Agents Discussion

moopli wrote:

You have hit the nail on the head.

Especially with your final comment -- this is exactly what I want to see happen. Evolving specificity of pheromones, venoms hijacking signalling pathways, the works.

Sounds beautiful. In a few days I might try to mock-up a quick demo of this mostly as an exercise for myself, but I'll post it here for general amusement.

~scio wrote:

...how many bits in the efficacy mask match the mechanism mask...

I just realised that there's a subtlety here we missed earlier -- it isn't about matching the mask. The mask simply makes sure only a subset of the bits in either the resistance or the toxin are actually important to this toxin. By and-ing, we make sure we ignore a bunch of data, which is how we reduce the coupling between toxins.

If we were to try and match the mask, then all the zeros in the mask would be important too, meaning every bit matters to every toxin, meaning we haven't reduced coupling at all. So it's a subtle distinction but an important one.

moopli wrote:

~scio wrote:

...how many bits in the efficacy mask match the mechanism mask...

I just realised that there's a subtlety here we missed earlier -- it isn't about matching the mask. The mask simply makes sure only a subset of the bits in either the resistance or the toxin are actually important to this toxin. By and-ing, we make sure we ignore a bunch of data, which is how we reduce the coupling between toxins.

If we were to try and match the mask, then all the zeros in the mask would be important too, meaning every bit matters to every toxin, meaning we haven't reduced coupling at all. So it's a subtle distinction but an important one.

So, if I understand you right, the agent's toxin bitmask just looks at a small piece of the overall resisitance bitmask?
If I understand you correctly, in order to ensure that there is no obvious winning resistance bitmask for any environment, the toxin bitmasks would have to overlap in some way.