Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Good ideas at the time.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Truth Graph

  [vote for,

Every since finding out that there are people who genuinely believe that earth is flat, the question "how do we know what's true" has been on my mind.

A ridiculous claim like earth is flat could be dismissed with: "crazy people believe crazy things". While the question "is earth flat?" is boring, I am very fascinated by the deeper question "how do we know what is true?" What are those mechanisms? How can those mechanisms be used without being abused? Also how do we measure what is true?

One such mechanism is a truth graph. A statement such as "the earth is flat" is backed up by connected statements such as "you can observe curvature of the horizon from a high flying aircraft" (and 1000's others). These statements are then linked to others to form a graph. Statements that have few links connecting to them are on shaky ground. They may, or may not be true. Other statements will be "pillars" that hold up millions of other statements. Those pillar statements will be harder to unseat, that is until scientists build a sufficiently strongly inter linked foundation beside it to become dominant.

Currently we do all this inside our heads, where it's the neurons connecting to each other and making sure that at all times our ideas are consistent with each other and with observed reality. But there are so many facets of observed reality these days that unaided human brains are having trouble keeping up. Truth graph is a way to get that process out of our heads and into a common framework we can work on collaboratively.

Collaboration also means competition. Flat earthers would be free to construct their truth graph right along round earthers, but my prediction is that the round earth graph would be 100,000x more richly interconnected than the flat earth one.

Now let's take something less clear cut. Let's say two competing quantum theories. There may be one that's dominant, but as evidence for the other mounts, it may be possible to cause a dominance flip. Or, if we go back in time to the statement "sun orbits the earth" even then a truth graph would be useful. Sure, in the beginning the truth graph of "sun orbits the earth" would be pretty sturdy... it would have all sorts of observations that support that fact... however, eventually it would be completely eclipsed by a much more richly interconnected idea which is "the earth orbits the sun" theories built on it, it would be an unstoppable snowball. The statement that "sun orbits the earth" would not die because at minimum it would be linked to the observation "sun appears to orbit the earth every time it rises and sets", but it would be so small (relatively speaking) that it would not be useful any more - other than to maybe understand what misconceptions people had back in the old times.

ixnaum, Dec 12 2019

Knowledge Graphs https://arxiv.org/pdf/1708.07239.pdf
[xaviergisz, Dec 12 2019]

Fall by Neal Stephenson wiki https://en.wikipedi...;_or,_Dodge_in_Hell
[theircompetitor, Dec 17 2019]


       Well, we have to give you a bun for your audacity in attacking one of the most fundamental problems posed in Metaphysics.   


       <Steps aside to allow through seething mob of halfbakers brandishing flaming torches and assorted agricultural edged implements with intent to inflict chastisement on [ixnaum] for hubris - presuming he exists/>
8th of 7, Dec 12 2019

       "I think, therefore I am right" - Des Cartes
pocmloc, Dec 12 2019

       read Neal Stephenson's latest (The Fall)
theircompetitor, Dec 12 2019


       //I think therefore I am?   

       //Neal Stephenson's (The Fall)?   

       Obviously I'm coming in this too naive ... because I don't get how any of these are related .... ok maybe tangentially, but I have a feeling like I'm not in the loop. It appears that petacycles of human brain activity have been spent on this already and I'm bit late to the party. Can someone fill me in?
ixnaum, Dec 12 2019

       Certainly; try any bar in Glasgow at 2250 on a Friday night. Just make some random remarks about Celtic, Rangers, Catholics, Protestants or the Independence Referendum - you will get "filled in" in very short order.   

       Cheaper and much quicker than Dignitas, no form filling or medical tests.
8th of 7, Dec 12 2019

       "I think I am right, therefore I am" - Des O'Cartes
pocmloc, Dec 12 2019

       "I am, therefore I am." Des o' Norfolk.
MaxwellBuchanan, Dec 12 2019

       // the claim of consensus has been the first refuge of scoundrels //   

       Odd, we thought it was patriotism - but that's the sort of specious rubbish you get from immigrants and foreigners with no pride in or loyalty to their adopted country.
8th of 7, Dec 12 2019

       //a dominance flip// I tried one this morning. The neighbours seemed quite cowed but my back is hurting now.
pocmloc, Dec 13 2019

       The proposal doesn't lean on a large number of people agreeing, but on a large number of statements agreeing. It's bollocks, but not because it's about consensus.
Voice, Dec 13 2019

       Sounds like the next platform for hypotheses of science! [+]
Mindey, Dec 13 2019

       The weakness that I haven't been able to fix yet is spam. Just like google's pagerank algorith, Truth Graph can also be gamed by non-scientific interests. For example, what if flat earthers put a lot of energy into creating self agreeing statements, while round earthers are too busy doing real work and can't be bothered to enter data into this thing and outcompete the nonsense?   

       The only hope is that scientist would not let it degrade and spend real effort on tending to it - it would be a matter of pride being a scientist. Or even better maybe this tends to itself, because a scientist working on ballistic projectile aiming, may make a contribution to the basic building blocks proving roundness of earth even though they are not working on that question explicitly. Maybe it could work the way the way open source libraries build on each other. A library function for TLS encryption will be made "true security" by it's presence in millions of other projects, while ROT13 will be delegated to the "toy encryption" category just by it's lack of interconnectedness within the ecosystem. When TLS library is obsolete by something better, the ecosystem will adjust the meaning of what "true security" is.   

       Maybe Truth Graph has to be linked to engineering as well. That would provide deeper verifiability of truth. If GPS time adjustment works according to Einsteins equations, then that's another link supporting Einstein's equations being true. Once someone comes up with unified theory of everything that supersedes Einstein just like what happened with Newton, and we finally get those hoverboards, those links for the new theory will dwarf the thousands that Einstein's stuff got. Newton isn't wrong, Einstein is more right. (and not just by a bit) Same thing for the next insight.
ixnaum, Dec 13 2019

       // Newton isn't wrong, Einstein is more right. //   

       Newton is right, for a given value of "right", i.e. on a macroscopic scale and where v << c.   

       Newtonian mechanics only really goes goofy at the atomic level, when objects (they might be waves) move at velocities approaching C.   

       Einstein is "differently right". Nothing that Einstein theorised disproves Newton's equations for application to unaided observation. It's not until you have a high-end telescope and seriously good chronometry that it's possible to physically detect gravitational lensing.   

       If Sir Isaac had had access to the Hubble, Arecibo, and a few supercomputers, who knows where your species might be by now ...
8th of 7, Dec 13 2019

       It's very unsatisfactory, isn't it ?   

       The answer is probably some sort of cull.
8th of 7, Dec 14 2019

       //Einstein is "differently right". Nothing that Einstein theorised disproves Newton's equations//   

       Well, no, in fact the opposite. Newton was utterly wrong; Einstein may also be utterly wrong (if and when we come up with a better theory), but the two sets of equations reflect completely different underlying theories.   

       It just so happens that both theories give very similar answers over the narrow range of conditions that we can easily observe. It's like the difference between Y=3X and Y=8X-15 : completely different equations reflecting completely different realities, but they both give the same answer (9) when X=3.
MaxwellBuchanan, Dec 15 2019

       // over the narrow range of conditions that we can easily observe //   

       That should entitle you to some sort of small prize, were it not that it falls outside the very narrow range of conditions for which small prizes are awarded ...   

       "A claim of validity for a theory can reasonably be made if no more than 50% of the data have to be discarded to obtain conformity"
8th of 7, Dec 15 2019

       // validity for a theory can reasonably be made if no more than 50% of the data have to be discarded//   

       you can go as high as 88% if undergrads were involved with the data in any way.
bs0u0155, Dec 16 2019

       That's useful to know.
8th of 7, Dec 16 2019

       Zooming out ( scaling from view perspective) from the vectored network graph might show discontinuous density areas. Places where more data and thought are needed.
wjt, Dec 17 2019

       /Neal Stephenson's (The Fall)? The reason I brought it up for those who haven't read it, it contemplates the ultimate collapse of trust -- the absurd conclusion to "it must be true, I saw it on the Internet" -- and the potential of using a next gen blockchain technology around vetting every piece of information you've ever seen and will see.
theircompetitor, Dec 17 2019


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle