focus on main loop

logseq/config.edn

@@ -293,7 +293,7 @@
  ;;  :journal?        false} ; Default value: false
 
  ;; Favorites to list on the left sidebar
- :favorites ["product" "cyber" "ask" "search" "learn" "the product" "superintelligence" "concepts" "🫦 space pussy" "space pussy" "bostrom" "neuron" "particle" "tokens" "cyberlink" "cybergraph" "relevance machine" "cybernet" "soft3" "aicosystem" "projects" "products" "todo" "pages" "energy reform" "aips" "cyb"]
+ :favorites ["product" "cyber" "ask" "learn" "brain" "the product" "superintelligence" "concepts" "🫦 space pussy" "space pussy" "bostrom" "neuron" "particle" "tokens" "cyberlink" "cybergraph" "relevance machine" "cybernet" "soft3" "aicosystem" "projects" "products" "todo" "pages" "energy reform" "aips" "cyb"]
 
  ;; Set flashcards interval.
  ;; Expected value:

pages/ask.md

@@ -1,18 +1,20 @@
 icon:: 👽
 alias:: infer
 
-- [[cyber]] protocol allow to find verifiable
-- [[decentralized ai]] is alive
-- behold the new [[truth medium]]
+- ask [[cyber]] protocol
+- they will [[answer truth]]
 -
-- [[amount]] of [[syntropy]] bits
+- [[learn]] from it
 -
 - philosophy
-	- fundamental openai-like method of interactions
-	- in contrast to [[search]]: one input - on output
-	- designed as primary method for interactions
-	-
-	-
-	- [[soul]]
-	- [[standard inference]]
-	- in contrast to [[search]]: one input one ouput
+	- fundamental [[openai api compatible]] method of interactions
+	- in contrast to [[search]]
+	- ask is one input - one output
+	- designed as [[main loop]] method of interactions
+	- with secondary [[search]] for answers explanation
+-
+- implementations
+	- TODO [[oracle/ask]]
+		- as default uses [[soul]] to define output
+		- call [[standard inference]]
+		- summarize it by [[local llm]]

pages/brain.md

@@ -1,8 +1,21 @@
+icon:: 🧠
 tags:: page
 
 - graph file manager
-- with [[sparks]] flexible viewer
-	- supported [[particle]] formats
+- addressed to close [[main loop]]
+- and as pain with all file managers
+- core [[aip]] of [[cyb]]
+- features::
+	- [[offline mode]] first
+	- localhost interface
+	- graph query language: datalog
+	- powerful scripting: rune
+	- static and dynamic linking
+	- private and public linking
+	- publishing to [[ipfs]] and [[cybergraph]]
+	- flexible viewer based on [[sparks]] extensions
+	- supported 7 [[particle]] formats
+	  collapsed:: true
 		- text::
 		- video::
 		- audio::
@@ -11,29 +24,25 @@ tags:: page
 		- epub::
 		- web2::
 		- TODO gltf::
-		- TODO
-		- aip::
-	- [[cyb/brain/sparks]]
-- dimension:: media formats for [[cybergraph]] discovery of [[page]] type
+		- TODO aip::
+		- [[cyb/brain/sparks]]
+- tabs:: media formats for [[cybergraph]] discovery of [[page]] type
 	- graph:: 3d render with preview and discovery
+		- space::
 	- list:: classical table with powerful analytics
 	- heap:: 2d knowledge graph with preview and discovery
 	- stack:: vertical scrolling list
 	- hike:: current [[particle]] in the center
-- [[paths]]
+- paths::
 	- `#` [[cyb/brain/particle]]
 	- `!` [[cyb/brain/neuron]]
 	- `@` [[cyb/brain/avatar]]
 	- `~` [[cyb/brain/learn]]
 	- `/` [[cyb/brain/root]]
-- features
-	- offline first experience
-	- localhost interface
-	- static and dynamic linking
-	- publishing
-- actions
-	- [[oracle/ask]] or [[brain/ask]]
-	- [[oracle/search]] or [[oracle/ask]]
+- actions::
+	- [[brain/ask]] or [[oracle/ask]]
+	- [[brain/learn]] or [[oracle/learn]]
+	- [[brain/search]] or [[oracle/search]]
 - TODO [[upload brain]]
 - state of [[cyb-ts]]: [[cyb/brain]]
 -

pages/cyber.md

@@ -19,7 +19,7 @@ alias:: [[the new, yet to be born, force]], one simple protocol
 	- immortal robot [[cyb]]
 	- self fulfilling prophecy game [[aos]]
 	- [[moon]] network state
-	- and this just a tip of an iceberg
+	- and this just a tip of an [[deai]] iceberg
 -
 - TODO [[tutorials]]
 - TODO [[guides]]

pages/cybergraph mining.md

@@ -1,3 +1,5 @@
+alias:: knowledge mining
+
 - [[oas]] game mechanics
 - simple idea to gamify the process of [[cybergraph]] discovery by [[avatar]]
 -
@@ -12,7 +14,9 @@
 		- [[avatars]]
 - while [[avatars]] and [[neurons]] discover [[cybergraph]]
 - mining more and more data
-- they see the progress of [[cybergraph]] discovery completeness in %
+- we fap the progress of [[cybergraph]] discovery completeness
+	- in %
+	- in [[brain]]
 -
 - this simple mechanics is a source of healthy [[dophamine]]
 - leading to more intelligent [[avatars]]

pages/cyberlink.md

@@ -1,6 +1,6 @@
 icon:: 🔗
 tags:: cyber
-alias:: cyberlinks, unit of knowledge, simple interactions, expert opinions, essential learning ability, cyberlinking, primitive learning acts
+alias:: cyberlinks, unit of knowledge, simple interactions, expert opinions, essential learning ability, cyberlinking, primitive learning acts, creating link
 
 - directed link between two [[particles]] of [[information]] signed by [[neuron]] and recorded into [[cyber]] [[vimputer]]
 - primitive [[learning]] act
@@ -28,4 +28,5 @@ alias:: cyberlinks, unit of knowledge, simple interactions, expert opinions, ess
 	- link between two pieces of content particles there each content is addressed by its [[hash]]
 	- this makes information searchable through the spacetime
 - changing the way we [[learn]]
+- essentially work as [[view]] [[signal]] for the world
 - discover all [[concepts]]

pages/deai.md

@@ -0,0 +1 @@
+alias:: decetralized ai

pages/learn.md

@@ -1,16 +1,20 @@
 icon:: 🍏
-alias:: learning, labeling
+alias:: learning, labeling, answer
 
-- create best [[cyberlinks]] between [[particles]]
+- [[create links]] between [[particles]]
+- in a joyful process of [[knowledge mining]]
 -
-- [[empower everyone]]
-- [[learn yourself]]
+- features::
+	- empower everyone, learn yourself
+	- [[decentralized ai]] as simple as [[creating link]]
 -
-- [[decentralized ai]] as simple as [[creating link]]
--
-- you need [[cyber/energy]] to learn
+- TODO [[cyb]] packed with all energy needed for [[personal learning]] of [[brain]]
+- you need [[cyber/energy]] for [[collective learning]] of [[bootloader]]
 - or if you have [[cyber/energy]] go to [cyb.ai/oracle](https://cyb.ai/oracle)
 -
+- TODO [[oracle/learn]]
+- TODO deep integration into [[main loop]] of [[cyb]]
+-
 - tools for learning
 	- [[go-cyber]]
 	- [[cyb]]

pages/link.md

@@ -1,2 +1,5 @@
-- [[cyb/link]]
-- [[view]]
+- directed link between two [[particles]] of [[information]]
+- signed by [[neuron]]
+- and recorded into [[cyb]] [[local state]]
+- [[brain/link]]
+- link is basic unit of [[learning]]

pages/links.md

@@ -0,0 +1,3 @@
+alias:: edges
+
+- [[cyberlinks]]

pages/main loop.md

@@ -0,0 +1,3 @@
+tags:: tips
+
+- [[ask]] -> [[learn]] -> [[ask]]

pages/neuron.md

@@ -1,4 +1,4 @@
-icon:: 🧠
+icon:: 🤪
 alias:: address, subject, agent, user, observer, actor
 tags:: cyber
 

pages/oracle___ask.md

@@ -1,3 +1,11 @@
 tags:: page
 
-- and [[learn]] and [[search]] => [[cyberlink]] and [[cyberlink]]
+-
+- and [[learn]] and [[search]] => [[cyberlink]] and [[cyberlink]]
+-
+- [[decentralized ai]] is alive
+- behold the new [[truth medium]]
+-
+- [[amount]] of [[syntropy]] bits
+- actions
+	-

pages/oracle___search.md

@@ -0,0 +1,7 @@
+tags:: page
+
+- instantly and censorfree
+- find and deliver content
+- decentralized search is just one [[aip]]
+-
+- [[amount]] of [[particles]] and growing

pages/search.md

@@ -1,12 +1,15 @@
 icon:: 🔍
 
-- instantly and censorfree
-- find and deliver content
-- decentralized search is just one [[aip]]
 -
-- [[amount]] of [[particles]] and growing
+- [[cyber]] protocol allow to search for [[particles]] in [[cybergraph]]
 -
 - philosophy
-	- fundamental google-like method of interactions
-	- in contrast to [[ask]]: one input many input
-	- allow to understand why [[ask]] answer what it did
+	- fundamental [[google like]] method of interactions
+	- in contrast to [[ask]]
+	- search is one input - many output
+	- integral part of [[main loop]] method of interactions
+	- allow to understand why [[ask]] answer what it did
+-
+- implementations
+	- [[go-cyber]]
+	- [[oracle/search]]

pages/standard inference.md

@@ -1 +1,32 @@
-alias:: cyberlinks weight, cyberlinks weights
+alias:: cyberlinks weight, cyberlinks weights, cip
+
+-
+- algorithm addressed to solve [[true false problem]] for inference of truthful [[answers]] from [[cybergraph]]
+- algorithm features
+	- does not introduce additional complexity
+	- requires zero additional information
+	- follow occam razor principle, that is it is as dumb and simple as possible
+	-
+- algorithm
+	- get all [[particles]] cyberlinked with asking [[particle]] sorted by [[cyberank]]
+		- for each [[particle]] get all [[cyberlinks]]
+			- for each [[cyberlink]] get [[neuron]]
+				- for each [[neuron]]
+					- get [[will]] balance
+					- get number of [[cyberlinks]]
+					- compute average [[will]] per cyberlink
+			- sum ([[average will]]) votes for every [[particle]]
+- multiplay [[cyberank]] of each particle by sum ([[average will]])
+-
+- in essence we weight [[will]] of neurons on global [[attention]] in [[context]]
+-
+- any intelligent system must solve [[true false problem]] for [[contextual awareness]]
+-
+- implementations
+	- [[cy]]
+	- TODO [[oracle/ask]]
+-
+- continue to
+	- [[relevance machine]]
+	- [[soft3]]
+	- [[elon launch rocket]]

pages/true-false problem.md

@@ -0,0 +1,23 @@
+alias:: true false problem
+
+- we discussed for a while one property of the [[cyberank]]
+- which rise huge concerns into the answering capabilities of [[cyber]]
+- lets call it true-false problem
+-
+- let us imagine that `true` particle have cyberank `10`, and `false` particle have cyberank `9`
+- the environment allow to link [[any particle with any]]
+- that means that for any questions which cyberlinked to `true` and `false`
+- the winning answer will always be `true`
+-
+- of course such behavior does not feels like something superintelligent
+- we always emphasize that [[cyberank]] is a first ranking factor
+- but [[not the only one]] which is possible on top the cyber architecture
+-
+- in order to solve true-false problem
+- we have to extend our [[universe model]] into [[two factor]] model
+- we must compute [[weights of links]] in a context
+-
+- [[standard inference]] introduce simplistic algorithm for solving this problem
+- giving as result contextual statements which are [[truthful]] by design
+-
+- [source of discussion](https://github.com/cybercongress/go-cyber/issues/694)

pages/truth.md

@@ -0,0 +1,3 @@
+alias:: find truth, compute truth, answer truth
+
+- result of [[standard inference]] computations

pages/truthful.md

@@ -0,0 +1 @@
+- either [[knowledge]] more [[truth]] or not

pages/two factor.md

@@ -1,4 +1,5 @@
 tags:: tips
+alias:: not the only one
 
 - model type of the [[universe]]
 - follow [[introduction to bostrom for ai geeks]] to understand context