The Future

While the old iterations have been fun, my goal for the future is to actively work on Miroha instead of it being a test bed for technology experiments. And I’d be incredibly excited if it’s playable online again, instead of locally.

Open Source

The first step is to start an open source version of Miroha. While this could lead to some accountability, it’ll at least avoid hiding the work, allow others to learn from it, and allow the less frequent experiments be a source of examples for other people exploring the same new technologies.


The second step is to deploy Miroha publicly on Heroku. This will allow people to actually play the game and it’ll make it easier for other people deploying their own versions as well. Being able to play with other people again might be the most exciting part, plus hopefully collecting feedback from them on how to improve the game.


Third is writing a story for the game, which is generally what makes or breaks a role-playing game. The players need to learn the history, become part of the story, and have it help drive how they play. This will be the hardest part of making the game, but has the added bonus of helping me improve my writing skills.


Fourth is documenting the process that goes into building a game. You could look at the changes to see what the result is, but how were product decisions made, what goes into planning the story, or how do you manage an open source game that anyone can suggest changes to. While it’ll be nice to share the learning experiences with others, having a record for myself will be great too.

The Present

While open sourcing the new version of the game is a start, these are pretty lofty goals. Even with trying to stay positive, I know all these future goals will be difficult to complete. At the time of writing I’ll consider this a success if the game is playable online and at least 10 people end up playing it. A rather small target, but it would have been an insane accomplishment to me in high school.