Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Logic Machine

No description

Jacob Curtis

on 25 November 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Logic Machine

is here
Logic Web
Insert a Question
There are many different lines of reasoning to reach a particular truth. Using lines of reasoning we should be able to derive every single truth from a grain of sand.
Lines of Reasoning
Process of elimination
Scientific Method
What's the easiest line of reasoning? How could've people from thousands of years ago come up with the truths we know now sooner?
Questions that be needing some answering
Criteria by Disorders
autism- we select what sensory input to focus on.
Science Decisions
What will the web look like?
Probably won't know until finished.
-How do premises relate?
Some premises are premises for more than one other premise.
How does the web connect?
-How many premises do we have?
-2 a day for 75 years= 54,750
How many premises can we hold at once to make a decision?
How many premises are possible? Every sentice is a premise. Figure out how many possibilities there are. Use dictionary to make all possible combinations.
Need to invent new unambiguous language. One word, one meaning, or figure out how to differentiate meaning with the language we already have.
Is the tree only a couple of layers deep?
Edited by world like Wikipedia?
Can I translate everything into numbers?
- Are there some questions we'll never have an answer to? Ex. Who was God's dad?
- Once the right basic level premises are realized, through that lense, all truth can be understood.
- We would learn a lot from asking all possible questions: why god create ____? Why God create spirits, Satan, etc.
- The first moments of the logic web being activated, will be spent correcting all of the logical flaws we used to create it.
- Find out how Wikipedia works.
- Why the logic tree will never be able to prove anything: There is a limited amount of evidence. (this was addressed in psych stats). We can only be in one place at one time. We can only focus on a limited number of things. Also, restricted by the limited amount of time.
Do we already know basic rules of logic at birth or do we learn it? Do we gain more rules over time or all at once?
How can the logic tree be emperically tested?
Evidence "falsifies" other theories
There's not so much complexity in the world that we can't organize it. Only a limited number of things, elements, possible ideas, etc.
-Computers can't tell what things mean. Can input data, but don't know why the data is important. But we can teach them.
- Would SM be most efficient and capture everything by just totally understanding and explaining physics and chemistry?
At least at first, there probably won't be one perfect LM, probably just competing machines that slowly get better over time.
-Will allows users to manipulate or turn off and on certain premises they agree with. Gives them more control.
- The fact that subatomic particles cannot be determined with accuracy is evidence against determinism.
- Maybe I can lay out premises and reasons and I can create a computer program to find the rules for connections.
- This is so hard because humans share basic assumptions about the world. Building a knowledge based from a blank slate is so hard.
- It's going to be constantly updated, so updating should be easy to do.
- Is it possible to adopt a world view that has no loose ends?
Whenever my friends and I got together in high school, we would spend most of the time arguing. It was usually about surprisingly deep stuff for teenagers: politics, philosophy, religion, morals, etc. I remember hiking with my friends once and realized I want to invent a logic aid. Something, two people would have in an argument and it would call people out whenever they made a logical mistake.

I was 17 when I opened up a word document, entitled it decisions and subtitled it going on a mission. I wrote, "We all have to make decisions. Some are generally insignificant, some are significant and some are life changing."
We're pretty much thrown into this world with no idea what's going on. Some decisions are unconscious, where our conscious mind doesn't have a say in the matter, but a great deal of decisions, usually the important ones require us to make decisions.. But others Guided by biological instincts, basic logic we muddy our way through our decisions, using hueristics, being right only a portion of the time. We seek outside help. We seek friends, parents, the internet, books, wise men, and fortune tellers or even ennie meenie miney mo and most decisions are made on a whim. Where should I go to school?
Sure, sometimes we know the right path, and go down a different one anyways, but a great majority of the time we just don't know the right way. Memory is incomplete (we can only hold mental represenations. We can't fit everyone's faces)
I came up with the idea of the logic web when I was 19. Neccecity is the mother of invention. Knowing that I needed to make a decision of whether or not I was going on a mission, I thought a lot about what I believed. At that point, I described myself as agnostic. At that point, the debate in my mind was over whether the LDS church or was true or there was no God-- I recognized, however, that really there were hundreds of theories about the world out there and any of them could be true. But, I certainly focused the most on whether the LDS church was true or not. That was the most pressing question. There was so much data, evidence on both sides that I realized I was overwhelmed. That’s when I realized I needed something, a logic tree, that could help me organize my thoughts and come up with conclusions for me .Created out of my desire to have something think for me. Sustained by my satisfaction from organizing.

Brain labyrinth. Didn’t expect. Lernaean Hydra. Every answer, just leads to more questions.

Evolution: At first, I thought it would just be like a geneology tree. Each premise leading to another premise. Than I realized it might look more like a web, since a conclusion could be a premise for multiple different conclusions. Than I realized it was probably going to be even more complicated than that, would probably need to use logarithms, decided it would have to be a computer, or machine. I realized later that their might not really be one web, but a few different interacting webs (one for the physical properties of something and then different webs will pull from that one.)

The week before I left on my mission, I furiously worked on the web thinking it possible to finish by time for my talk.

In the MTC, I prayed about whether I should continue to work on the logic tree and I felt like I should.

I thought about the logic machine whenever I could on my mission. I'd write notes in my journal, and on a p-day copy them into a logic machine folder.

To my surprise, when I got back from my mission, didn't really focus on it. I'd go months without having a new thought. I think the logic machine was a place I go to do as a defense mechanism of how hard the mission was, but I didn't need it when I got back.

Hopefully, I think more about it this year. Going to try to see everything through the lense of the machine.

We get stuck in our individual logic webs. There is some premise out of line that we can’t find. WE can think and think about the problem but not get any closer. It takes someone who doesn’t have that limitation to show the light. The logic tree can do that.

Overall, the brain is a pisspoor logic tree, having the abiliity, but much more focused on survival (we only recently develped forebrain, withint the last 2 million years, majorit of brain used for basic survival, (breathering, temperature, heartrate, fight or flight), racial bigots (cling to belief even though lots of evidence around them it's not true) pleasure and social success to produce much truth. I primarily learned this on my mission, talking to hundreds of people about their beliefs. The telephone game. So many people had objectively crazy core beliefs about the world. We just can't digest all the data around us. We don't have time. There are hundreds of thousands of research articles alone. " illogical- If hungry at store, buy tons of stuff even though they don't need it.

Logic web would be involved in every part of our lives, because we're always thinking, and it aids in thinking. Imagine having all information at our fingertips. It would know more than the smartest man that's ever lived multiple times over, making all books obsolete. It would watch us and give advice without us even asking. We could tell it our thoughts and it would write an essay for us.

Every time I learn something new. Incorporate it into what it means for logic web. So, I don’t lose interest because it evolves to be relavent to my interests.

On mission, wrote thoughts in planner. Near end of mission, looked through planners, copied it on to paper, trying to organize. When got home, pinned on flat cardboard box (was thrown away).
Three different types thought about.
1. The logic webs in our minds. As the best example of a logic web, good to know how it works.
2. The logic web of a group of people.
3. The actual logic weboi

Everything thing that has ever existed in this world has left behind evidence that it exits.
Right now, I don't have the time to analyze every conspiracy theory. Would be able to so can actually learn the truth of those that are actually true. Won't have to use heuristics.
would it be a fact machine allowing us move beyond teaching and learning facts and instead focusing on big picture ideas and theories.
Cognitive Map
What types of premises are we born with? Claims of fact, policy, and value?
How "sure" he is.
F it. We'll do it live!
Is the LDS Church true?
Take all the different things humans do (disorders) and eliminate theories what that means about the way the mind works.
People with dissociative identity disorder can sometimes have a personality that’s straight and a personality that’s gay, but they can also have different eye prescriptions.
Truth is not easily accessed by gradual progression of logic web, but through God through the medium of the Holy Ghost.
Each line of reasoning will have
it's own premises.
How much control do we have over evidences? Are we basic robots who are slaves to the evidence we see or is there more control ( Maun’s kids leaving church).

How are our minds and logic related. Logic is something it uses, but it’s independent, right?
The evidences and premises provide limits of how the world works. Rules out possibilities. Creates possible theories.
If evidence is true, what is the world like?
The evidences and premises provide limits of how the world works. Rules out possibilities. Creates possible theories.
If evidence is true, what is the world like?
Logical Fallacies

Post hoc ergo prompter hoc

False analogy- analogies without relevant connection

Ad Hominem- “against the man”

False dilemma- Black-white- only 2 alternatives

Slippery slope

Begging the question- assumed premise

Straw man- attack on view similar but not same as opponent holds

Two wrongs make a right

Non sequitor- “it does not follow”

Ad populem- Appeal to prejudice of the people

Appeal to tradition

Faulty emotional appeal
A List Of Fallacious Arguments

"The jawbone of an ass is just as dangerous a weapon today as in Sampson's time."
--- Richard Nixon

Several of these have names in Latin, but I mostly ignored that and used English.

If anyone is bothered by my using "he" everywhere, note that "he" is the person arguing

Ad Hominem (Argument To The Man)

Affirming The Consequent

Amazing Familiarity

Ambiguous Assertion

Appeal To Anonymous Authority

Appeal To Authority

Appeal To Coincidence

Appeal To Complexity

Appeal To False Authority

Appeal To Force

Appeal To Pity (Appeal to Sympathy, The Galileo Argument)

Appeal To Widespread Belief (Bandwagon Argument, Peer Pressure, Appeal To
Common Practice)

Argument By Emotive Language (Appeal To The People)

Argument By Fast Talking

Argument By Generalization

Argument By Gibberish (Bafflement)

Argument By Half Truth (Suppressed Evidence)

Argument By Laziness (Argument By Uninformed Opinion)

Argument By Personal Charm

Argument By Pigheadedness (Doggedness)

Argument By Poetic Language

Argument By Prestigious Jargon

Argument By Question

Argument By Repetition (Argument Ad Nauseam)

Argument by Rhetorical Question

Argument By Scenario

Argument By Selective Observation

Argument By Selective Reading

Argument By Slogan

Argument From Adverse Consequences (Appeal To Fear, Scare Tactics)

Argument From Age (Wisdom of the Ancients)

Argument From Authority

Argument From False Authority

Argument From Small Numbers

Argument From Spurious Similarity

Argument Of The Beard

Argument To The Future

Bad Analogy

Begging The Question (Assuming The Answer, Tautology)

Burden Of Proof

Causal Reductionism (Complex Cause)

Changing The Subject (Digression, Red Herring, Misdirection, False Emphasis)

Cliche Thinking

Common Sense

Complex Question (Tying)

Confusing Correlation And Causation

Disproof By Fallacy


Error Of Fact


Exception That Proves The Rule

Excluded Middle (False Dichotomy, Faulty Dilemma, Bifurcation)

Extended Analogy

Failure To State

Fallacy Of Composition

Fallacy Of Division

Fallacy Of The General Rule

Fallacy Of The Crucial Experiment

False Cause

False Compromise

Genetic Fallacy (Fallacy of Origins, Fallacy of Virtue)

Having Your Cake (Failure To Assert, or Diminished Claim)

Hypothesis Contrary To Fact


Inflation Of Conflict

Internal Contradiction

Least Plausible Hypothesis

Meaningless Questions

Misunderstanding The Nature Of Statistics

Moving The Goalposts (Raising The Bar, Argument By Demanding Impossible


Non Sequitur

Not Invented Here

Outdated Information

Pious Fraud

Poisoning The Wells

Psychogenetic Fallacy

Reductio Ad Absurdum

Reductive Fallacy (Oversimplification)


Short Term Versus Long Term

Slippery Slope Fallacy (Camel's Nose)

Special Pleading (Stacking The Deck)

Statement Of Conversion

Stolen Concept

Straw Man (Fallacy Of Extension)

Two Wrongs Make A Right (Tu Quoque, You Too)

Weasel Wording

Some other Web sites:

Critical thinking vs. Specious arguments

The Nizkor Project

Propaganda Techniques Related to Environmental Scares

Logical Fallacies

Critical Thinking Across the Curriculum Project

The Atheism Web: Logic & Fallacies

South Shore Skeptics

Introductory Logic

Elementary Logic

Practical skepticism

Bruce Thompson's Fallacy Page

Critical Thinking On The Web

Conversational Terrorism: How NOT to Talk!

Love is a Fallacy, a funny short story by Max Shulman.

How to Win Any Argument On The Internet, a cheerfully obscene guide to flaming.

How to Argue and Win every time. Alcohol and lying appear to be involved.

Recently changed entries:

Ad Hominem (Argument To The Man)

Two Wrongs Make A Right (Tu Quoque, You Too)

Changing The Subject (Digression, Red Herring, Misdirection, False Emphasis)

Argument By Personal Charm

False Compromise

Argument From Age (Wisdom of the Ancients)

Least Plausible Hypothesis
Logical Fallacies
Logic Tree use in everyday life
Smoking gun- evidence that proves the certainty of a certain premise.
Theory- a base premise.
We all have a few big “T” truths.
Beliefs can cause us to overlook valid evidence.
- If you want a reason to disbelieve, Satan can cause one.
Doctors and lawyers are evidence weighers.

Socially bad: Quick to assume you’re right and others are wrong.

Mistunderstand: when two people’s premises conflict.
Sense Input.
2 + 2= 4
2 + 2 = 4
Math is fun
I understand!
1, 3, 5, 4
Consciousness/ Metacognition/ frontal lobe
1, 2, 3, 4, 5
1, 5, 3
2, 4
Collect data.
Inductive logic
Schemes- A claim of a perceived reality. 3 types: 1) Claim of fact (santa
real). 3 subcategories: Past (was), present (is) and future (will be). 2) Claim of policy (we
not kill), and 3) Claim of value (I am
than you (some action belief or condition is wrong, good or bad, beautiful or ugly, worthwhile or undesirable. You cannot dispute tastes - Latin proverb.
Updating software. After playing a Monopoly Deal game, using strategy from that to pick a tie.
Can adjust at any part.
What does the structure of our sentences tell us about how we think? We've just seen a tool, the Book of Mormon, that's really helped us out (as opposed to being at the end).

How much are our sentences accurate representations of what we're thinking?
Basic Schemes/self-evident truths/ core domains
- We're born with them. Instincts. Either given by God (Light of Christ) or result of evolution.
- e.g., hurting people is wrong and physical, numerical, linguistic, and biological.
- Language learning capacity
Sense Input: Premises and evidences through senses/ scientific tools. Input from textbooks, what we sense in our experiences through our senses or tools(cameras) with us.

Are there more than 5 senses?
Evidences- little pieces of reality. "something that makes another thing evident. Sign. How many evidences are there? Can I demarcate the size/weight an evidence has?
February 2009- Now
Perceive/consider/label evidences-
The smoking gun- The evidence that weighs to much it destroys all doubt. It "proves" a scheme. (Thomas wanted a smoking gun.
“In many ways we know more about the moon than the brain.”- Ab Psych
Faith vs. Reason
There shouldn't be a faith vs. reason debate. I believe in faith as a motivating force to do something that your reason tells you is probably true. The notion of faith as something that contradicts reason would be, perhaps, the most dangerous notion in the world, except for the fact that no one actually does it. We all do what makes sense to us, what we reason to be true.
Weighing new schemes/ finding the most likely one: How many possibilities? infinity?
- Occum’s razor (urges a preference for the simplest hypothesis that does all
we want it to do.
- Which one has the least I don’t knows
Weighing new schemes/ finding the most likely one: How many possibilities? infinity?
- Occum’s razor (urges a preference for the simplest hypothesis that does all
we want it to do.
- Which one has the least I don’t knows
Weighing new schemes/ finding the most likely one: How many possibilities? infinity? Always a best. Make lemonade.
- Occum’s razor (urges a preference for the simplest hypothesis that does all
we want it to do.
- Which one has the least I don’t knows
Creation of new schemes based on evidences.
Does every thought produce a feeling?
Emotions are made up of schemes. Attitudes are made up of emotions?

4. How decisions are made today- by debate with all leaders
5. Reason Bible similar to BOM insight, the prophets had records written by earlier prophets that both used
6. 3 Witnesses
7. Patriarchal blessing
8. The difference between J.S. and other religious founders is that Joseph encourages others to confirm his religious moments by their own.
9. W.O.W. Temperance Movement. Lots of religious groups had similar health codes. Has a historical/social context. Didn't just pop out of nowhere.
10. Noah's flood impossible.
11. Other religions have religious experiences
12. Holly rollers sinners- Jerry
13. Rev 6: Dueteronomy 32:4- He is the rock, his work is perfect; for all his ways are judgment; a god of truth and without iniquity, just and right is he
14. Jacob 4:13
15. Polygamy
16. A lot of BOM is same verses as from Bible. On the other hand, a lot are different and very insightful
Circumstances neccesary for making of BOM (Missionaries little book of teaching tools p. 14-18
17. BOM racist against dark skinned (Alma 3:6)
18. Few new revelations from modern prophets
19. The reason blacks couldn't hold the priesthood was because they were less valiant in the preexistence
20. Reve 3: Spirit is complimentary. J.S. wrote D&C and other prophets wrote their revelations differently. 95% of preparing for District meetings was my own thoughts. Just got a nudge every once in a while.
21. Not a ton of historical evidence for the BOM.
22. Your church hasn't been around long enough to have a bad prophet- Lyn Schelp.
23. - If not from God, where would commandments come from?
24. -Tied in to social norms.
25. - Strength of Youth changed
26. -Tattoos, homosexuality
27. Correlation dept.
28. - Quorum of the 12 discuss policy. know it's from man because it was debated
29. Other similar relgious experiences like Joseph Smith. They contradict each other. Can't all be true.
30. J.S's papryi.
31. DNA of indians
32. Circumstances of how the BoM was written. 3rd grade education, etc.
33. Our prophet acts like the leader of any other church.
34. "Lamanites today extant upon the land as American Indians." - JTC Talmage
35. Jacob 2, D&C 132. Contradiction over polygamy.
36. Wow. Incling of science at the time.
37. Church leaders don't distinguish between when what they're saying is coming from themselves and from revelation.
38. earth created in six days
39. No death before the fall - D&C 101:29
40. If J.S. would've said God spoke with him, instead of just introducing him, would've ruined everything.
41. Was the garden of Eden really in the garden of Eden?
42. No new significant revelations. Counter: could be because God just hasn't needed new revelation for a while.
43. That Isaiah verse in the BOM that was changed or something. I learned about on my mish.
44. The papyri from The Book of Abraham
45. At age 8, you’re considered accountable. At age 8, brain size of adult.
46. At age 8, brain size of adult.
47. Egypyian endowment
48. Patriachrcal blessing tribe
Evaluating evidence
What adopting certain premises leads to
When learned, ye think ye are wise (2 Nephi 9:38-29)
My Conscious Premises
Tower of Babel
Fluid and crystalized intelligence
2 parts to machine
Crystalized and fluid
What we want to know
Past (was), Present (is/are), future (will be) and how we/others should act (action).
Our brain will make best guess. Even if no evidence, need to find answer. Will guess if need to.
MMPI: factor analysis on MMPI. Random questions. Found connections using factor analysis.
Analyzing Organization of Arguments
1st layer: What debatable premise is
Unorganized thoughts
- man doesn't like not knowing

- Gottlob Frege sought to break math down into pure reason. He had to use self-evident truths.

1 There is no revelation. The things we describe come from the reasoning power of our minds, unaided by any supernatural force.

2 Their is some revelation. Some of the things we describe as revelation come from God (these statements are true and commandments are moral), while some come from the reasoning power of minds, unaided by any supernatural force.

3 Their is some revelation. Some of the things we describe as revelation come from God (his statements are not true and commandments not moral), and some come from the reasoning powers of our mind, unaided by any supernatural force.

4 There is some revelation. None of his statements are true and no commandments moral.
5 All the things we describe as revelation are from God (all of his statements and commandments are true and moral, respectively).
6 All the things we describe as revelation are revelations (some of his statements and commandments are true and moral, respectively).
7 All the things we describe as revelation are revelation. (None of his statements and commandemtns are true and moral, respectively).
Whether the things we (Members of the Church of Jesus Christ of Latter-day saints) deliniate as revelation (personal, scriptural, and leaders on every level) come from God and whether they're true and moral
NoneSomeAllNone MoralSome moralAll moral
1. Achan's story4- There is some revelation. None of His revelations are just and true
1. Achan's story4- There is some revelation. None of His revelations are just and true
1. Achan's story4- There is some revelation. None of His revelations are just and true
Input evidence
Evaluating website: authority, accuracy, objectivity, currency, coverage.
Current enough, author expert.
We use emotion when we don't have a logical answer.
Elephant story
Elephant story
Weighing evidences

3. Priesthood and blacks

Priesthood and blacks (the lineal restrictions of the Aaronic (Levitical) priesthood were lifted when the law of Moses was fulfilled, and thereafter the offices of the Priesthood were conferred upon worthy men without limitation to the tribe of Lehi.

Priesthood and blacks (the lineal restrictions of the Aaronic (Levitical) priesthood were lifted when the law of Moses was fulfilled, and thereafter the offices of the Priesthood were conferred upon worthy men without limitation to the tribe of Lehi.

Priesthood and blacks (the lineal restrictions of the Aaronic (Levitical) priesthood were lifted when the law of Moses was fulfilled, and thereafter the offices of the Priesthood were conferred upon worthy men without limitation to the tribe of Lehi.

Priesthood and blacks (the lineal restrictions of the Aaronic (Levitical) priesthood were lifted when the law of Moses was fulfilled, and thereafter the offices of the Priesthood were conferred upon worthy men without limitation to the tribe of Lehi.
What would the world be like if this were true?/ testable hypothesis

- Would have to couple this with other premises. Would it just pair it with all other premises. No. "Peanut butter is brown" is not a premise that can be used with premise number 1. Peanut butter is brown + There is no revelation = ?. What premises are relevant?
What would the world be like if this was true?
What would the world be like if this was true?
What would the world be like if this was true?
What would the world be like if this was true?
What would the world be like if this was true?
What would the world be like if this was true?
Don't pay attention to everything. Change blindess. which nickel is right.
Is the brain a terrible logic machine, designed to survive and have social success, and not necessarily to find truth? Yes. I think so.
We only internalize, learn, when we can associate a premise with what we already know.
- Contiguity (associate things that occur at the same time.
- Similarity (associate things similar features or properties)
- Contrast (associate things that show polarities (hot/cold, etc).
What would the world be like if this was true?
Metacognition (higher-order executive process used to plan, monitor and evaluate our logic machine.)
Weighing/scoring premises: Are there negative evidence that makes premise impossible? How strong is that evidence on a scale of 1-10? If a 10, must reject premise. (don't necesary have to accept as true if a supporting evidence gets a score of 10.

Almost everything can be explained through every premise.
- Whichever is simplest.
- Fool me 3 times, shame on me. One counter-evidence might be a coincidence, but if there are enough, you need to adopt the oposing premise.

don't throw the baby out with the bathwater.

The truth isn't always any of the theories we currently have. The logic web would have to creat theories, not just evaluate them.

Falsifiability view: The temporary nature of inductions makes negative evidence more important than positive support. If data confirm a prediction, one cannot say the theory is true. However if a data leads to a prediction that is not supported by teh data, then Popper would argue that the theory must be false and should be rejected.

Categorize evidences:
Smoking gun-
If multiple theories work, whichever one is the most "simplest" wins. The fewer the statements in a theory, the better the theory (occum's razor)
We select what to focus on by our interests and purposes.
Characters need to be very specific.
- Our language is too ambiguous. Allows for too many miscomunications. How many times someone said one thing and the person takes it as another. The Bible. No, what he's really saying here is....
Wars have been fought over misscomunication.

Every character will represent a specific wavelength?

Language tends to evolve over time. Very strict rules will be placed on this one. Efforts would be made to keep it from evolving. It wouldn't need to. It'd be perfect. And not all languanges evolve as much as english has (Chinese).

Will be no exceptions. No more of this: i before e, except after c, or when sounding like A, as in neighbour or weigh.


Just written. We can get more specific written because we're not limited to only producing characters represented by sound. We can possibly do words, with smaller marks surrounding it with what it means.

What about new words or ideas. How can we incorporate them into the language? I'll have to have a way to do that.
Different ways we like to talk about everything:



Look these up:
Nouns (construction)

Articles (a, an, the) are useless. Any other parts of language useless?
For Prezi: Background on first languages.
Pictures of ideas --> sounds ---> thoughts.
Sumerians. Cuneiform. For business transactions.
Akkadians. Semitic language
Advances in language spurred great cultural movement. Greek alphabet (Homer's writings, Aesop's fables, Hesiod's writings.
Previous writing systems communicated ideas. This one communicated sounds.
Less ambiguity.
Brought them out of the dark age.
History, laws, business (all improved)
More democratic, easy recorcing of details, expression of subtilties
How many languages there are today.
Realized had to be specific.

Give example. Teacher asking: How many uses can you think of for this cup? A student responding.

Some of the implications, and consequently, why it makes me happy:
Defining Lds: Prophets. Church correlation department, scriptures
Thalamus: relays sensory information to cerebral cortex.
Lemonaide, secrets of the universe from a spec of dustWe need only one bit of empiracle evidence and rationalism can explain the rest. A perfect rationalist can tell the secrets of the world from a grain of sand.
Frontal lobe:
motor processing and higher thought process (abstract reasoning, problem solving, planning and judgement)
Parietal Lobe
So, what's it going to be?

-Computer code (1's and 0's)
-Java script
We know language can be specific because computers work that way. Every complexity over the internet is communicated precisely. That's why when you open up a website from one computer, it's the same as on another.
Will also be useful because we'll speak the same language as the logic machine.
- Brain waves (bypassing language all together)
- We're still trying to map out the brain. Eventually, that'll be finished. and then what?
EEG, MRI, etc
Not precise enough right now.
We know what parts of the brain language is in. Wernick's and Broca's area.
$550 for an hr in MRI
Lie detector test
Everyone would be able to speak and write.
Patients in comas.
Wouldn't need to take tests. Just take a snapshot of the brain to see how much of the material we know.
Would be able to read animal's minds.
- Numbers
but, not everything can be compared nominaly (Pythagoras thought it could; as did Aristotle "reality is highly organized and can be represented by small whole numbers". "Things were numbers and numbers were things""The world is becoming quantum"-stats byu), but can logic, premises?
"Modern quantum chemsistry states certain small whole numbers strictly describe the energy levels allowed for electrons an, therefore, determine the shapes of molecules>"
Authority: Religious, parents, professors, news
Look for patterns/algorithms
Create a list of all the different theories
Create a list of reasons people give for doing things.
Calculates what world be like if that premise was true and matches up to reality based on collected evidence.
If different, reject. If same, accept.
Deductive Logic
Action-Jackson List
I will not be able to make the logic machine all at once. First, I'll need to create a simple working framework. Something that's not perfect, but is a good framework that can be evolved.

If have to narrow down, focus on church or science.
How do computers do logic?
- http://computer.howstuffworks.com/boolean1.htm
3 criteria for evaluating theories: parsimony/Occam’s razor (The fewer statements in a theory, the better the theory. If a theory needs a separate statement for every result it tries to explain, clearly nothing’s been gained by the theory) (LDS church?) (If two theories can explain the same number of results, the one with fewer explanatory concepts is to be preferred) (Every thing can be explained by anything). , precision, testability (ESP, the church? What about the spirit?.
Unconscious premises
Filing cabinet
How many researchers have been done? We've only looked at a small percentage of the world. We should be collecting data from everywhere.
Argument Map
Start with church doctrine (see "F it. We'll do it live: 1) Could even be an app. What do we believe.

After this, move on to whether or not the church is true or not.

LDS Doctrine
What is "LDS Doctrine"?
What type of question is it?
General reason database. Which ones apply?
Who said it?
Belief it to know as
inconclusive is to conclusive
Can write essay for us. We give it thoughts and it puts the thoughts into an essay form. It asks you questions. Ie. What is your paper about. What do you think about it. Why do you think this. Find some evidence this is true.
14 points of a prophet.

Seer- sees things not visible with the naked eye
The one I believe
What about poetry, songs?

A long time, but we’re on an inevitable path. Magnetic devices becoming more precise.
Measuring energy of orbits of nuclear particles in the molecules of the body.

Measuring blood flow when doing a technique. Doesn't use radioactive particles.

But you have to be in the tunnel.

Can detect changes in state. Can tell how long we think about a certain thing.

x-ray technique
blood flow. die.

x-ray technique. 3-d.

Increases in oxygen.

Aristotle's 4 scientific method questions
- Material: What is it made of?
- Effiecient: What caused it?
- Formal: Its essence? What it's categorized as.
- What is its purpose?
How many natural phenomena does it explain?

3. Priesthood and the blacks

4. Achan's story


2. Members have a hard time telling the difference between when their own mind and revelation

3. Priesthood and blacks


2. Members have a hard time telling the difference between when their own mind and revelation

3. Priesthood and blacks


Input data: _____. Did you mean this, this or this? (just different definitions from a dictionary). Thank you! Your data helps support this theory and this theory.

Categorizing Variables:
Tense, parts of speech (noun, verb), artifact/natural, amount, possessive.

Thematic Roles
- The agent
- The patient
- The benficiary
-The instrument
- The location
-The source
- The goal (p. 385)

Input descriptive statistics: mean, medium, mode


People can input their own data or data already completed.

After data is inputed, website will ask how many people witnessed this. If they say multiple, they'd have to verify.

Will give suggestions based on holes in knowledge and what questions people have been asking. Ecological validity.

Ask a question:
Score "Is the LDS church true" premises 1-10
Implications: Ramon Lull-type disks

Are the data inconguent with implications
Alternative names:
The God machine
Science Machine
Reality simulation
Decision Machine
Decision Calculator

HOC project due 11/29

Learn how to write code: Learn how to code pythagorem theorum, ticktac tode

Y combinators hacker news message board
First slide: a picture of aristotle.

Second slide: a picture of science (Arisotle, measuring tube, equations) merging with a picture of a computer

Algorithm comes from al-khwarizmi (Persian, 9th century)

Summerians. 2500 BC. Clay tablets. Repeatable division of frain.

Arisistotle was the first to classify things.

Mechanical Turk
Look up: game theory

Mark-up languages
Cannot input subjective
Descriptive Research:
Everything in the world can be reduced to what we see (unobtrusive observation) and what we can't (unobtrusive measures). What we can't see though, leaves footprints.

Footprints: we examine it after the fact, by measuring the results of the thing, not the thing itself (everything that exists impacts its environment and leaves some type of footprint).
- But what about things that leave the same footprint as something else (Psychologists used to think moms with autistic kids were bad parents because the kid wouldn't make eye contact with them. The kid was really just not making eye contact because of his neurobiology.

- Nothing leaves the same footprint. That's why precision is important.
How do you get from the footprint to the dinasaur?
- Relationships. People inputting, "this dinasaur walked across my field and left this footprint." And then we know that footprint someone else finds belongs to that dinosaur.
1720 volumns
9 stacks

Evaluating the cost of a printed Wikipedia is fraught with difficulties. As of 14 March 2010 there were approximately 14 billion characters so assuming 5,000 characters per page that would yield 2.8 million pages. If you then add 25% for extra space for photos, tables, and diagrammes that would yield 3.5 million pages. This would produce 8,750 volumes of 400 pages each. As an example, allowing US$0.05 per page would yield a cost of US$175,000 without binding.

Mechanical Turk
Suggestions for data to look at: In your city, we know there are these rocks and this, but we don't have their location.

How many variables can be on a contingency table?
Before you figure out the whole world, do LM on one room with a limited number of variables.

Input current scientific knowledge
Will need algorithm to detect liars.

Scoring system to incentivice people to participate.
- More points for being more precise

Should a majority of information come from scientists? Their scores are dramatically higher?

Start vague, get more specific. Fundamental variables first.
After data is inputed, the website will report back what theories they contributed to. And give them their total score.
One of the first things would be to incorporate already existing scientific thought.
Is/are web:
LM should be able to know normal curve for everthing.
Deep Thought
How could sight and footprints come up with evolution?
Science Web
My website probably won't take off. But I think a logic machine is inevitable and we should see a GPS for ever decision we make in the next 20 years.

Connected to Googleglasses
Anything science does now to get rid of biased untruthful conclusions, a computer can do a better job.

Still be a place for the spirit.

No one will force anyone to follow the logic web.

God would never want us to disavow logic.
LM can find correlations. Can it draw theories based on those correlations?

Could it figure out the correlation between head size and intelligence? YES, it could.
What you can do (input data, ask a question)
Walk them through the process.
Input data: 25 grapes are purple.

Ask a question:
What religion should I join?
Is there a God?
What is life like after death?
Should I invest in Google?
If I try to jump over this river, will I fall in?
Should I marry Jane?
What are all the uses of a paper cup?
Sometime people say science can't answer all questions. The science web can, including humdrum day to day activities and questions about religion.
Imagine if everybody used this. If every idea and thought someone had were compiled together.

An arguement moderator app on your phone.

More stratified samples. Not just white college studentes.

Will eventaully be used to be able to manipulate all variables to control the world.

It will be practicle. Suggestions will come from questions people ask.

Will be answer deep questions of the universe from the properties of a specific spec of sand.
Right now, most of our thinking is done on our own.
Simplified World
Simplified World Room
Get a box. Put two objects in it that you know a lot about.
Learn how to code:
Pull existing scientific data from scientific journals.

A cardboard box with a stick and piece of paper.

Goal: Input data then be able to predict everything that happens in that box.
Simplified World variables
The stick
The paper
The future
When wikiscience is connected to googleglasses. You would know what everything is and all of it's properties. And you can know the result of all your actions. The wikiscience glasses would recommend certain actions.
Which variables interact?
Can I convert everything from the Logic Machine to if/then statements.
ACT John Anderson 1976, 1993, 2001, 2004
How it works

Abstract Concepts
-Show website-
How does one know egg will fall when thrown off Eiffel tower when we've never seen it happen?
Will answer questions of theology: is there a God, which church is true, etc.

Who should I vote for?
Historical Background
Parts of the website
Science Web
There are 80,000 Morphemes in English.
Ask a question
1. Ask a question: Which planet is the biggest?
2. Operational translation: Planet: size: biggest

$1=first row;
if ($1=="planet") search in the array 2nd row// How do I get it to look through Descriptive Web?
if (number in 2nd row is the most) echo that number.
3. Find merger of relevant variables.
Create an array for the descriptive web:
$Descriptive_web=array("planet", "Jacob Dee Curtis")
$Descriptive_web [0]="planets in our solar system";
$Descriptive_web [1]="Jacob Dee Curtis";
Assign values:
$Planets in our sun's solar system = array("Planet"=
Create an array for each subcategory within the web
$planet=array("Mercury", "Venus", "Earth", "Mars", "Jupitor", "Saturn", "Uranus", "Neptune", "Average")
Will have average for most dropdown boxes. If someone wants to ask about the planets, but not a specific one, they'll choose the average option.
Characteristic Web:
Phrase-structure grammar:
Determiner (The) Noun phrase (boy), verb phrase (made dinner.)
How many licks till you get to the tootsie roll center of a tootsie pop?
Each characteristic will have a rating on a scale of 1-10 based off of reliability and validity.
Relevant characteristics of variables to interact:
- Force of tongue muscles, specifically at the angle of licking.
- 8 tongue muscles.
- Either intrinsic (act to change the shape of the tongue. Not attached to a bone) or extrinsic (genioglossus, hyoglossus, stylogossus, palatoglossus)
- Muscles supplied by the hypoglossal nerve, except for palatoglossus.
- Texture of outside of tongue/coat of tungue/the thing making contact with the sucker: gastric Qi (white coat).
- Taste buds
- How much surface area of tongue is over sucker.
-How soft the tongue is.
- How many cracks.
-95% water, enzymes, salivary proteins and ions.
-Temperature of tongue

Tootsie Pop- Made of sugar, corn syrup, condensed milk, artificial cocoa flavoring whey, soy lechithin, orange extract, partially hydrogenated soybean oil, and other artificial flavors.
- A Tootsie pop is about 2.5 cm, with .8 till the center is reached. - How many licks?: Or, How to Estimate Damn Near Anything (Aaron Santos)
- Hardness of lollipop
- Rate of dissolving depends on particle size. The smaller the particle, the larger the surface area, the faster it dissolves.
Will eventually be all computerized (sensors from smart phones), for now, like Wikipedia, others can input dat from scientists.
Other programs out there that do similar things, like Google. To my knowledge, this is the only one focusing on life's deep questions with the long-term goal of explaining everything.

Computers will aid us in every decision we make.
What characteristics will be relevant?
- Force
"Men at some times are masters of their fates,"
- Cassius

We live in the information age. Unfortunately, most of it is wasted.

"The fashionable term now is "Big Data." IBM estimates that we are generating 2.5 quintillion bytes of data each day, more than 90 percent of which was created in the last two years."
Is there a God?
Systematic Knowledge
The more specific the data, the more specific the results.
Aristotle's categorization on steroids.
How will it know which variables and characteristics are relevant?
Can be both characteristics of variables and correlation/causation. Correlation/causation shows what variables are relevant?
Create log book for all web pages created.
Variable + Characteristics Web
What things are isomorphic?
Keep It Simple Stupid!
It looks like the variables it's describing
Show how comes up with Tootsie Pop question. 1-10.
Use metric system
Science easily accessible
We take the relevant characteristics of the tongue and the tootsie pop, the two interacting variables.

The tootsie pop is a candy about 2.5 cm in diamater and .8 cm till the tootsie pop center is reached.
Solubility by water.
- 60 calories
- Total Carbohtdrates 15.0 g
Sugars 10.0g

The tongue
- 98.7 degrees temperature
- 8 muscles.
- Thin film of saliva.
-Salivary gland emit liquid 95% water, enzymes, salivary proteins, and ions
- 1500 milliters (ml) a day
- .77 (total in mouth before swallowing) and 1.07 mL (total in mouth after swallowing) (http://jdr.sagepub.com/content/66/8/1300)
- 70-100 micrometers thick ( Collins LM, Dawes C. The surface area of the adult human mouth and thickness of the salivary film covering the teeth and oral mucosa. J Dent Res 1987;66(8):1300–1302.) (http://jada.ada.org/content/139/suppl_2/18S.full#ref-25)
Mucosa of the tongue (the top/dorsal part with taste buds the interacts with the tootsie pop).
- Lingual Papilla (tastebuds)- includes conical, filiform, fungiform, and vallate
-Surface area-(http://jdr.sagepub.com/content/66/8/1300)
- Pressure
Tongue (Pressure from tongue + temperature + amount of saliva) + Tootsie Pop (amount needs to be dissolved + how soluble it is)= amount per lick removed.
Maybe main point of the thing is the web. Introduce that first. Then describe inputting data and asking a question. Then advantages
Types of questions and how it'd be answered
- Does something exist? Ex. Is there a God?
- What the world would look like if it were true.
- Did something happen? JS story
Steps: input question. data web, equation (relationship) output answer
Data from existing scientific knowledge.
New data
Give suggestions
<iframe width="560" height="315" src="http://www.youtube.com/embed/sf9pxFW6I6o" frameborder="0" allowfullscreen></iframe>
Youtube Vidjo
Which smartphone should I buy:
1. Look at all datum of smart phones and come up with a score for each aspect.
2. Weigh importance of different data.
3. Multiply score of each aspect with weight.
How answer different types of questions
Input Data
Smoking gun evidences: Contradictory evidence will automatically falsify a theory.

If comparing things to make a decision, for some people, a school costing over $20,000 is eliminated.
Which car should I buy?
Relevant factors: Cost, maintenance costs, insurance costs,
Bayes's Theorem
If the data is 1, then the answer is _.

If the data is 2, then the answer is _.

Do this for all numbers.
Modus ponens argument
Take probability of something happening times payout if you do. Compare to other choices.
80% chance * 25 = $20
40% chance *55 = $22
Is there an algorithm that can answer every question instead of a separate algorithm for each type of question?
Apply induction/deduction/syllogisms (500 different combinations)learned through 375 ch 12. Use data web as one premise. Create other premises.
Use yahoo answers
How do I compare apples and oranges?
- The fewer the exceptions in a theory, the better the theory (Occum's razor)
- If someone didn't say something on a topic, what does it say about what they thought about it?
After SM
Code (C++)
using namespace std;
int main()
cout<<"Ask your question:";
int question;
cin>> question;
return 0;
Questions to throw into the Logic Machine
Comparison of different ideologies
Contents [hide]
1 Anarchism
1.1 Anarchism without adjectives
1.2 Environmentalist anarchism
1.3 Individualist anarchism
1.4 Religious anarchism
1.5 Social anarchism
1.5.1 Anarchist communism
1.6 Other
2 Communism
2.1 Marxism
2.1.1 Left Communism
2.1.2 Leninism
2.2 Non marxist communisms
3 Conservatism
3.1 General
3.2 Regional variants
3.3 Other
4 Environmentalism
5 Feminism
5.1 General
5.2 Religious feminism
6 Liberalism
6.1 General
6.2 Libertarianism
6.2.1 Libertarian socialism
6.2.2 Right libertarianism
6.3 Radicalism
6.4 Other
7 Nationalism
7.1 General
7.2 Fascism
7.3 Regional variants
7.4 Zionism
7.5 Unification movements
7.6 Other
8 Religion
8.1 General
8.2 Buddhism
8.3 Christianity
8.4 Hinduism
8.5 Islam
8.6 Judaism
8.7 Mormonism
8.8 Sikhism
9 Socialism
9.1 General
9.2 Libertarian socialism
9.2.1 Libertarian Marxism
9.2.2 Social anarchism
9.3 Reformist socialism
9.3.1 Democratic socialism
9.3.2 Social Democracy
9.3.3 Regional variants
9.3.4 Religious socialism
9.3.5 Other
9.4 Revolutionary Socialism
9.4.1 General
9.4.2 Marxism
9.4.3 Anarchism
9.4.4 Other
Mormon Psuedocode
Answer to a question based on statements from church leaders (based on how recent statement is, height of the position of leader and number of church leaders that concur), the cannonized scriptures, and spirit (cannot compute at this point But a reminder that it's important.
- Translate every concept into relevant characteristics, or implications.
- Make these rules (operational definitions) able to tell which circumstances they apply to.
There's a finite number of variables and possibilities from variables, which is why history is always repeating. Could SM look for patterns in data and make assumptions from that. Ex.

and rules from that. Ex. If something works...?
Vision Machine
How transduction occurs
Isomerization- a photon of light lands on the retinal, changing its shape.
Yahoo Answers
Data Web possibilities
Dewey Decimal System
Yahoo Answers Categories
Arts & Humanities Beauty & Style Business & Finance Cars & Transportation Computers & Internet Consumer Electronics Dining Out Education & Reference Entertainment & Music Environment Family & Relationships Food & Drink Games & Recreation Health Home & Garden Local Businesses News & Events Pets Politics & Government Pregnancy & Parenting Science & Mathematics Social Science Society & Culture Sports Travel
Yahoo Answers
Books & Authors
Types of Questions
*Sometimes people leave out the questions word as in, "Last names for Olivia?"
What should I ... name my daughter
What... was in a book, are some suggestions , is the first step, is the definition of
Is... this ok, something older than something else,
Are there (implied: what are they)... books about
Where can I find... a specific website,
How often should I...
8:53 1/19/13
When defining a religion: what is taught to members.
Every premise will have a truth score out of 100.
How will it choose which premises are relevant?
1: Insert theory; Does revelation in the LDS church exist?
2. What would the world look like if this were true?
Other People
LDS People
# of statements
or Logic Tree, Logic Pyramid, Logic Machine, Science Machine, Reason, the Future Machine, Prediction Machine, Screw You Guys; We'll Do it Ourselves, Deep Thought, Reasoning Machine, Decision Machine, What's Better?, knowledge, Masters (from the quote), wiki Decisions, Decision-pedia, smart decisions, Calculated Decision, Science Decisions, or CIRUS
This is the one I believe
Differences in religion: a mixture of input from God plus our own interpretation.
We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes.
—Pierre Simon Laplace, A Philosophical Essay on Probabilities[3]
It's all about facts and the relationship among facts
God wouldn't be immoral
Is God guiding the LDS church?
First, question needs to be specific and what the user is looking for (accurate).
- Accurate: Subtle changes in the meaning behind the question can dramatically effect the answer.
- Specificity: question needs to be specific enough (for example, is the LDS church true could mean does the LDS church exist or are its claims about the world correct. I ended up changing to "Is God guiding the LDS church?". Perhaps to make it specific enough, program can present the user with clarifying questions. (i.e., by true do you mean ___ or ____?). What the questions aims to find out needs to be specific enough to tell the program when defining the variable in question what things will be relevant in its definition.

Second, the program needs to understand the question, or give it meaning. The program can be divided into variables and relationship.
- First, figure out relationship. X guiding Y means x tells y what to do.

-Second, define objects (x & y) or give them meaning. Can't have a "complete" definition of an object. A complete definition would be the thing itself. It's got to be determined by the type of question being asked.

Define them based off of the world aggregate definition of them (but the user can choose to narrow the definition to smaller group (e.g., just use Christian's definition of God or the user's definition of God).

Will need to define variables by data we won't necessarily know we'll need.

-Third, look at tense. In our example, "Is" implies present.
Step 1: Specify what the question means.

- Define initial question.
- Second, define variable based off of what's relevant to what you want to find out.
My definition of God: God is Omniscient (all-knowing), Omnipotent (all-powerful), and Omnipresent (everywhere at once).
Defining questions: use pictures when defining things and symbols to express relationships.
Perfect love.

"For behold, this is my work and my glory- to bring to pass the immortality and eternal life of man"

A body of flesh and bone.

Our Father.

Communicates directly infrequently (baptism of Jesus, transfiguration, witness to Nephites and Lamanites and in Joseph Smith's first vision.

He works through a mediator, his son.
-Stephen E. Robinson (http://eom.byu.edu/index.php/God_the_Father)

Only when clothed in the spirit can anyone endure presence of God, otherwise light and heat that can't be withstood by humans.

-Rodney Turner (http://eom.byu.edu/index.php/God_the_Father)
Probably Important Variables
Time: Is, was, will be will determine result
Now that the program understands the questions, it starts using arguments to find an answer. This will probably be different depending on the type of question.
Step 2: Find an answer
The church started in 1820 by Joseph Smith.
If (X) were guiding the LDS church (Y), Y's path would look different than other similar organizations presumed God is not guiding. Presumably, Y would be more like X.
Learn about Bayes Theorom
Possible Routes
Compiles data. Finds unexpected differences between groups thought to be similar.
- Able to tell people what things they should do to produce a certain result.
You input your predictions. It gives you insight into what predictions you're missing.
Employ different lines of reasoning.
Can we look at beliefs through brain imaging techniques? This is what I would like to study.
How much of our mind is devoted to beliefs?
My Self-Evident Truths
Storing and weighing premises

First, given a truth score. This is based on its validity and reliability based on lower level premises.
Second, a separate, different strength scores are assigned determined by the different types of premises it could be used to support.
Truth = matches reality
The leaders wouldn't follow the immoral trends of the day.
blacks and the priesthood
Priesthood and blacks (the lineal restrictions of the Aaronic (Levitical) priesthood were lifted when the law of Moses was fulfilled, and thereafter the offices of the Priesthood were conferred upon worthy men without limitation to the tribe of Lehi.
-Perhaps, for simplicity's sake, the premises which are super weak are discarded (like brain elasticity).
- Can you automate connecting data to conclusions?
- Premises are made stronger or weaker depending on how they match reality.
- Instead of multipying by weight, multiply by strength. Makes more sense.
- Because it's not practical to run every possible theory, maybe wait to run numbers until specifically asked about a theory.
- Doesn't have to be perfect for now, just better than what's already out there.
- No microcosm is a perfect representation of the things it's representing, so my microcosm of the world will not be 100%, however, there's enough redunancy in the world, that's it'll be almost the same as if it was the whole thing. i.e., don't need to represent every animal from every species, just all the species.
-Is there a max for how many variables a computer can manipulate? Wikipedia has 4 million entries. Would you be able to create a variable for every entry?

How do I try it?

Is it possible for the computer to mess up and delete everything?

Will produce novel insights.
How will sm know what information is relevant?

Difference between hypothesis and theories is theories are premises that can relevant to (change) more premises.

If you dig deep enough into any theory, you find shit. After the first couple layers of premises, you can no longer find justifiable reasons to do something. It's shit. Everything's shit.

How small can break premises up: what are the least common denomiator/ prime numbers of different variables?

You won't need to do experiments to see how one variable effects another variable to find causal relationships if have all variables mapped out.

If starting from the ground up, it will take a lot of premises. Most of our premises our unconscious socialization.

Lots of different categories of questions. Not just the three.

what will the ultimate goal be? If can explain every behavior by a few specific variables (time, space, ?), it will be fully practical, but won't neccesarily say anything about what is the original cause of everything. In other words, being able to predict every behavior is separete from finding the reason for the beahvior??????Find the proximal (closest) cause

The problem is there's too much data in the world to process everything. Solution: put your money on reasoning as opposed to experience/data/evidence. If reasoning is good enough, all you need is one piece of evidence, i.e., one grain of sand, and you should be able to tell all the truths in the universe from it. Because everything that's existed will have effected that grain of sand in some way.

Right now in science, data is collected after a hypothesis is created by humans. The bottleneck is humans. Sm would bypass that. The data would be collected firsta dn then the sm would form hypothesis from that.

Because value judgments are subjective. Leave it up to the individual. Either ask them or base it off of prior decisions they've made.

The problem is that in order to code anything, you have to do it 1000 times. So, I need to do the sm process 1000 times before I can automate it.
Weighted sum of inputs with a threshold
Each premise has a seperate reliability score and validity score. I guess both weighed the same.
Lanaus? decided to classify things according to the importance to man.

One advantage to numbers if you'll never run out.

Another is that they're already in the form ready to have equations to run them.

Also, if someone gives you the name, you can already know exactly what it is, even if you'd never heard of it before.
I'll need other people.
Build a prototype or demonstrate it can work. Need evidence to show people.

How to define things. Things are defined based on context (comparison to other things). It doesn't make sense to define a table as made from wood as its defining characteristic, but it would if everything else was made ouf of metal.

- What things will useful to define things as with SM?

Have a number of different functions. One that takes a newly accepted premise and sees how it would change other premises. Then, maybe those other premises would be compared against reality.
App: Knows what situation you're in in life and dispenses scientific information relevant to you. Depending on your job, work, what time of day it is.
GPS for life. Constantly collecting your data.
Just answer the questions, "what should I do?" Has the potential to be able to be able to build one prototype at a time. Ex. First, just look at biggest problems in area: death, sickness, etc. Next, can start calculating those premises.
Every thing has more than one meaning. A marker is a marker and a throwing object.
Look up connectionist and network models
Top-down vs. Bottom-up
discover a way for a computer to run through all the different possibilities for a theory. Is there a managble way to do that? Only so many possibilities?
Look for evidence that
violates current theories.
cout << What premises you want to accept?
Depending on repsponse, you should believe this.

Can create this now
You are not so smart David McRaney
a bunch of equations. math

Modify it so it fills in your calendar for you depending on all the possible things for you to do that you entered and when there's conflicts, it puts the best one in for you.
Symbolic Systems
Maybe I need to start using stats and probability in my own life before I'm able to turn it into a computer program.
do SM for card game
The purpose is simple: organize and manipulate all knowledge. The execution is not.
We name not only species, but every time we want to distinguish something from something else. i.e., every time we save a file under a specific name.
The Analytical Engine - Babbage
Ideally, the language would come from brain signals
Religion is within people.
People in the community of Christ church feel like they feel the spirit too.
Fundamental Data
Time, Space, Material
Classification: mine needs to be the first that allows for future unknown categories.

We can't assign everything just one number because where something would lie on a spectrum depends on which spectrum is being used. Bread would have a high number on the nutrituion spectrum but a low number on the weapon spectrum. Every variable would be assigned a different number for each relevant spectrum. So, it will actually be a combination of numbers.

The variables's numbers are not assigned based on comparison of other variables, but on their raw data.

What's Better?

cin >> Who's better?
What's better?
Where's better?
When's better?

How to decide what variables are relevant: Based off of "at what"
>> everything and their variables (i.e., height, weight, IQ, criminal record
cin >> "at what"
i.e., being president, soccer player, leadership, sports, team captain
"at what" determines which variables to use
What's better?
Sometimes science doesn't even know the right question to ask. I.e., auditory cortex, phrenology. Can the Science Machine automate the questions?

Different somatosenes in body: exteroceptive (sight, hearing, touch, taste, smell), proprioceptive (position of the body (comes from receptors in muscles, joints and organs) and interoceptive system. SM would not need pro or intero. Could use exteroceptive + thousands more.

- When science doesn't even know the right question to ask. e.g., The auditory cortex phrenology. They were asking what are the differences between different races? Can SM know the right questions to ask?

- Information is surprise. SM must give the user something they don't already know.

- Maybe some questions don't need to be answered. I.e., how many angles on the pin of a needle.

- Decisions are central to thinking. That's what seperates animals from plants. Animals or do other things that indicate they are making decisions.

- Decision machine will combine all mathematical knowledge in all subdivisions of math. For example, to know if can jump over something requires geometry and ...

- Maybe have certain options be a function that can influence other options.

What are the footprints?

Emotions aren;t just influenced by our logic. Our mood determines our logic.

"Light et al. (1990) once said "You can't fix by analysis what you bungled by design"

Information age forces us to realize that our reasoning powers are limitedd because every argument has smart people on both sides each with dozens of seemingly plausible arguments.
Sensory information feeds conscious and unconscious mind

Study: Use Wiki Decisions to see how often people guess the correct decision.

My self-evident truths are the code.
F it! We'll do it live!
ND Y TX N Cost 3 Y Location 2 Y Prestige 10 N 100
200 10000 20000 7 8

M Y B Y W N Cost 10 Y Proximity 10 Y Taste 10 Y garbage 10 N
0 5 10 0 5 10 0 5 10 0 5 10
Example: ND TX
Cost (3), Location(2), Prestige (10)
ND: cost- 100 Location- 10,000 Prestige- 7
TX: cost- 200 Location 20,000 Prestige – 8
ND: 1.33 TX: 1.866
Total for Standardization: cost 65000, location 55,000 prestige= 19
Standardized ND: cost- .46 location- .45 prestige .52 = 1.43
Standized TX: cost - .54 Location- .55 Presitge -.48 = 1.57
m Y b Y w N
c 10 Y p 10 N

10/.5 10/.5

C: M 1, B 5, W 10
P: M 1, B 5, W 10

results: cost m:[.0625] c b:[.3125] c w:[.625] prox m:[.0625]
p b: [.3125] w:[.625]

results: [.03125][.15625][.3125][.03125][.15625][.3125]

total_option: m:[.0625] b:[.3125] w:[.625]

Output: W B B

You should go to TX!

m Y b Y w N
c Y 10 Y p Y 10 N

10/.5 10/.5

C: M 1, B 5, W 10
P: M 1, B 5, W 10

results: cost m:[.0625] c b:[.3125] c w:[.625] prox m:[.0625]
p b: [.3125] w:[.625]

results: [.03125][.15625][.3125][.03125][.15625][.3125]

total_option: m:[.0625] b:[.3125] w:[.625]

Output: W B B'

a Y b Y c N
d N 10 N

d: a5 b5 c0
total_options [.5][.5][0]
a b a b c
Wiki decisions
Enter Data
What is the reason you gave ___ this score on ____ ?
User can hover or click on each layer to find the reasons behind the automated score (e.g., avg from users input or this person inputted it in on the enter data page)
smart decisions
Simple like google. Least amount of information as possible
Enter Data
Add a new option
Every time they add a new option, another "Add a new option" box appears
Option names:
Add a new option
Option Attributes:
After they enter options, give them best suggestion of what you think it is. Give name and definition (scores on its relevant attributes).
If the sum of scores is negative, standardization wont work. It'll reward the negative numbers. Are any negatives ok?

User can hover or click on each layer to find the reasons behind the automated score (e.g., avg from users input or this person inputted it in on the enter data page)
Option Scores:
If they try to click next before they have any options, give them an error message
Option names:
They click up here to edit options
- After 1st option, goes through all options. If it finds it, suggests the ones in that vector.
- Store what options go together.
Also store how many people start but don't finish. Just create a start variable and an end variable and look at the difference. Hopefully this improves as it becomes more automated.
Ultimate goal would just be that the user puts in options, next, next and then have their winners
- Will know if its subjective if the score are widely different
Maybe give the user the chance (percentage) that the data you're giving them is correct.

- Cross loadings had too much of an impact because differences are often large between best and worst compared to like CFA. Maybe incorporate SD. Either that or start all Mods at like 5. So instead of 1 and 2, it'd be 6 and 7.
- Some data can are the level where they can be viewed in the world. Others might have lower premises E.g., safety.

- Classification: How will you keep people from entering things with different names that are actually the same thing?

- Two goals of classification: 1) speed (as little steps as possible)f for the user to find the option they’re looking for. 2) As all-encompassing as possible.

McDonalds won the last one. Was it right?
From best to worst:
Was it right?

Was the data wrong on a variable or did you forget to add a variable?
- Most variables are weighted. But some variables have a “Smoking Gun” power where they negate everything else. Allow user to put in smoking gun. Allow for a smoking gun (i.e., if Mom would be willing to make mashed potatoes) Just make the number big enough to outweigh everything. =

- //I'm tring to make it accept a line at a time
//cin.getline (compare_input, "\n");

- Look through prezi to find names for things.

- Allow for equal scores

- Simplify by creating functions
- // Use tolower to make first letter lowercase. Something like this, but it's not right: char ch = tolower( variables[i] = tolower (variables[i]) In “On a scale of 1-10, how important is Funny to you?,” make funny lowercase

- Can results be turned into a 2d vector?

- Am I content with the word attributes? Maybe "Properties of the system" - spence. Instead.

- Change variable names to thisWay

- Fix reverse scoring

- Just do the basic code on javascript.

Back up database before making changes.
How do they pinpoint what data to add? Start with library codes. Zoom in to deeper and deeper categories. or they can just use a search box. Or just let them search by name (option) or definition (attributes).
you can use to make decisions. and data will be used by others in their decisions.The more people use it, the better it gets

Anything that the computer can't do, we allow humans to.
Inform them what data needs to be researched based on either questions people are asking or contradictory evidence.
What would be need to be logged in to do?
Recommend attributes based off of what attributes all options have in common.
allow upload of text for data. A lot of data from research is in that form. Make it easy for them to input it.
This works
10/.5 10/.5

MC cost: 1 BK cost10
MC prox:1 BK co: 10

results: [.1][.9][.1][.9]

output [.1][.9]
output_copy [.1][.9]

comments or suggestions: jacobdeecurtis@gmail.com
Attributes: specify. E.g., instead of time, say years
After they select the attribute, rank on scale 1-10. Automate suggestion based on how other people ranked it.
can add picture
Burger King
After they type first option, autopopulate best guess of what they mean.
if checked, recommend other options
Create an Account
Set "Options to fixed"
- All attributes all the options have in common.
- Won't work perfectly because there's more than one thing a person would want to do with a particular set of options. For example, someone might list McDonalds and BurgerKing if they're looking for a place to eat but also if they're looking for a restaurant franchise to buy. So, a common variable could be net income but someone looking for a place to eat won't care about that. A farmer doesn't care how ugly a cat is as long as it catches mice.
- One possible fix would be to have the user input the question before the option names.

How many examples? 1, ___, all of them?
if more than one: mean, median or mode?
From: website, scientific article, own eyes?
Have the recommended attributes in a pile

Maybe just start with automating one thing at a time.

Make it so users can go back if they made a mistake

- Create functions (asking for input, standardization)

- add: if cin fails, reask.

- Make it so the user can keep putting in strings as long as they want and end it with "end" or something.

Go deeper. Get rid of subjectivity. Don't allow the user to input how important something is to them. Calculate importance based off of ...
Subjectivity is leaving it up to the user and the user is dumb. It condones illogic.
Factors that go into how important an attribute is to me.
- Cost is higher if I don't have a lot of money.
- If options have very similar scores on an attribute, that attribute is less important. (i.e., Sam with taste between Mcdonalds and Burger King
string optionNameInput;
Cout<<"What are the options?(end with 0).
while (cin>> optionNameInput) {
- Make DM a repository where people can input data from any source (e.g., scientific papers, their own efforts, etc).

- Create a separate collection that holds all the options with their accompanying attributes, importance and scores.
- What gets updated in this table when a user makes a decision? What can a user update when "Entering data"?
- Is possble to predict a score from a sliding scale if there's another compareGroup with that option and another option from the current compare group.
- Old: MCD taste= 7, BK taste = 1; New: MCD taste= 4, BK = ?. 7/4 = 1/x = 4/7

- 1st, if the people only said it was wrong. Use that information to compare to those who said they were right and spot the differences.
What would you change?
It'll know the difference between objective and subjective data because subjective data will have more variability. Can find a specific cut off based on looking at the variability of known subjective attributes (taste). It will also distinguish between variability because of subjectiveness to variability because data changes often (e.g., NBA rankings) based on whether the scores are consistent for a period from everyone (NBA rankings).
-Maybe for the app, have it so people can enter scores differently from their phones and the result will be averaged between them.
Nothing is in the understanding, which was not first in the senses - Francis Bacon
To find: Users input Option, then attribute, then unit of measurement. Tell users if they think people could confuse what they're option is to be more specific. E.g., Instead of "chicken," search for "chicken at Texas Roud House."

One way to tell if an option is the same as another option with a different spelling is if they have the same attributes with the same scores. Calculate confidence level. Specify alpha (.05) and H0
H0 is always that they're the same thing.
If not found:

If found:
Things we'll want to do with "Enter Data" data
Cause machine: See correlations of attributes. I.e., cost goes up as proximity goes up to know what attributes causes other things to increase. How will this be useful? I need to think about this more.
Just like Mplus does with CFA or SPSS with EFA, have DM tell me if options/ attributes are actually the same thing.
Science Machine

The more you use it, the smarter it gets
- Placeholder. Light gray text in input box, "Option1"
Can see value when using sliders
Calculate mean if n is big enough

To do confidence level, has to be random. Is this random? N also has to be large. Needs to be able to be plotted and look normal.
Make it just like moving shapes around

DM: make the user interface intuitive. Somehow model what happens/ what we see when we make decisions

DM: Maybe make colors get progresively darker
DM: Output results using bar graphs. Make bar graphs equal in width.

DM: knowledge representation: the study of how to put knowledge in a form that a computer can reason with.
DM website: use www.colorcombos.com

DM standard width was 960px and now it is 1200px

DM- protect your data
If a distribution is skewed, use median instead of mean. Can tell if skewed if large difference between min to median and max to median.

DM: Average or the mean won't be super useful for bimodal distributions. Would want the average of the two peaks.
Use phonegap to turn mobile thing into app.
DM WS: if the user clicks on the thumbs, switch them and reverse score.
Theory Machine: Comes up with, falsifies or proves theories.
Decision Machine Experiment
Hypothesis: people use DM algorithm when deciding which decisions are best and are bad at it when there are too many comparers or qualities. Purpose: 1) To find out if people do use the DM algorithm to make decisions
2) If they do, how good at it are they? At what point, do they stop making good decisions?

Just ask a question right before show results which they think is better. Use the Decision Machine which they decision they think is best.
- Qualtrics, app or website?
- What I'd expect to see is that they'd be correct with
- Put a timestamp on everything.
- Find the difference between smart people and dumb people when they choose what's right.
- Difference in number of options or qualities?
- Difference in quality of qualities?
- extrinsic/intrinsic?

To test 2: Give them fake qualities compares and compare scores-- has to be fake otherwise . Randomly give them different compare groups and have them choose which they think is best. Hypothesis: people would follow DM when a few attributes or items but
Disadvantage of doing it this way is that maybe mock options and qualities don't mirror how we make decisions with things we know about.
Or actually, maybe the priming doesn't matter because it'll be the same for both people. I don't think there's a reason for why people with ability to hold lots of qualites and options would be more successceptal to priming. Maybe still do the priming test to show that there's no difference between the two groups. Still a problem for testing 1.

Method: Demographics: Ask years of education (to make it continuous)l, age and maybe some way to measure intelligence, sex

Truth Machine
Many decisions depend on a quality score being true or not. E.g., whether I should go out to eat or not depends on if I'll actually do work if I don't go. Allow the user to input
Not Going Out to Eat's score on productivity towards future
If condition is met, what's the score? 10/10
If condition is not met, what's the score? 3/10

What're the odds it happens? 4/6
Condition: I'll actually study.
What'o I do with the input? Just weight give a weighted (based on the odds) score of the average 2 scores. So, (10/10 * 4/6) + (3/10 * 2/6) = .777

For now just have the user enter this, but it'd be fun to make it do it on it's own. This would be the truth machine.
What are the different types of evidences/footprints? Can I categorize them?
"Compared to the other attributes, how important is..."
Which is better: [smallest] [quality] or
[largest] [quality]?
DM C++: Dynamically create vectors by just having like 15 by default and only use the ones used.

DM C++: I could just have all the possible indicators I could think of and whatever ones they don't think are important, they can value at zero. It will also have the option to add a new one.
Measure twice; cut once (I don't want to end up like the Stats camp campers)
Maybe see how people's qualities they care about and care the most about correlate with Maslo's Hierarchy of needs.

See if different qualities were more important to different people based on demographics

Not just the more the better (linear)
Also on/off
When someone asks a question, go through all options. If an option doesn't have relevant qualities, kick it out. Just an if statement.
Could be non-continuous qualities that matter. How incorporate that?
How do we incorporate non-continuous data? Do we turn everything into continuous?
An infinite number of possibilities. How do we figure out what those are and limit those.
I don't believe in free will.
I don't believe in free will.
I don't believe in free will.
I don't believe in free will.
Ideal is .7 (e.g., CFA)
Anywhere from .3 to .5 is ok. Average the goodness. The farther it gets from that, the worse it is.
Maybe instead of focusing on those, focus on what the brain uses (what, where)
copula modeling
How automate options: thinking of differnt options than usual is thinking outside the box
After a certain point, it doesn't really matter.
e.g., Nokes said once you get to 90 degrees, hot is hot. No difference between 90 and 100 degrees.
How to tell that certain options make up other options. e.g., Sue Curtis + John Curtis + Kirsten Curtis + Zane Curtis + ... = Curtis family
- Know they're part of a group if they move together. In different circumstances, as one of their attributes move one direction, the others do as well.
What should and shouldn't be included in importance scores? How similar a quality's score on a quality? CIRUS is taking that into account right now.
Eventually enter what they want in an option and DM would decide which attributes are relevant for them based on correlation with success.
wikidecisions website: People get a truth score. If stuff they enter is confirmed, they get more points.
wiki website: after people input data, "here's the implication of your data:"
You can't compare apples and oranges because they have different qualities.
Cause Machine
"A cause is that which makes any other thing, either a simple idea, substance, or mode, begin to be... and an effect is that, which had its beginning from some other thing"- John Locke
To determine causation:
1. temporal precedence. Cause happens before effect
2. Covariation
Type 1 Error- think covary. They really don't
Type 2 Error - don't think covary. Really do.
3. No plausible alternative.
halo effect
Allow for them to say they're the same.
Link between success and judgement.
Other factors are important. e.g., how good people are at following through on what they think is right.

Item stems should be short, unambiguous, simple containing one proposition, should avoid negatives, use simple and conventional vocab.
inus condition - an insufficient but nonredundant part of an unnecessary but sufficient condition (match starting a forest fire)
-insufficient because match alone can't do it. Needs other things like oxygen.
-Nonredundant only if adds something fire-promoting that is different from what other factors (oxygen, dry leaves) contribute to the fire.
-It's part of starting a fire in combination with the full constellation of factors. But not necessary because there are other conditions that could also start fires (lighter).

3 causes of correlation
1. One causes the other
2. Extraneous variable causes both
3. Shared components (e.g., with AMSEX parents stuck together)
Hypothesis: Ability to juggle lots of qualities and options leads to better decision making.
Is it possible better decision making causes ability to juggle lots of qualites and options? No
Is it possible better decision making and ability to juggle logs of qualities and options are caused by the same thing? No

Truth Machine
Figuring out what's true
Feeds into <-
1. Correspondence Theory: A knowledge claim is true if it confirms to the world (It is raining is confirmed by looking outisde; Causal book p. 35)
2. Coherence Theory: A claim is true if it belongs to a coherent set of claims (I think just based on solid premises)
3. Pragmatism: Claim is true if useful to believe it. (say electrons exist because it brings predictability or meaning.
Albert Einstein once said, "Thinking without the positing of categories and concepts in general would be as impossible as breathing in a vacuum" Research cannot be done without constructs - Causal book p. 65

Assigning qualities to options
"This difficulty arises in part because deciding which features are protypical depends on the context in which the construct (option) is being used. For example, it is not that scientists are right and laypersons are wrong about how they classify trees. To a layperson who is considering buying a house on a large lot with many trees, the fact that the trees are deciuous means that substantial annual fall leaf cleanup expenses will be incurred."
Another way is to find out the qualities that usually go into making a decision. Can be something simple, e.g., holding the door open for someone. Observe. Treatment (Change the calculus enough to rebalance the equation.) then observe again.
My study is only tapping conscious judgements.
How much is unconscious and how much conscious?
Are most decisions made unconsciously?
Does my judgment model hold up for unconscious decisions?
New thought: I think it applies to conscience and unconscience decisions. I wasn't sad when Britt broke up with me without realizing why. It wasn't till later, that I pulled the reasons out of my unconscience.
Control for the method effect. Do it with different methods. One online, one watching behavior.

DM ability as an indicator of intelligence - How does fluid and crystalized inteligence have to do with this?

Background research: The magic number 7 (we can only hold 7 things in our memory at once)
The DM process (specifying attributes, standardizing, weighing, adding to come up with a total score) is not just for making decisions. The same method determined whether I was sad or not when Britt said she didn't want to be my friend anymore. I wasn't sad because 1) clearness of conscience was high and 2) my desire to continue to be her friend was low.
Potential programming project:
Generate questions by randomly combining 5 Ws (who
what, where, when, why) with nouns, adjectives, etc.
Point users to where data collection needs to be based off of attributes for options people are using in the machne but don't have data yet.
Store Xs and Ys. Find slope: y = mx + b to predict certain things.
Type 1 error: find innocent person guilty
Type 2 error: find guilty person not guilty - would rather have this one.

"The earliest known reference relating to Multiple Criteria Decision Making can be traced to Benjamin Franklin (1706 1790), who allegedly had a simple paper system for deciding important issues. Take a sheet of paper. On one side, write the arguments in favor of a decision; on the other side, write the arguments against. Strike out arguments on each side of the paper that are relatively of equal importance. When all the arguments on one side are struck out, the side which has the remaining arguments is the side of the argument that should be supported. Supposedly Franklin used this in making important decisions." - http://www.mcdmsociety.org/facts.html
Nonfoundationalist - Truth is not absolute. Not a mountain you can climb to the top of. Only better or worse than other theories. Truth machine will be a constantly changing mountain.
"The human brain is a relatively inefficient device for noticing, selecting, categorizing,
recording, retaining, retrieving, and manipulating information for inferential purposes. " - Grove and Meehl Clinical-Statistical Controversy
Surely we all know that the human brain is poor at weighting and computing.
When you check out at a supermarket, you don’t eyeball the heap of purchases
and say to the clerk, “Well it looks to me as if it’s about $17.00 worth; what do
you think?” The clerk adds it up. (Meehl, 1986a, p. 372)

Decision Machine App
Long term goal of the Decision Machine. The DM will give advice on what to do without the user having to ask.
Fun with the data
By comparing values for a person over different types of decisions can see what context change values. Also over time (e.g., cost becomes more important at a later point when I was broke).

If eventually ask demographic questions, find out what attributes are different in ppl with a particular disorder.

DM calendar: The user inputs all the things they need to do, either lots at once (eg importing class syllabus) or one at a time (just realized I should clenan my room). Calendar will then populate calendar recommnedations for them. If things conflict, it will choose the best option.
Will the things people need to predict be included in the database? Will liklihood of cheating be an attribute for someone else. If so, can see differences in scores between when that is high vs low.
One of the self-evident truths is that things cannot have mutually exclusive values. Grass cannot be more than one color. Either green or blue. Although, I'll tell you what, quantum mechanics might shit on this rule.
Evidence can't prove any one theory. All it can do is rule out certain principles. Evid
Full transcript