I had an idea. And I was convinced that it was the greatest idea ever idead.
In three words: Zachtronics… but ants!
Here’s the premise. In a colony of ants, a single ant has a very basic set of instructions that it follows. A finite set of “states” that it can be in and a few methods of interacting with other ants in the colony. But when you have thousands of ants following the same ruleset, something emerges that is greater than the sum of its parts. Something like a SuperOrganism, where the colony as a whole seems to have an intelligence. A Hivemind in the most literal sense.
One of the most common examples of this is the food collection process. When an ant finds food, it lays a trail of pheromones pointing to that food source, and very quickly, a large number of ants will be able to find and collect that food.
Here was the idea that I had. What if the player could write that simple ruleset that an ant follows. Just the exact if-statements and state_machines, that each ant would follow, and explore how that would impact the intelligence of the hivemind.
And a prototype emerged, where I got the basic food collection working.
A Promising Prototype
It was a prototype, so a lot of things are a little more complicated and wordy than what it would be in the final version, and it definitely felt like there was something there.
I shared it with a few other game devs, and there seemed to be consensus on a couple of things:
- The idea is promising.
- The block-based programming is limitting.
The next set of ideas was a long and knotted journey exploring a whole lot of different unique paradigms. There were some ideas that I think were pretty unique and pretty out there.
Bespoke Programming Paradigms
There was one which was a geometric approach, based on the excellent talk given by Brett Victor, Stop Drawing Dead Fish (clip of relevant excerpt).
Then there was another, where all information was encoded as a direction, so you would do checks that were all directional.
Two things from that video:
- It’s called Loomy 4, so it’s the 4th iteration of this interface.
- At the end, I said it was fairly easy to understand…
Needless to say, I had dived in very deep, head-very-much-first.
The deeper I got, however, the more the voice in my gut resisted. There was something off about it all, but I just couldn’t figure out what it was. So I kept making more prototypes, and exploring more paradigms.
A Moment to Reflect
Eventually, I spoke to Radu, my mentor from Astra, and he got me back on the right path.
The issue was that all of my design had been bottom-up. I liked the idea of programming ants, so I was figuring out new ways that programming could be done, just so that it could be used in this game about programming ants.
He asked me to take a top-down look as well. Why are we programming these ants? What are they going to be doing?
I made a list; and here’s some of the cooler ideas:
- An enemy clan of ants attacks us, so we move dirt around, and create a funnel for the enemy ants to come through, where we can hold our ground, and defeat superior numbers, like Gerard Butler and the other 299 Spartans held off the Persians.
- Wood has fallen into a deep pit, which we cannot climb down, so move some dirt around to create a canal, and then fill up the hole with water, and collect the wood once it floats to the top.
- In order to break a large piece of stone into smaller pieces, we move dirt around to create a small cliff, then throw the stone off the top so it falls down and breaks apart.
And a bunch more. It took some mental flexing, but there were a few ideas there.
A Moment to Realize
And that’s when I realised what my gut had been saying all along. All of those ideas, when broken down into mechanics, were just two things:
- Ants following trails
- Ants picking up and putting down things. Rocks to break, or dirt to terraform.
If all the ants are doing is going to be following trails and moving things, then what was the point of creating a programming game in the first place?
I guess I had misunderstood what was exciting about the idea and prototype. Maybe it was not the programming that had been cool. Maybe it was seeing the large number of ants impact change on the enviorment that had been interesting all along.
And it seems like all of that is built on two simple “mechanics”. So why not go ahead and make a different game, and just focus on those two mechanics.
So now, I have another idea. And I am convinced that it is the greatest idea ever idead.
In three words: Factorio… but ants!