I’m good with ‘global mind’ attached (search engine interface to web as my extended brain) but isolated, I don’t fare well.
I’m good at rewriting questions so long as I have a ‘gist’ notion of what a term within an ontology _might_ have a physical or visual analogy to which allows me to often help folks who are “stuck” in a quandary due to being lost in the ontology+theory that they’re operating within.
This is why I can find answers for students ‘stuck’ in theoretical physics or mathematical theories, as both utilize a modified first-order logic, which commutes to any other, and the solution is often shifting to a “fixed point” type of answer.
If it isn’t, a function-only solution can get them out of the dilemma.
But usually I can help because I’m good at finding answers in their language even if I can’t read it or produce it very well myself.
I know that when folks talk about transhumanism, they USUALLY have no idea that we’ve already arrived.
Ah, yes. I think the key to that will be having a proper notion of cognitive development as the first waypoints likely do not resemble human intelligence but rather a characteristic that all life shares.
Affordances. The environment provides the degrees of freedom allotted to the actor, which in turn, through its constraints, guides the actor towards its goal choosing but does not make the choice for the actor, as the actor will have many, just constrained.
I don’t think goal seeking is action minimization. That is good for efficiency, but efficiency is a property of physical systems. Living systems operate on effectiveness rather than efficiency.
I don’t know. The concept of fitness always seemed a little ad hoc to me.
or rather “just so stories”
Here’s me complaining about stuff I want to see from back in 2015. It’s based upon cellular automata. Silicon substrate, carbon nanofibers, copper etc. Each molecule of paint would be a CA but it would specialize for function as needed. (networking, touch, display, etc) It would stay flexible but the top and bottom layers would be “plain old silicon” for weather protection. So, that’s what popped in my head when you mentioned the colors so you have context. “They’ll happen. I’ll be dead but idc – It’s like computer paint. I have the plans and know what materials will work to make it happen but our technology has at least 80-120 years to go for it to happen. Basically, paint with embedded self-organizing computer ‘stuff’, including display, ccharged by whatever is around [people stepping on it if it’s the sidewalk, air… sound… and the display is human touch sensitive. and connected to everything. That way our computers are painted on every surface. Easy. But, we’re not there yet.”
I’m likely Asperger’s / high functioning autism, although not sure. A famous autism researcher claimed that autistic folks lack ToM – the ability to imagine how another is thinking/feeling. That never seemed right to me But later I found something that showed that it’s possible that autistic folks just form ToM differently – through rational processes. Slower, less ‘automatic’ but it gets them there. I think for your conundrum, you might find inspiration in looking at various findings.
How I do a ToM is complicated and situational. I’m always learning. Forming theories, sticking them to my ontologies, discarding. Creating LOTS and lots of rules that I could enumerate if asked.
I don’t form stereotypes in my mind very easily. But I do have LOTS of rules. Lots and lots of rules.
I’m on board with Danko. I know his theory is correct.
If you want a ToM that works for AI, learning how different humans do it can help you. People are both anonymous to me and also individual. I don’t know what somebody is thinking by looking at them. But I need to form a notion in order to communicate safely and effectively. I think it maps.
Distinguishing anonymous from anonymous:
a) ping vs internal metric. If ping > internal metric, there is an “other”.
Now you have location and distance and a channel potentially open.
b) How do you unlock other anonymous?
A key. Or are you the lock? Ok. Key-lock must be the same mechanism. I’m sure that exists in code form.
c) Is getting unlocked / unlocking another safe? How do you establish trust between anonymous agents?
That’s where I think it gets complicated.
You know it’s a message by its header or by a recurring pattern. There are several of those. Radio communication has schemas for that mapped out.
Ok. For that you need built-in rules. GO/NO-GO conditions.
The automata have their own internal clocks so long as they exist. That clock is DISTINCT from system clocks.
ID is birth timestamp from origin system * timealive, which is unique for every automata.
if the automata is running at a different clock rate than the host system it’s on, it won’t matter as its clock rate is internal and goes with it so long as it’s running.
Well, the anonymity is public but it has a private identity that it carries with them and can be masked or ignored.
Automata won’t share or look at the identity, but simply HAVING identity could be useful on a per transaction basis.
Hey stranger, can I call you Phil”? sure
“Can you perform X operation?” : perform system check if X is in set(system resources), send “YES’
First four are simple. Focusing on process message is good. What kinds of messages (just some examples) might it be expected to process
CASE should work for that. If it’s a limited number, you can step through. Or if it’s organized, you can run through in more efficient ways.
How much processing time does it have? It sounds like you’re operating in “long ping” ranges. (1000ms and under)
(fyi – that was missing the last part: this is the full). https://www.youtube.com/watch?v=xIuBmOufbls
If I was working with uncertain total computation time, I might seek out CA with LONG ping times.
That allows both of us time to process, compensating for deficiencies in underlying systems.
I’m starting to think that REPEAT communication between two CA nodes might be impossible (if you’re designing as such). So, computation time can be ‘stolen’ via long ping time.
https://www.youtube.com/watch?v=hiHP0N-jMx this is recommended among gamers and lag. Game programmers must take all lag into account, and for a “simultaneous experience”, a few tricks take place for this illusion.
So, in some games, long ping is good. Gives you an advantage due to compensation algorithms. Other games, low lag is better.
As gamers send very few instructions, coordination isn’t difficult in theory. But in reality, it’s tricky.