Taikyoku Shogi
en.wikipedia.org
en.wikipedia.org
I love chess and all kinds of board games :) but I can't help thinking about stuff like this.
I struggle with this question for myself, certainly. I know I could be working on some kind of revenue-generating side project, or volunteering my time on a political campaign I believe in, or working to feed the homeless, but instead I play video games. Is that.. okay? Is it "right" or "good"? I don't know that I have a perfect answer.
I know this much: without some amount of leisure time, I'd go crazy, nor would I be much use in any of my "productive" capabilities; so clearly, there's an amount of leisure time that is a net good for me. I think the question becomes: what is that amount?
I think that is false. Enabling your leisure is not the sole value of work.
The GP said:
>> I know I could be [...] volunteering my time on a political campaign I believe in, or working to feed the homeless
Those are work too. If you are a certain kind of person, you do those things and value them, and they don't enable your leisure time, indeed they reduce it.
I don't think it's for fun either. Another commenter said:
>>> the number one reason I spend my time there—by far—is because it was fun
It was fun for them, but I know people who do volunteer work to help others who don't find it fun, but do it because they think it is right or essential to do anyway.
I think most people do some kinds of work which is not for the sole purpose of enabling their leisure time, at some time in their lives.
For example every key worker (nurse etc) who feels underpaid but remains in the job because they feel it is a good thing to do. Every unpaid carer of their parents. Every volunteer who delivers food to others during pandemic even when they have become fed up of doing so.
I think you could approximate this function of the value of leisure time if you had enough time and money to go around polling people. My hunch is that there would be some variation around work ethic and virtue signaling, but that there would be a clear relationship between net worth, free time, and the value of free time.
Its funny that you mentioned “feed the homeless” as an example here. When I lived in San Francisco I spent every Tuesday going to my neighborhood anarchist space to cook for Food Not Bombs and later served it on 16th and Mission. Why did I do that? Sure helping people is a noble goal, spreading anarchist propaganda through direct action is productive for a cause I believe in. But the number one reason I spend my time there—by far—is because it was fun.
Thus, you're better off playing games or reading. Being inactive and happy is by far the most rational thing to do.
Why is the fashion industry so big? Apparently it causes >10% of carbon emissions. 85% of used clothes end up in the dump.
Why is marketing so big? It has long since surpassed the point where it simply matches products with demand. The best marketing is designed to create demand. I view that as a market inefficiency.
The human emotional system has been tuned for local optimization and self benefit.
Huh? How do we know that we could? I thought we knew that we couldn't.
Because these endeavors all lead to novel tools and learnings that are at times applicable in surprising ways to help solve "whatever important problem."
For example, number theory, on its face, seems completely useless, and since 1800 BCE, years were "wasted" by scholars toying with useless factoring problems...
...until we got RSA cryptography in the 1970s.
If I had lived in the year 1900, I could have never imagined number theory having such immensely important applications, but in 2020, I sure as hell believe AlphaGo and these game solvers have relevance because it's already bearing fruit.
AlphaGo, for one, inspired many application papers. My favorite example though is of a neural net trained to solve CSPs through sudoko which was applied to the inverse protein folding problem! https://ostrokach.gitlab.io/project/proteinsolver/
Chess, Go, etc. are constrained environments. Playing these games makes you better at them. Maybe you'll find a new opening, a new sequence, etc., but I don't think you'll ever be able to apply it to another field.
Implementing AlphaGo is no longer playing the game, it's analyzing a computer science problem and finding a solution. Number theory was still about understanding some objects that seemed natural and given, rather than arbitrary ones with artificial constraints, like these games are.
- Game Theory
- Was the "Seven Bridges of Königsberg" not a fun puzzle of sorts?
- Was probability and statistics not founded on trying to hack gambling?
- Did knot theory start out with advanced applications in mind or people deep dive into making knots for religious reasons?
- How are Fermat's last theorem, the Collatz Conjecture, Goldbachs conjecture, or the twin primes conjecture not unnatural, constrained environments?
I'd argue DeepBlue and AlphaGo and Watson /are/ playing the game, just in a far more sophisticated manner, and our collective learnings as humans provide the foundation for academic endeavors. If you remember, the first variant of AlphaGo studied games played by expert humans.
Also, playing, understanding, and thinking about these games has been a boon to the AI community for decades. McCarthy has an article on the subject called "Chess as the Drosophila of AI".
My point is that these things start off as games and puzzle divorced from a problem, as a pure pursuit of leisure or curiosity, and over time, perhaps millennia, these mental explorations bear fruit in the most surprising ways.
Game Theory, Seven Bridges of Königsberg, etc. were not raised by the game players, but mathematicians.
And the original question is not arguing against the "usefulness" of these games themselves either; it's about the practice/endeavor of merely playing these games.
>DeepBlue and AlphaGo and Watson /are/ playing the game
DeepBlue and AlphaGo are playing the games. But they don't "think" or, in the OP's words, "waste their mental energy", since they are not even human beings.
The designers of these AIs are not playing the games, though. That's what the question is about.
>Also, playing, understanding, and thinking about these games
To me, thinking about these games' mechanisms and/or how to optimizing them is totally different from the mental energy used in playing it.
Again, I'm not disagreeing or agreeing with this sentiment (actually I strongly disagree), just trying to explain what the OG questions is in my understanding.
Whether you can get funded for your explorations is a separate question, but no justification is needed for them.
Yes, sitting on a couch arguably has value to someone as leisure, but its pretty clear that studying puzzles and exploring the possibilities of rigorous logical thought also happens to be valuable.
This is why we pay and award people to think about these problems rather than sit on couches.
And joy might be the best thing we can have.
Is that merely pleasure, or is it joy?
I don't accept that life is about happiness. Happiness is too vague of a term, used to describe multiple different things. Does it mean a measure of how much pleasure one is feeling in a given moment? Satisfaction with one's own life? How much entertainment someone manages to have?
We could reduce the argument that life is about simple pleasure to its logical conclusion, and ask why it wouldn't be optimal to use drugs and/or a neural device to induce a semiconscious, lasting state of maximal pleasure. I don't accept that most people would choose this life. And to anticipate that objection that it's because you have to look at a ultiliarian view of everyone's happiness, I don't think most people would choose to have everyone collectively in that state, either.
I assert that life is about meaning. That's why the above doesn't sound appealing (probably). What do you do that is truly meaningful, to yourself? That is what makes a life well lived and gives you satisfaction, and it's separate from how happy you feel.
And it could be chess games. It's up to us to look inside and figure it out for ourselves. I'm just tired of people expressing existential angst and it being met with "it makes you feel good so it's good".
If you really want to reduce, then you might as well go for the physics explanation: everything is deterministically following the laws of physics and any perception from your meat brain about what's desirable or noble or whatever are merely entropic microstates of a large complex physical system. Therefore, there's no "choice", and whether one specific individual chose to play a game at one time and another chose to study cancer at another time are entirely inevitable outcomes given the exact arrangement of atoms at their respective times.
The idea that an individual conscience has control over any outcomes would have to somehow prove that it's possible for some metaphysical force to change chemical reactions enough to "flip a bit" in a neuron in some way that is capable of affecting a macroscopic human decision that would otherwise have taken a difference course, had one "let" chemistry run its deterministic course.
I've read of experiments where lab rats lost the will to live so to speak, when they had their pleasure chemical receptors short-circuited (meaning, given the choice after experiencing the "neural device", they chose to submit to it). This suggests that motivation is indeed merely governed by a series of chemical reactions, rather than being a supernatural actor in its own right. It just so happens that from a natural selection perspective, organisms that take outside stimulus into account in their pleasure mechanisms are more likely to reproduce. Acts like choosing to play games or choosing to study are, under that interpretation, merely variations of evolved neurological mechanisms attempting to generate dopamine through convoluted means.
For some people yes, for others not.
> Frankl points to research indicating a strong relationship between “meaninglessness” and criminal behavior, addiction and depression. He argues that in the absence of meaning, people fill the resultant void with hedonistic pleasures, power, materialism, hatred, boredom, or neurotic obsessions and compulsions (Frankl 1992, p. 143).
https://en.m.wikipedia.org/wiki/Viktor_Frankl
https://www.pursuit-of-happiness.org/history-of-happiness/vi...
- why play a game when you could work for society?
- why work for society when you could create replicateable value for society's future?
- why do that when you could change society?
- why do that when you could save lives today?
- why do that when you could save humanity from extinction?
- why do that, any of that...at all?
At the end of the day, the only remotely satisfactory reason for doing anything is to pursue individual fulfillment (even if that be in pursuit of a collective goal).
Additionally, progress in any valuable-to-society pursuit does not happen singularly within the confines of an academic or professional field. Human brains work in mysterious ways, and giving them a wider space of things to bounce off of and different systems to analyze only strengthens it, not take away from it.
I would be willing to bet the time (if/when) we find a cure to cancer will be concurrent with the most board-game-shaped-activities being participated in by humanity, not the least.
TL;DR: human brain utility is non-zero-sum, unlike computers.
Then you need people who actually dedicate serious resources to the game, to bring it to its limit and to be able to teach students the relevant parts.
A great quote I heard about physical training is, "if you try to peak all the time you will plateau."
Work is important but it is helpful to have a couple of things to do with your life besides it.
You'd be turning work into fun, and that into solving hard human problems. If you get there, maybe you'll notice that curing cancer wouldn't help humanity in the long run, so you'd turn to the cause of that. Could end up in a vicious circle. But as long as it is fun, go for it.
That's the catch. We don't know what problems are important. We think we do, and sometimes we're right, but we've gone down dead ends too. And sometimes something that feels like a dead end turns out to be pivotal.
I'd bet there's many scientists and researchers out there who had their eureka moment while their brain was disengaged during leisure time. Disconnecting and play are essential parts of discovering the important things.
There are charities that focus on reducing human hunger. There are also charities that focus on reducing child hunger. Why should the latter exist, since it's a subset of the former? Isn't everyone working on the child hunger charities ignoring the problems of adult hunger? Of course not. There are just some people who, for whatever reason, are specifically interested and motivated in fighting child hunger, and I strongly suspect that overall outcomes are better when those people have the opportunity to focus their work on child hunger.
The problem is that without some eventual justification the whole value system becomes circular. You can't put all of your efforts into working to keep people alive so they can... put all of their efforts into working to keep people alive, etc. All value systems are ultimately circular or unjustified, but the latter is IMHO superior since it's more honest.
Six days ago, I had brain surgery. I made it home a few days ago and I'm still mostly bed ridden. Despite this, I'm starting to work again because I love my job and I'll be bored out of my mind if I stop...
If I was at differently stage in my career, I would probably be playing Civ and shooting for a religious victory under Canada (the religion would be named "Tim Horton's" of course).
People with harder surgeries (like brainstem) take months to recover.
Here's my story if you're interested: https://twitter.com/rclmenezes/status/1321160909360943109
Is this to say that you enjoy your work that much at the present (more advanced?) stage?
Also, have you tried Europa Universalis 4?
Another take on this is that there is no "noble goal" at all, whatever one individual finds interesting is their own choice and freedom and there's no reason to create a hierarchy of goals which would be completely arbitrary.
What you are suggesting is a nihilistic approach. Despite our collective inability to determine any "root goal" of the universe to aspire to, we collectively benefit from certain "local goals" - such as trash collection, trying to stop people from running around and killing others, having food (ideally good) available, etc.
A "noble goal" is a social construct. You might actually define it as a goal with positive impact on a society that some members would rather not or cannot do themselves.
I agree that people should not have their goals forced. However, the concept of a "noble goal" is useful in describing this evolutionarily emergent social phenomenon. And I am in favour of a societal incentive system that rewards positive impact to society.
Most of my goals are pretty neutral, and I guess I don't really put much effort into any particularly noble ones. Open source software, I guess. Though I can see that some goals really have a large beneficial impact to the experience of life, and can't help but appreciate people who engage in those goals.
Much better than being at home watching TV all day which is what many elderly people are forced to do due to poor mobility.
In Japan many elderly people play Go and Shogi. In China people play Weichi and Xiangqi, the equivalent games respectively.
They are also not necessarily a waste of time. The intuition and insights gained in board games can be used in other aspects of your life.
We do many wasteful things as humans. We cut down trees to turn them into supermarket marketing that go directly into the trash.
We drive 5 seat vehicles with one person in them and spend a lot of time waiting in traffic so for other 5 seat vehicles with one person in them.
We spend money on overpriced shiny stuff with no purpose.
We spend years in college learning about computer science so we can disregard most of it and push dirty code full of tech debt.
We spend in scientific research so then a climate change denier, flat earther or anti vaxxer refutes it with no proof.
In the first world model, wasting energy and time is indeed suboptimal and humans indeed must focus on survival. In the second world model, curing the cancer is about as valuable as extending the lifetime of cars to thousands of years, while figuring out how the world works is the number one priority. I believe that humans intuitively lean towards the second model, and although they can't say it out loud, they do things that are optimal in that model. Thinking about chess won't extend anyone's lifetime, but it might reveal something important about the underlying principles of our world or of our thinking process and we secretly recognize that making that little discovery would be worth everything.
I think what it comes down to is access. It's easier to buy a chess board and books about chess strategy than it is to buy the tools you would need to do cancer research. The chess community is open to anyone who wants to join; to join the cancer research community, you need years of school. There's an institution sitting in front of the problem saying "trust us, we're working on it" -- and if you want to work on it too, you have to do it as part of that institution. Most people aren't motivated enough to climb this institutional mountain. They'd rather stay at base camp and enjoy themselves.
(This is one nice thing about software, by the way. You don't need a degree or an expensive lab to play around, learn, and contribute.)
Frankly I'm much more worried about engineers wasting their mental energy on improving tracking of users to sell useless stuff that will end up in a landfill.
Hardcore optimization of one's life is the path to becoming soulless imho.
It's ok to have fun.
In shogi there is no queen and there is only 1 bishop and 1 rook. Pawns capture forward and can move only 1 tile at a time and there is no en passant. There are lances which are like rooks but can only move forward. Knights have 2 ways of moving instead of 8.
There are many more differences, like drops, different promotion rules, etc. It is a cool game.
Shogi gets pretty exciting towards the endgame.
An easy way to play shogi is in 81dojo. That site is available in English and is pretty active.
It can take a while to get used to reading the pieces and memorizing moves and promotions, but there are piece sets available that show you the available moves for each piece.
Illegal moves are the easiest way of losing. e.g.: nifu. And, you are expected to recognize a checkmate and resign.
And there is a "beginner" version, doubutsu shogi. Which consists of fewer pieces and a smaller board.
I think both games have their interesting parts and it's fun to play both.
Is that different from chess?
Although it is rather hard to explain for a beginner like myself. Mating in shogi is complicated...
In chess you have literally no legal moves, but in go a good player just knows that it's hopeless to continue, and it's rude to keep going.
If you ever do reenactment of a sword-and-shield line fight it'll make a lot more sense. You naturally counter the person in front of you and neither makes any progress, the people who are diagonal from each other are where the action is.
Also, people do play crazy games these days, look up "monster wargames". For example "Advanced European Theater of Operations" has 2200 pieces played on 2 poster-sized maps...
I don't know what the Japanese word for Grognard is, but whomever played this qualifies.
Sounds fun though. I enjoy playing new boardgames on steam against friends without having read the rules. In an electronic game, you can't make illegal moves, so you're on even footing, clueless, but still playing in some approximation of correctly.
Somehow I don't think I could do that for this one, electronic or otherwise.
You'll never get anywhere on a real grognard game without spending hours poring over the rules first. The complexity is inherent to the system. Many of the better ones will try to compartmentalize it however, so you can start play on a relatively simple ruleset and only dive into the complex stuff when pieces interact.
Also, thanks for adding Grognard to my vocabulary lol
And can be linked to Advanced Pacific Theater of Operations for those who think it is too small of a game on its own.
Then some one made a functional Taikyoku Shogi flash game, and they started playing a match, by posting each moves on the thread one by one. It took about a year or so to complete. That may have been a first time the game is actually played in history.
The second time may be on the TV show トリビアの泉, by using real restored pieces mentioned in the article, and two professional Shogi player. It was around 2007.
The pieces reminds me of Pokemon Wack [0], which has more than 3000+ pokemons, and 49 types, including Sound, Food, Meme, Plastic.
I wonder if the extra space and pieces would increase the length of training polynomially or exponentially.
I'd expect AlphaZero to hammer the best humans given a similar amount of training (~100,000 2017 TPU-hours) in this shogi variant, so no increased training time at all by that metric. Maybe a more meaningful metric is how closely it approaches perfect play, but we don't have a good way to measure that.
As a problem solver, I'd like to think that I'm always subconsciously working on unsolved problems at the back of my mind. Taking my mind off a problem for a bit might help me get a fresh perspective on things, and lead me to eventually solving it faster.
Mental energy distractions like board games and TV might help in similar ways.
The game they spend most of the time playing is the fictional Gungi [0].
- Longest article title: Cneoridium dumosum (Nuttall) Hooker F. Collected March 26, 1960, at an Elevation of about 1450 Meters on Cerro Quemazón, 15 Miles South of Bahía de Los Angeles, Baja California, México, Apparently for a Southeastward Range Extension of Some 140 Miles (253 characters)
- Longest article title without spaces: Lopadotemachoselachogaleokranioleipsanodrimhypotrimmatosilphiokarabomelitokatakechymenokichlepikossyphophattoperisteralektryonoptekephalliokigklopeleiolagoiosiraiobaphetraganopterygon (183 characters)
- Longest full typed number with an article: 9,223,372,036,854,775,807
EDIT: Er, it makes more sense for me to have interpreted that as article length rather than title length. :P There's actually an entire autogenerated page for tracking that: https://en.wikipedia.org/wiki/Special:LongPages
Ideally, the length of an encyclopedia's article about a subject would be proportional to the subject's importance and to the amount of information available to the subject, not proportional to the contentiousness, propaganda value, or recency bias of that article.
Huh. Without looking, I thought that would be a dinosaur species that someone named in jest. It turns out it's a dish in Aristophane's Assemblywomen.
Proceeds to detail information on 257 different unit types
Based on the link kibwen provided: https://en.wikipedia.org/wiki/Special:LongPages