My faith in humanity has been restored

Some-faith-in-humanity-has-been-restored

I don’t know how, I don’t know when, but I feel better now. Not sure for how long I’ll feel better, but certainly I feel somewhat more confident of my abilities now.

That said…

LameBoy

LameBoy debuggerIt’s going okay. Right now it’s clocking at about 5.5k lines of code, so evidently still in its infancy, but we are making progress fairly quickly. I had to break everything so that I could add a layer of abstraction. It’s just a matter of cleaning up Denton’s crappy code and future-proofing it.

World War Spades

Apparently everyone hates the idea now, so the best coders in the block (LeCom, ByteBit, Chameleon, and I) are meeting up and making a voxel shooter in D. It’s a nice compromise between C# and C++ although I’ve personally never used it for anything larger than “Hello World”. Thus it is now…

AoS of the Future

Probably a more imaginative name will be produced in the future… but hey, it’ll have planes, and hopefully backwards compatible with 0.75 servers 🙂 We’re still waiting for ByteBit to finish the renderer (he’s making good progress).. then we’ll be ready to add the other subsystems!

Disparity

We don’t… talk about Disparity.

LilithPort

Still waiting for the spectate reports to arrive. No dice trying to fix the code.

Type Wild Release 3

Unfortunately some BG animations broke tinkering with some things (seriously I never touched that stuff, how’d that happen? :-?), so now when I have time I need to go back to the old versions and find the working sprite sequence and see if I can replicate it.

“Final project” – to be planned soon

I believe this will be some sort of an AI project. I must do this if I want to go to a nice college. Even if I don’t take this to a science fair, it would be great to show off. If it is good enough, it might take me somewhere, I don’t know. But regardless, it must be something nobody has ever done before, or a continuation of something someone else has done.

I have an idea for an AI model. It’s still in my head and needs to be put on paper before I can make something, but it goes along the lines of this (please note that I’ve never read a textbook on AI in my life, but I do plan to, so consider this entirely hypothetical (and incredibly incorrect) in nature):

There are two memory regions: one with data and the other one regarding the metadata of regions of that data. There are three default subregions of data: in, general, and out. Metadata has a few fields:

  • Data being addressed.
  • Life – once the life is 00, the connection will die.
  • Lifespan – maximum possible life of the data.
  • Address of data to be connected with.
  • Weight of connection.

So how is this supposed to generate any activity at all? Good question. The answer is stimulus. No input, no output. The second answer is learning. The most basic type of learning is repetition, I would say.

Say you input “Hello world”. So the input subregion would contain:

[‘H’ -> ‘e’ -> ‘l’ -> ‘l’ -> ‘o’ -> ‘ ‘ -> ‘w’ -> ‘o’ -> ‘r’ -> ‘l’ -> ‘d’ … …]

with a default life of 255 and lifespan of 255. The standard input bytes are connected in sequence to differentiate between extraneous data. Without any connections, the data is forgotten and should be replaced with zeroes after a couple of ticks.

Now here’s the problem: how does the AI know where to put stuff and how? If I wanted to move this byte sequence onto the general subregion, would it be done by default? After how many repetitions would it find itself there? How could I tell the AI to drive a connection into output? There’s no buffer, it’s all a stream of data.

Which shows you how much more research I still need in order to figure this out.

Well, that’s enough for progress. See ya.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.