This evening I found a few tweets from Jonathan Blow that disappointed me a bit. Here they are:
I wrote code to do A* today, for the first time since I was in college. Long time!
— Jonathan Blow (@Jonathan_Blow) April 6, 2013
In other news, this puts me pretty close to the state of the art in video game AI research…
— Jonathan Blow (@Jonathan_Blow) April 6, 2013
This has driven me to finally start doing something I meant to begin a long time ago. I’d really like your help, too!
My first thought when reading Jonathan’s tweets were that he meant one of two things – AI in games generally, and AI for NPCs (which is what people often mean when they say ‘game AI’). For the former case, well, there are too many examples of great, vibrant, exciting research to mention. Even in the latter case, research into pathfinding and planning has made huge surges in recent years, with Monte Carlo Tree Search cracking every search domain known to man, and even Jonathan’s plain old A* having new life squeezed out of it – by researchers working on AAA games, no less.[1]
However, it’s easy to point the finger. I am upset by Jonathan’s tweets – because I believe very passionately in building bridges between scientists and independent developers, moreso than ties with the mainstream industry. But instead of moaning, I’m going to try and do something about it. As I tweeted myself:
Perhaps Jonathan Blow’s comments reflect a failing on our side to communicate our work though. Maybe we can change that!
— Michael Cook (@mtrc) April 6, 2013
What can I do? Something I wanted to do ages ago! Every fortnight from now on, on a Saturday, I will post The Saturday Paper – a blog post where I highlight a cool paper from or related to videogame development, with a summary written by me that cuts through the density of the paper itself. I’ll try and provide pictures and video where possible, and links to the paper if I can gain permission (or if it’s not open access already – many conferences, including CIG and AIIDE already are!)
What do I need for help? For one thing I need criticism. It’s been a while since I wrote regularly, and I’m new at science communication. I’ll value any feedback or comments you can give me – feed them through to the usual address, mike@gamesbyangelina.org. The other thing you can do, of course, is spread the word – Twitter, Facebook, Reddit and more. Show gamedevs, get the knowledge spread, and maybe next time an indie mentions game research on Twitter it’ll be because she’s just adapted it to make her awesome game even better.
- [1]I should probably say – I did say I was learning about science communication! – that A* is a very popular algorithm used to find the shortest route between two points. As you might imagine, it’s used a lot in videogames for getting NPCs places, among other things!↩
This is an excellent idea! Perhaps it may draw some of the more inquisitive minds to look further into the field of artificial intelligence, too.
Thank you! And I definitely hope that can be one of the side effects.
This is fantastic! I really can’t wait to start seeing what you post!
A good idea. Though I would not take JBlow’s tweet too seriously, the man has a tendency to be glib and uncharitable with his opinions.