> From: [email protected] > To: [email protected] > Subject: RE: [agi] Couple thoughts > Date: Fri, 27 Feb 2015 08:04:27 -0500 > > > -----Original Message----- > > From: Nanograte Knowledge Technologies via AGI [mailto:[email protected]] > > > > To me, the internet's issue would be issues of bandwidth and delay and > > routing (latency) and so on, and off course availability and robustness > and all > > such stuff. Bad choice overall, in my view. Skype apparently uses the > spare > > resources on personal computers well, and probably other webapps too, but > > not for this level of computing. > > > > Keep in mind "the internet" isn't only a collection of homogeneous nodes > separated by high latency low bandwidth connections it's very non-homogenous > with many super low latency clusters of high performance computing regions. > Adaptive systems can also adapt to their computational resource topologies. > Software that can run on different substrates, or at least be ported, > usually live longer lifetimes. > > But I won't try to sell on that I see you have a requirement for a GPU > centric enabling parallelism. > > > Thank you John. The technology might change again. On has to continuously > > sniff around for the best-fit solutions architecture.
> > The "problem" I would like to address is how to connect a 3D printer to > the > > wormhole in our universe, via an adaptive, learning communication service. > > > > OK. That's either dreaming or very ambitious or both :) I think, the exponential, technology advances happening now is set to continue. Every new step increases the viability of such a dream becoming realized. I suppose, it is very ambitious, but I would not be surprised if at least 30 other universities are working on a similar concept. As soon as the super gene is uncovered - which I believe it would be - it would open the door for the realization of this particular step in a a much, bigger dream. I may not be a part of such pioneering work, but I could probably die happy knowing that at least, I had one, original thought. Still on my quest though. I sense I'm getting there, but how would one ever be able to verify such a thing? Moot point. I'll even be okay learning that no original thought was possible, which would support my theory of irrelevance. The journey would be worth it. Worse-case scenario? It was helluva lotta fun! > > > Fractals? Yes, the knowledge methodologies employed, to date, are fractal > > based, each component being a pure object, some recursive, others self > > recursive. Each one represents as a system of systems on their own, and > then > > collectively at a quantum level. I have taken upon myself the job to > integrate > > these. It would seem to be my area of expertise. > > > > I used to use pieces of Fractint code for various things back in the DOS > days that was really the last time I've worked with fractals in depth. How > does the system you are thinking of work? Does it inject fractal to learn > from complexity? Is it like fractal compression? Fractal compression could well be utilized at a holographic level, whether logically, or as physical output. I don't know yet. Various applications seem likely. For now, the core, integration methodology is rooted in fractal theory, thus N scalability. All other schemata and/or supportive methodologies and methods would have to conform to this requirement. In essence then, one would not be amiss to say that the whole system enjoyed the benefit of a symbiotic, fractal architecture. Randomness would be a key objective for research to resolve. If the system cannot embrace true randomness, it would be quite useless. Thus, at the logical-systems level, one has to consider any method, which would be pattern-of-pattern friendly. Recently, I conducted an armchair experiment to assess the reliability of a name-brand random generator, which claims to use cosmic noise as random, input stream. Still, using a basic algorithm, I discovered a few, consistent patterns in all the patterns that on-line system generated, according to my specifications. I was thinking for some, bright mathematician to flow this algorithm over a Markov chain, to see if we could not detect the "pattern of one", as a symbol of effective complexity, which I have the notion might exist. I'll keep that one aside for the team and an appropriate, research environment. The conceptual movie already plays in my mind. But logically it's a different story altogether. At some stage, I'm quite certain of it, I'll fall off the edge, due to my own ignorance and inherent fallibility. Further, my resources and effort has been considerable, so one can only drive as far as one has fuel for. It's a highly-conservative approach then, which is a good incentive for me not to waste any resources on wild-goose chases. By then, my hope is we'd have a dedicated team working on this, with unshakable funding in place, so my weaknesses would find support with them and the organizational structure, and so on. To my reckoning, we have about 3 years left to conclude this work and have a working prototype ready. I just think that the shit is going to hit the fan after that, and the increased freedom to do research, almost at will, would become severely impaired. So, dreamer, or not. Someone has to slog on with it. Already, I noticed some benefit of that flowing into this community. It was my pertinent choice to come share here, not coincidence. If we all did that, we'd get there a lot sooner, or just in time. The magic continues... Rob > > John > > > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/26941503-0abb15dc > Modify Your Subscription: https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
