"Abstract Board Game": the game

So I've noticed that in a lot of what are called "abstract board games" like Chess and Go, there's a kind of common theme to the strategy. I've played a lot of abstract board games, and here are my thoughts. It turns out that in a lot of them, you win by making what's known as a "fork." Let's think of "checking" someone in chess, or "putting someone in atari" in Go. In these cases, one player puts the other one step away from a loss, and that other player MUST respond in order to, in chess, not lose immediately, or in Go, not have a piece or pieces captured from the board (chess can also have threats to individual pieces of course.)

A fork is a situation in which there are *two* immediate threats made simultaneously, such that one's opponent cannot respond to both at the same time and so has lost due to having to choose between two losing situations.  This can be checkmate in chess, or, say, "double atari" in Go, where two stones are put in danger of capture simultaneously and only one can be saved.

In one of my favorite abstract board games, Trax, there is a thing called a "corner" which is basically being one step away from check, or one step away from atari (maybe there's terminology for this in chess and Go also, I'm not sure.) The equivalent of check in Trax can be a "loop threat" (the game is all about winning by making a loop of your color before your opponent makes a loop of theirs.)

In Trax, loop threats aren't always that great because they're easy to get out of. Instead, one tries to make lots of corners, since they give leverage later in the game. Maybe you will be able to "upgrade" two corners at the same time, making two attacks simultaneously, or a fork.

But the other player will try and stop forks from happening. So there ends up being more and more layers of strategy. Some corners which are obviously one step away from a fork can be stopped easily in Trax often, so maybe you want to make two forks at the same time! Like, a kind of "double fork" or maybe you want to make two double forks, a quad fork. It gets complicated.

So anyway, I thought to myself, also in chess, if you can see your opponent is about to win, you can stop them. So you want to be a few steps away from winning, but when you can be a few steps back, you can have lots of ways of winning, and maybe your opponent won't see all the ways to stop all of those ways of winning (maybe you won't see a way either.)

So comes my point, what about making a game that abstracts abstract board games? I've been trying to think of a way for a bit. Had some thoughts for an outline of how the game might work. Here's some rules that won't actually work but I think maybe they could be modified to work.

So here's my non working version, but that I think might be on the right track who knows: You literally have "checkmate" and "check" and "two a way from check" etc cards (in this game the cards will just have values which say how many steps away they are from a win. A "zero" card in your hand is an immediate win let's say.) Now I say this should probably be played on a computer because the cards can change values, and there can be a *lot* of cards.

A breif asside: I also like games with infinite potential so I happened to think this might be one which allows for infinite play in a sense. Let me explain. Preliminary thoughts on possible game rules: on one's turn, one can do one of two things. One can either add cards to one's hand, or upgrade a card in one's hand

To upgrade a card in your hand, you may decrease the number on your card by one, and simultaneously increase the numbers of every card of a certain class in their opponents hand.

To add a card to one's hand, one may do something interesting. One may add one card one away from winning, two cards two away from winning, three cards three away from winning, etc. The infinite comes in in the fact that one could in theory add a million cards a million away from winning (in practice I guess you'll have to draw limits somewhere, but one could just keep a list of the number of cards of each type so it could just *say* someone has a million "million away" cards)

Ok I realize now that the rules as I just wrote them down aren't going to work at all I don't think. But something like this might work.

(Very slowly) turning the entire English language into computer code

TL;DR;  I'm very slowly working on defining all of the English language in code.  Here is the link.

So when I was a kid, I thought dictionaries were interesting.  You define a word, and that's defined in terms of other words, which are defined in terms of other words, on and on.  What, I thought to myself, if you could expand a word out and see what it's made of down to the lowest levels?  What is the smallest set of words you need to define other words?  An issue is that dictionaries are often circular.  Life is defined in terms of death, death in terms of life.

Well I really was happy much later in life to discover later in life that I was not the first to think along these lines.  Enter the Natural Semantic Metalanguage.  This is supposedly a list of around 60 words from which any word can be defined down to.  See the link for words, words like "do" and "happen" and "part."  Supposedly?  Did I say supposedly?  Turns out people on a website called Learn These Words First actually made a non-circular dictionary.  They boiled down the definition of 2000 words, in layers, down to the 60.  Meanwhile they selected the 2000 words from an English learner's dictionary that depends on only the 2000 words that the LTWF boiled down further.

This is all super great.  However, I thought to myself, English is such an imprecise language.  What if I could make the definitions look more well defined?  I decided to define them like one would define computer code, very specifically.  I am boiling down the definitions to something similar to a symbolic logic system called predicate logic where one might say something like "all dogs have tails" as "ForAll(x, dog(x) Implies Exists(y, tail(y) And has(x, y)))."  Basically you are making statements about what kinds of things exist, what kinds of things all things have, what things imply other things, things like that.  It's a language with a pretty small number of words in the end, if you can boil down further words like "dog" and "tail" and "has."

So bad news first, I'm nowhere near finishing the project, and it needs work even as it is.  Good news, I've started it, and I feel like sharing it because I just want people to know I'm working on it.  I'm loosely following the LTFW non-circular dictionary posted above (it has been super helpful.)

The code doesn't really do anything right now besides definitions, but I think the definitions are fun to look at.  At some point I do want to make it so you can expand definitions out by clicking or something, or perhaps even make it so that code you write like on(house, hill) would actually draw a generic house on the screen or something.  Some day I might try swapping English with an already more well defined language like Lojban.

I've posted it on GitHub if anyone wants to make suggestions.  Here is a link where you can easily view the code.  I wrote it in Mathematica.  I might at some point convert it to Python or Haskell or even maybe Prolog if that makes things easier for people.  You can use Mathematica online for free now, also someone's working on a free Python interpreter of some of the language and someone else a Haskell version.

Coding in non-editable code

Weird programming language idea.  So in one of my favorite video games, Riven (the sequel to Myst)  there's this thing where people write books that describe places and then they can literally visit those places.  An oddity of this is that they are basically writing code, but since they are literally writing symbols on paper with ink symbols cannot be literally erased.

It is mentioned in the lore somewhere that there's at least a symbol you can add to the end that can cancel out stuff before it but anyway.

The idea then is this:
A language designed to be written once, in ink, where anything you write can no longer be directly erased.  All edits to your code, if you didn't write it correctly in the beginning, must thus simply be added on to the string of symbols already there.

As a bonus!  This could maybe work in the direction of a spoken programming language, since what you say in real life to another person cannot be unsaid.

Explaining self-awareness to non-self aware intelligences

A philosophical zombie is someone who acts exactly as if they are self-aware, but they are in fact not self aware, and the question is, how could you know.

I have a different question.  What if we came upon a new intelligence, let's in this case say it's an alien race.  What if these aliens are INTELLIGENT BUT NOT conscious, and actually KNOW that they are not conscious!  They can talk and write stories and do math and science, but actually, if you ask them if they are self aware, they just look at you funny and ask what you mean, and if you try and explain to them they don't even know what you are talking about.

And how WOULD you explain to them what consciousness means?  You're like, "hey aliens, I know I exist" and they say "what does that even mean?  The brain that is in this body, the one that is speaking to you, contains the information that that thing on the table is a stapler, but this pocket calculator can also contain that information.  The brain in this body contains the information that this brain is a thing in this world, but this pocket calculator can also contain the information that 'this information here is contained in a thing called a calculator which is in a world.'  Heck, a piece of paper can contain information about itself being in the world.  What is special about you that you call this 'self awareness?'"

And I don't know how I would explain to these creatures what I mean when I say that I am self aware.  It seems to me that people tend to assume that if something is intelligent enough to converse with you and have poetry and math and philosophy etc, that it will also have self awareness, or at least that it will SAY that it does.  But what if it does NOT and is actually clearly capable of expressing the fact that it doesn't even know what you mean when you say you have "self awareness" because it has not even had a "simulated" philosophical zombie type of self awareness let alone the real thing?

I like to imagine one of these aliens looking at this blog post and scratching their head.  "What on Zaxar [their planet] is this blog post even about?  This brain in this body cannot even conceive what self consciousness could possibly refer to so what could the fuss even be about."  I have no answer for you here alien.  You'll just have to take my word that there's something out there that I can't explain to you, and I have no idea how it even could work, and you probably think I'm just crazy.

"Base infinity" number system

Base ten has ten symbols, base two has two symbols, base infinity logically has an infinite number of symbols, one for each number!  The first thing one might think of with base infinity would simply be something like a QR code, which you've probably seen before.

However, I wanted to make something in which the symbol would always be one continuously connected symbol, and also tried to make it obvious to the human eye where on the grid each bit is placed (neither of these things is guaranteed or expected with a QR code.)  There is a simple formula in my system for determining from a given symbol what the corresponding number is.  With these limits on what base infinity would look like, I made this:

More recently, I came up with another system.  In this system, the binary number swaps around parts of shapes, and it's more fractally.

 

One nice feature of this system is that when a number is written in binary, two numbers which differ by exactly one digit are identical in my system except that one "arm" is flipped around.  Another feature is that digits which are further to the left, and hence have more influence on the size of the integer, are generally represented by larger parts of the shape.  (The leading "1" is mostly ignored, which works since all positive integers start with a 1 anyway.)  Here are some more numbers in my system: 

Here's a large number, 3^3^3, or 7,625,597,484,987

There is actually a simpler way to display these numbers, which is to just mark the ends of the lines, instead of showing direction by having different sized lines.

But I think having using different sizes to mark the direction of the "arms" for one thing makes it easier to tell what order the digits are in when reading one of these shapes, and for another thing just looks cooler ;-)

Non-arbitrary decks of cards, or Uno with Set cards

So this is something I thought about a while ago.

A deck of cards has 13 numbers + face cards, and 4 suits.  This seems a bit arbitrary to me.  I thought to myself, we can do better than this.

How about something like, 3 numbers, 3 suits, and 3 colors?  This gives you 3^3=27 cards, which I think is much more pleasing.  Also this could be considered a "3 dimensional" deck, as each card could be uniquely placed in a cube by treating each value of each attribute as a distance from a cube edge.

Or, you could make a 4 dimensional deck, with 4 numbers, 4 suits, 4 colors, and, I dunno, 4 shadings. 4^4 = 256 cards.

Well, after thinking about this, I re-stumbled upon  a game I had played a while before that, which I now suppose could have at least unconsciously led to my idea in the first place.  This is a card game called Set, (the family game of visual perception it says on the box.)  This game has the 4 numbers, 4 shapes, 4 colors, and 4 shadings.  There are 3 of each instead of 4 of each, giving the maybe less "mathematically pleasing", but maybe easier to work with, 3^4 = 81 cards.

The game they designed for it is fun I  guess, I guess it's gotten awards and stuff.  It is mostly just about being faster than the other person at recognizing patterns.  But anyway, I bought a set and set to thinking about my old idea of using these cards to play actual games.  The game I started with worked pretty well I think.  That game is Uno, or I guess Crazy Eights.  The rules are, as I have tried out,

-Each player starts with 7 cards.
-A card is drawn from a draw pile
-Now players take turns putting down cards that "match" previous cards on a growing pile.
-"Matching" cards have either two OR three things in common.
-Two things in common means it's the next person's turn to go
-Three things in common means you can go another time if you choose to (my fiancee says you are REQUIRED to go a second time, which I guess makes things more challenging, but also maybe a bit more complicated...)
-If you have less than two things in common with the card already down, you have to draw until you can play again.
-The first person to get rid of all their cards wins.

This is basically the most basic rules of Uno but with more than just an arbitrary number of colors and numbers.  As my fiancee points out, nobody likes you when you call someone out for not saying "Uno", (wasn't me!)  so I leave it up to the prospective players whether or not to add a rule requiring the calling of "SetUno!" when down to one card.

I've only played the game with two players, but I see no reason it wouldn't work with more (I'll probably try that soon and update here.)

And yes I like the name "Setuno" for this game.

I feel like Set cards could with not to much trouble be adapted to work like other card games as well.  Maybe I will try something with a bit more strategy than Uno next, like Rummy?

Wavelength of 2^95 Planck lengths = red

Ok, so for my first post since the thing about Rule 110 tiles over a year and a half ago, I will be talking about something interesting (in my opinion) that I discovered. 

First of all, I have always been kinda fascinated by Planck units.  The  way I understand it, Planck units are fundamental amounts of stuff in physics.  For instance, the Planck length is considered to be the smallest unit of length that has meaning in current physics theories.  The Plank time is the corresponding smallest unit of time.  Light, by the way, apparently travels at exactly one Planck unit per Planck time, which is interesting. 

Anyway, I got this idea a while back I decided that the metric system doesn't go far enough in having non-arbitrary units of measure.  Sure it's all "a meter is 100 centimeters, a centimeter is 10 millimeters" but a millimeter is arbitrary.  What about doing things in multiples of 2^n Planck units?   
If I did things related to 2^n Planck units, maybe I would feel more "in tune with the universe" or something silly like that.  Now the Planck units for length and time are very small.  By trial and error and using Wolfram Alpha, I was able to figure out that 2^144 Planck times is about 1.2 seconds.  (I probably could have found an answer more efficiently using logarithms or something but I was feeling lazy.)  I used to give myself two hours to try and get as much done as possible when I had things to get done, but now I try to give myself an extra minute and a half, since 1.5 x 2^155 Planck times is close to 61.5 minutes.  (Or perhaps I should have gone for 2^155.5 Planck times and subtracted a couple minutes instead...)

I also started thinking about plank lengths.  Then I started thinking about wavelengths of light.  What color, I wondered, corresponds to some wavelength integer n of 2^n Planck lengths?  With more trial and error, I figured out that 2^95 Planck lengths corresponds to a wavelength right in the middle of the part of the red labeled region of the visible spectrum!  This seemed kind of interesting, since 2^96 Planck lengths would be twice the wavelength, and in the infrared (invisible to humans) whereas 2^94 would be half the wavelength, and also out of the range of human vision (ultraviolet.) 

I have a feeling that most likely there is no real significance to this, but who knows, maybe there is.  Humans evolved to see wavelengths of light that are available for us to see, which depends on the wavelengths most emitted by our star, which depends on nuclear/quantum physics, which depend on the Planck length and other Planck constants.  I'm sure I'm stretching here, and mostly I just like to think that there's something special about the color red for me now. 

I did some google searches for 640.2 nm and 640.3 nm (wavelengths of visible light are generally given in nanometers, 2^95 Planck lengths ≈ 640.5 nm) and found a few things that at a glance looked possibly interesting, but I also found hits for things at 640.6 nm, etc.  It would be interesting to do some analysis to see if there are more hits for 640.2 and 640.3 than for other wavelengths.  At a glance I didn't see much interesting for 640.25 nm, which is interesting in and of itself that as far as I know I'm the first to say anything interesting about that particular wavelength :-P

As an aside, there's also the Planck mass, which is interesting in and of itself.  This one is different from the Planck mass and Planck length, in that it's a maximum value, not a minimum value.  There is apparently no minimum value for mass.  One way I have heard the Planck length described is that it's the maximum amount of mass (or equivalently energy) that will fit in a Planck length x Planck length x Planck length cube of space.  (At this point it becomes the smallest possible black hole, and only becomes bigger with more mass added!)  The upshot of all this though is that unlike the Planck length and Planck mass which are much much to small to observe in any way, a Planck mass of something could actually be visible to a person.  A Planck mass is about 21.8 micrograms, which, according to Wolfram Alpha, is about 7X the mass of a "typical small grain of sand," for what that's worth. 

More interestingly, and again using Wolfram Alpha, I find that a cube made of a Planck mass of gold wold be a bout a tenth of a millimeter on an edge, and currently be worth about a thousandth of a penny!   A Planck mass of ultralight Silicon Aerogel on the other hand, would according to my calculations be a slightly less easy to lose cube 2.5 millimeters on a side!  You can buy a whole bunch of those here!  (The Planck mass of Aerogel costs a bit more than the gold D-:)  This could be an interesting marketing scheme for selling tiny amounts of stuff to nerds...yay!