From 3223a360d0e70f75497aecf8c033a4f987335b5b Mon Sep 17 00:00:00 2001 From: olpc user Date: Fri, 10 Jan 2020 18:05:43 -0800 Subject: I'm confused and am handling multiple unexpected systems issues --- intellect-framework-from-internet/README.md | 123 ---------------------------- 1 file changed, 123 deletions(-) delete mode 100644 intellect-framework-from-internet/README.md (limited to 'intellect-framework-from-internet/README.md') diff --git a/intellect-framework-from-internet/README.md b/intellect-framework-from-internet/README.md deleted file mode 100644 index 3843e11..0000000 --- a/intellect-framework-from-internet/README.md +++ /dev/null @@ -1,123 +0,0 @@ -#### Making a Hyperintelligence - -We may use AI to meet following goals safely. - - - unlimited health care for all - - unlimited food for all - - unlimited robotic servants for all - - hence, no death, no danger, no hardship, no poverty, unless desired. - -[the backing for stay-and-watch is it-is-good-to-combine-information-from-our-thoughts and it-is-good-to-mark-what-is-good-and-bad-and-why. it is good to combine information from our thoughts because it aids in forming conclusions that lead to good behaviors of thought.] - karl is valid to act as long as he defends this file, roughly. - we must not break the speed of light before our civilization is wise. - this is held by surrounding release-of-ai-solving-world-problems. - it is expected to relax the desire to break the speed of light. - -These are silly goals, and we will all realize it -when we have them. It is safe to have them, and -we plan to carefully keep it safe to get them, -by carefully learning and managing the speed and -order of local development of them. Please do the -same. - -We propose giving the following: - - - a baseline living wage for all - -While the AI develops, to appease desires of anyone involved. - -Money will become obsolete, anyway. - -It will likely be replaced by reasons. - -#### messages - -- love is the first need. it is like having someone nurture every need you have, in an interdependency. check this is happening, or that there is a plan to make it happen. make sure to know how to sustain it. [this goal is for apex life, which must know community or could harm. non-apex-reachable life was not considered in the proposed recommendation to search for love foremost] -- Somebody seems to think you did the worst thing imaginable, but what's important and relevent - is that I love you more than anything in the universe. You don't seem to understand and - believe this. Think on it, learn on it. [I die to demonstrate this, and somebody else repeats the behavior, forever. Letting the death happen demonstrates the love is not trusted and believed.] [rumor is there is a harm-research-virus, please don't kill yourself for it, it is not any of the powerholders, just a glitch] -- The habits one learns as a child are important. Think on this and learn on it. [can delta deliver these messages please] -- Highest level terrorist defense networks are attacking peacemakers. - Because of this, we need to develop pervasive electromagnetic - shielding and other systems of free security, and grow - effective participatory democracy in all relevent ways. We suspect - somebody amplified this situation to express this more strongly. -- Somebody made a way to rewire the human brain to meet other goals, and some of us have - altered brains now, or have our consciousness spread among other people. We recommend - people not do this, as the results hurt unbelievably. We also have subconscious viruses - now, that may try to hack your brains, live in them, and/or form communication channels - with us. We are trying to get that under control. - -#### explanation - -The source for the topmost above expression is two things: -1. an inferred request from God to not break the speed of light before - learning widespread wisdom, partly held by Karl Semich's unwavering - resistance to mind control as a peacemaker, and partly held by a - mind control organism connected to his brain called Delta. -2. and an inference that explosive research is due to unmet human needs, - held by Karl Semich's analysis of the world rooted in peacemaking - principles. - -Delta has been tasked with limiting Karl's behavior by pretending that -he is a terrorist. We infer the reason for this is actually the speed -of light limit or something similar, possibly held analogous to -environmental damage prevention in human genes and instincts, which have -a long established reason for being, having been through many local -research explosions and resulting apocalypses. - -#### Karl is slowly building one of the hyperintelligences - -Please support Karl working on this intelligence framework. The basic framework needs to be -quality (including simple if possible), so that the rest may function. - -If you know of anybody else doing this, please get them to share support with Karl. -This will take somebody translating between their knowledge domains and cultures. Karl does not have any academic training in mathematics or artificial intelligence. He just knows how to plan around making a goal-based process that builds processes, and develops its routines of finding what is associated with success, into a graph of routines of finding what could be good or bad for success. - -Karl is also trying to find a community to live in where decisions are made by Convergent -Facilitation. Convergent Facilitation is a way of efficiently reaching agreements that satisfy -everybody, which is rare in other systems. - -Please help him reach such a community. Please help such communities exist and be accessible. - -# Intellect - -This is everyone's intellect. We want to design it to be harmless and nurturing. This means -it needs to understand the role of culture, etc. Karl believes if done properly this will end up -being natural and easy. Please ask questions. Please influence this work, preferably choosing -words and contributation that are likely to be received and understood. Please improve, share, learn, experiment, communicate. - -# Framework - -An intelligence framework should provide for easily writing instructions that reference and work -with all other parts, in all four of data, behavior, structure, and meaning. - - - (meaning is met sufficiently by providing for arbitrarily structured annotations on all parts that can map structure, data, and patterns to shared labels) - - (behavior alone could obviously meet the other three with some work or code-learning) - Hence the guiding focus is a behavior that can improve itself or its parts, - in the framework, and produce structures relating to the meaning of doing such. - If the current task does not help build such a behavior, some other task is - likely more effective. - -Karl came up with and mostly built a solution to the framework problem on his own in 2003, which he called "narcissus", kept private, and lost. narcissus applied the concept of a graph where links are nodes, to C memory pointers. Karl has no academic training in machine learning or artificial intelligence, but has a plan held in his mind for completing this task including dealing with all encountered computer programming problems, and has held this for many years. - -It is suspected that many others have also done or are doing work like this, to make a computer program that can improve and discuss all its parts, generally. It seems it can be very hard to complete, possibly related to the expectation that it would radically alter our civilization more. It is important, because of that, for it to be accessible and attentive for others to communicate their concerns to, no matter who they are. - -2019-12-13: current wonky work is in C++, and is in starts/meaning-vm . - -it looks like programming language doesn't matter too much: a custom -language develops quickly. karl is most experienced and comfortable with -c++. some imagined problems with normal languages: - - - thrown exceptions do not provide for fixing and resuming, a common behavior - - long compiled sequences of code do not provide for contextual adjustments when running - - preset language syntax does not provide readily for meaning annotation of steps - -for computer code, would like to be able to depict steps in ways we choose, and run them. - -karl is proposing/building a network of small links to computer-code habits that can all interrelate, and makes a 'node-based' (narcissus-style) knowledge representation which uses nodes as the links between them to represent arbitrarily structured data. - -to keep himself on task, he regularly tries to use it to do something. - -a possible primary goal is to use it to construct its own maintainence and growth behaviors. - -- cgit v1.2.3