The Past, Present, and Possible Future of Magic Duel
I do not believe in hard work when it comes to programming. I am doing hard work for MagicDuel, for the last few years, almost daily, and it’s persisting and improving, but that’s only “maintenance” work. The beauty and results of the programming required for such “unstable” systems, such as; MagicDuel requires a different type of programming/programmer, a bit of a crazy mind like mine I dare to say. It requires that connection between “out of programming world” and code, I do not consider myself mainly a programmer, programming is just a tool, I think that tells us everything.
Even if it isn’t successful, after, I will experiment with this a bit and see what “damage” it can do, I will release a part of the basic concepts for others to study and improve, but if it works, well you might not hear about this again 😀 but you will also not be able to tell if Muratus del Mur is a bot reading your mind or just a busy programmer that pays little attention to what you’re saying.
I honestly hope to find people out there that have the required mindset and will to work on these types of experiments, but unfortunately, even if 60% of you have great ideas, less than 0.01% have what I need for this particular experiment, so there’s no way I could share the workload with someone about this. All I can do is to tell you about it and challenge you to read endless posts like this one 🙂
Yeah, I’m not sober right now, but I do love you all and I can think straight, partially at least 🙂
and if I was sober I would never tell you about my crazy experiments at all.
My first experiment of this kind was a neural net that did something very similar, it could find synonyms to ideas based on what it learned from previous inputs. It was a fun thing to do, about 3 days of hard work, but not much excitement at the end of it when I realised that neural nets are far more complex and I can’t reach their full use because of server limits and of course because I know no programming language that could be used for this, and neural nets for php are more like a slow monstrosity. Also it was a disappointment because I knew all the info came from previously learned patterns. I wanted something MORE. I started to imagine computing systems that use much more than on/off bits, and have more states to flavour a complex thinking, but it was still no use, I have no technical way to put that into practice and I have no advance knowledge in microchips either, so it was fun thinking but futile. Then I realised I was looking the wrong way. I couldn’t see the forest because of the trees. Too much complex programming to try to mimic natural algorithms is pointless, the solution is much simpler if you look at it from a different perspective.