Website founded by
Milan Velimirović
in 2006
17:56 UTC
| |
MatPlus.Net  Forum General The Fairy Chess Classification Project @ Julia's Fairies |
| |
|
|
|
|
You can only view this page!
| | Page: [Previous] [Next] 1 2 3 | | (21) Posted by Andrew Buchanan [Tuesday, May 4, 2021 05:11] |
Thanks all. I haven't done anything yet, but I am excited to be part of this great endeavour.
I want to share two links:
(1) https://gatherer.wizards.com/Pages/Advanced.aspx
This is the front page of the card database for the collectable cardgame Magic The Gathering, which allows browsing of currently 22,105 different Magic cards. All this is underpinned by an information model. Some of it is indeed hierarchical (type, subtype...) but there are many orthogonal dimensions (e.g. expansion, colour). There is multiple inheritance too. It used to be chaos, but now a better structuring of the elements of their world has many benefits, not least that it highlights "fresh design space" where new cards can be imagined. Without the software development of an online version of the game (after many failed attempts), the rationalization of ideas behind this database could never have happened.
(2) https://en.wikipedia.org/wiki/Celestial_Emporium_of_Benevolent_Knowledge
Jorge Luis Borges is one of my favourite writers, and this is one of his best conceits, showing the folly and imperfection of classification. And yet we must do it! I hope this reminds me to keep my sense of humour. Apart from anything else, any modeling process requires clear judgement, and I tend to lose my edge if I am emotionally stressed. And this is all just a bit of fun really.
Gens Una Sumus,
Andrew
| | | (22) Posted by Hauke Reddmann [Tuesday, May 4, 2021 10:00] |
@Kevin: I definitely agree with you on one point. If the new AIs
(read: deep neural nets) manage to create "true art" (however
that shall be defined!) from scratch, chess problem, with its
high degree of possible formalization, would be a very probable
"first victim".
But then, beauty is never tarnished. Would, say, the Saavedra
study be less beautiful if it had been an accidental find in a
4 men tablebase?
Or a trained chimpanzee threw it on the board?
I still ride my bike to the Baltic sea even if a car would bring
me there faster, and I eagerly want to come back to the board
after Covid is over even if the computer program of my 40 year
old ATARI always clobbered me. Assume an artificial human will
be built in 200 years, who can do everything better than the
natural stuff, does this make human life worthless?
I, for one, welcome our new insect overlords. :-)
| | | (23) Posted by Kevin Begley [Tuesday, May 4, 2021 11:21] |
I don't know that Magic (the gathering) is what you want to be using as an organizational model -- FYI: that website doesn't make your case clearly (expect perhaps to those previously inclined to that strange card game, who may appreciate what you are trying to express there about categorization).
Instead, I would strongly encourage you to adopt a scientific approach.
Begin by inventing scientific terminology to describe the fairy chess elements.
This will help your team overcome conventional biases.
Imagine you live in the year 1535, and you assemble a group intent upon classifying all the plants and animals (it's not so far a stretch from reality).
If you present evidence to such group that there is a strong relationship between jellyfish and coral, and you advocate for classifying them accordingly, you are likely to get laughed out of the room.
Worse, imagine if your evidence based presentation should leak to the public -- the entire effort would be exposed to untold ridicule (from scientifically clueless hecklers).
In truth, those who would laugh are actually expressing their ignorance and their bias (I would argue their goal is not to understand nature, but to assert their false dominion over nature).
That's exactly where we are, at present, in fairy chess classification -- somewhere near the year 1535.
Except that our scientists lack the courage to make such a presentation!
If they step out of line, their contributions may be called into question.
In fact, jellyfish and coral both belong to the family group called cnidarians, and they are closely related organisms.
Instead, imagine you convince the entire assembly to start from scratch.
The term "fish" is not yet well defined, you insist, therefore the term "jellyfish" carries with it a false notion.
Conventional names are nothing but a hinderance (and a liability, should your progress leak to the public).
If your goal is to understand nature, you need to adopt scientific names.
Jellyfish = Scyphozoa.
Corals = Anthozoa.
Start with completely unrelated scientific names (random, unpronounceable words), and you allow for alteration (in accord with discoveries which establish a distinct relationship -- which you can define).
This liberates you to discover nature as it actually exists, rather than to impose ignorance and biases upon it.
If Morphy taught us anything, it's that you can not exert dominion over a chess position (you must listen to what the position naturally demands).
First step: empty your cup. Unlearn everything you think you know about fairy chess.
Frankly, everything you think you know is wrong -- as is evidenced by the fact that no fairy chess taxonomist to date has been willing to define their most fundamental divisions; all they can do is attempt to laugh you out of the room if you challenge them.
That's why rookies could outperform our most experienced problemists in modelling fairy chess elements: their cups are already empty!
They don't have the ego investment, they are not tied to naming conventions, and they don't suffer under the absurd notion that the "jellyfish" must be classified in a group containing the "goldfish."
They are not inclined to exhibit hostility toward anyone who challenges our faulty foundation.
People are going to be outraged by your findings, quite naturally, if you are bold enough to allow nature to inform your classification system.
Better to present your findings using scientific names (this tends to discourage the lazy hecklers, leaving only those who care enough to study your findings carefully).
Do not express any findings in unscientific terms (nowhere should you use the term jellyfish -- to do so constitutes an inherent admission of ignorance and bias).
Outsiders are already doing this, and they are going to own this franchise if we continue to demonstrate an unwillingness to invest ourselves in understanding the natural framework of our own artform. Mark my words! That train has already left the station.
If you have spent an hour studying artificial intelligence, you understand why data classification is fundamentally important (it's step one).
That is the light at the end of the tunnel, and it's steaming our way.
Use scientific terms, and implement your model (as a software developer would).
If that's what Magic (the gathering) has done to overcome conventional bias, maybe you can relate that better (for those of us who are not inclined to fantasy card games).
| | | (24) Posted by Neal Turner [Tuesday, May 4, 2021 13:48] |
Kevin likes to use the word 'logical', what I'd like him to explain is what it actually means in this context.
Does it mean that we need to have 'logical' relationships between the various elements?
What is the nature of these relationships?
How do we know whether the parts of the system are logically consistent or not?
Could he give us an inkling of what his logical classification would look like?
I ask these questions because in Chess we have the paradox that, although everything about it - the board, the pieces, the rules - is well defined, the game itself resists any and all attacks from the Logicians.
This is one of the very things that attracts people to it!
| | | (25) Posted by Kevin Begley [Tuesday, May 4, 2021 14:47] |
@kalyan,
I'm glad to learn that somebody in this project is willing to listen to an honest critique.
I am freely offering honest advice here (not necessarily for your group, but for anyone willing to listen).
As for the atmosphere -- ha! -- let me tell you something about atmosphere...
When your group's leader calls into question the contributions of anyone who offers an intelligent critique of her product, you would be better served to address the hostile atmosphere radiating from within your group.
If you can show me where I have personally insulted anyone involved, I'll issue an immediate apology.
I have demonstrated nothing but sincere respect for everyone involved (including those who have attacked me personally).
I have been gracious, even to those who have falsely claimed my motivation is something other than generous, constructive criticism (note: I have not retaliated in kind).
Beyond that, I harbor nothing but good will toward everyone who has volunteered to help.
I hope they will succeed.
They are not going to succeed if we ignore their profound failures. Failure to develop logical definitions for ANY of their four key terms, which they use to divide problems, is an obvious blunder in any classification effort.
There is no denying this.
I take no joy in delivering that bad news -- I came into this thread hoping to congratulate the entire team (this is apparent from the very first sentence in my very first post).
I do not aspire to find controversial positions -- truth be told, I don't like to involved in these petty altercations.
I wind up at the forefront of these controversies because I value truth far more than I fear slings and arrows.
Maybe there was a more diplomatic way to tell that truth. Maybe. Perhaps I didn't sugarcoat it enough.
But, I didn't disgracefully suggest that a member of your group has made no beneficial contribution to chess problems.
That was not me.
Where's the apology for that?
Why has your group not demanded an apology -- as this statement was made by a person who purports to represent your entire group.
Your failure to call for an apology makes you complicit (as if you've made that disgraceful statement to me, personally).
If you're going to lecture the critics about atmosphere, sir, you're going to first need to own up to your own failings in that department!
That was a shameful statement.
Does your group demand six months of benevolent contributions from rookie volunteers?
Should I call upon you to demonstrate your six months of contributions?
I would never do that to you, and I would certainly never allow somebody who represents me to speak to you in such a disgraceful manner.
Nor would I deign to lecture you on presumptions that your failure to join this group constitutes a shameful lack of commitment.
Furthermore, I would not uncharitably ascribe to you foul motivations.
I'm not the one attempting to silence opinions here.
That's the atmosphere coming from your group.
I'm trying to help a broader audience (especially any rookies who may stumble upon this thread) to understand that the next generation of problemists will inherit a bad deal; I am trying to embolden them to correct what corrupted conventions we leave behind.
I'm trying to show them that they need not fear these slings and arrows (which are routinely launched against anyone who dares question bad convention).
That's why I wind up in these controversial debates: because I don't fear the petty attacks that tend to silence some problemists (those you might consider diplomats). It's not in my nature to cower with diplomats. I say things they will not (things they know are true).
Everybody knows your team has failed in the primary stage of your task (you have failed to define the four fundamental terms by which you divide fairy problem elements).
I don't want to be the person who tells you.
Dmitri tried to warn your group (internally, and diplomatically I trust).
I wish you'd have all listened to him, so I wouldn't be in this position!
But, whatever the case, I certainly am not the party responsible for any hostile environment here.
If you have a different view, I'm happy to discuss it with you (either here, or you can send me a personal message).
I'm open to discussion.
If you convince me I'm wrong, I'll issue a sincere apology.
If that's considered a hostile atmosphere, I don't know what more I can do to make you happy, but I'm certainly willing to listen to your suggestions!!
Probably more than anyone who has volunteered, I want to see your group succeed (in spite of all the bad vibrations that have come from your leadership and your defenders, that remains my wish for all of you).
If you somehow misperceive that as creating a hostile environment, I'll be happy to parse everything I have said (word for word, if necessary) with you.
I believe you are profoundly mistaken, my friend, but I shall always welcome (and value) your opinion.
The same goes for those whom I believe have wronged me personally in this thread (truth be told, you wouldn't believe how much I respect their work -- so it's rather sad to find myself in the crosshairs of their misbegotten animus -- but that's not going to inhibit me from delivering hard truths).
I wish your group's leadership had the grace to acknowledge the glaring failure in the classification scheme presented.
It's undeniable. So, admit it. Big deal. Yes, we made a fundamental mistake, and we're going to correct it in the next version.
I'd be overjoyed. You'd be happy. Everybody would be a winner!
The classification document would be improved.
Thank you, and you're welcome. I could be on my way back to relative obscurity.
If you want to suggest I am responsible for the way your group officially responded to an honest critique, I look forward to having that discussion with you (and any ambassadors you want to bring with you).
I'd like nothing better than to meet you all amicably, and discuss ways we can all improve the environment here.
Let's schedule that meeting. I'm happy to volunteer my time for that good cause.
| | | (26) Posted by Dmitri Turevski [Tuesday, May 4, 2021 22:50] |
Let me explain how I see these matters and then I welcome everyone to criticize my point of view and me personally and my involvement.
There are very many basic concepts familiar to the chess player: position, piece, square, move, etc.
It looks like if you introduce a new (fairy) element to the rules of the game (a nigthrider or a vertical cylinder board) then these concepts don't need to be redefined at all (to start, they were never defined at all, they were so natural), and it seems so natural that if the bishop can move from a1 to h2 on the v-cylinder board , then the nightrider can move from a1 to e3.
No. This is where it all goes wrong. Once the inventor of a new fairy element introduces a step away from the rules, they expect everyone to adjust their understanding of the basic concepts to shift accordingly. Unfortunately, it doesn't work so, everyone understands it a bit differently. Popeye and Winchloe give different solutions to the same problem only because the inventor wasn't clear enough. Can a rose-hopper hop over itself? :)
How can this be fixed? One way is to develop a very formal system of rigorous definitions. Like really rigorous in mathematical sense, this is what I believe, Kevin is talking about. Could it be fixed in some other way? No idea.
Now, compiling different sources, that each may have a slightly different notion of what a basic concept is (piece, move, chess problem, etc.) into a single document. It obviously solves a different problem if any.
Above, I intentionally tried to stay away from the chess software developer point of view.
Kevin:
QUOTE
Dmitri tried to warn your group (internally, and diplomatically I trust).
Not very internally, basically in comments to:
https://juliasfairies.com/fairy-elements-statistics/
https://juliasfairies.com/fairy-classification-preliminary/
| | | (27) Posted by Kevin Begley [Wednesday, May 5, 2021 05:16] |
Dmitri,
In fairness, inventors generally do a good job of defining their new elements (and that is their primary mission).
Yes, some (even the best inventors) will overlook special cases.
Consider Circe Parrain -- even in this smartly defined condition, where the inventor was careful to consider special case moves, it has been noticed that special case ambiguities exist (units immediately reborn behind a pawn advanced two-squares *could* give rise to a double-capturing, and double-rebirth possibility; and castling *might* be possible across check when a virtual rebirth provides a shield).
Each ambiguity may give rise to a split in the condition -- which is not necessarily a bad thing, except that nobody can provide a good method to determine which direction the older problems should be reclassified (read: which version should be considered the default condition?).
Favoritism remains our preferred methodology when logical discernment proves a hefty brain tax.
Historically, these matters were decided by favoritism (which problemist do you like better?).
Of late, the developers have the ultimate say (database and problem solving utilities are the ultimate arbiter of the rules, and the names).
I favor the idea that our fairy problem community should reclaim their authority to logically govern these matters.
Unfortunately, the people in charge of this effort have not demonstrated a commitment to a logical (and just) classification system.
Plenty of fairy inventions come from rookies.
Until our community provides them a better framework, I can't fault the inventor (I don't even expect them to consider every special case rule -- it should be up to the community to help them identify oversights, and guide them in arbitrary decisions, and provide intelligent naming conventions).
We offer them no capacity to petition for advice, we offer no document suggesting default conventions, and we refuse to establish a body that might sanction (or commend) well considered inventions. When a fairy inventor gets it wrong, we are at fault -- we were all complicit in accepting these inventions (just as we are all complicit in accepting the status quo of our every historic classification failure).
This lack of conventional clarity is why pawns reborn onto the first rank (in a variety of circe forms) can sometimes advance three squares, sometimes two squares, sometimes one square, and are sometimes reduced to dummies (no capture and no advance).
The people who are not "bitching" about this are most culpable. The diplomats who cower to authority are to blame.
Not the inventors! Please, let's don't discourage inventors. Let's instead help them.
The fairy problem community has allowed this wild west atmosphere, and we have made no effort to remedy (nor to even acknowledge the maladies which spring naturally from our failed system).
Worse, elites have conditioned our community members to mindlessly defend against any attempt to confront our failures.
Try to make any improvement, and you immediately become the issue. The collective will march upon you, in lockstep, to shut down any attempt at resistance.
How dare you "bitch" about a classification system which offers no definition for their most fundamental divisions!
Perfection is mathematically impossible. You haven't contributed enough to warrant a fair hearing. Your motivations are suspect.
Some years back, one defender of the status quo actually protested that Kurt Gödel had proved it might be impossible to define the rules of a fairy condition (and who the hell am I to argue with such a towering logician?).
I do wish Bobby Fischer had heard that claim, back when he introduced Chess-960 -- it would have been a dream to hear his reaction to this mindless, defensive hive inclination (alas, he was probably already insane with hate).
Such absurd claims emerge from a highly intelligent people, and accomplished composers.
Nobody is safe from getting completely turned around in our toxic fog atmosphere.
There'd be little to argue about, I suppose, if the fog should ever lift -- perhaps that's the purpose for preserving it at all costs!?
In my experience (as a 7th grader), it's not very difficult to develop a logical classification system (even when tasked with unfamiliar subjects, any random group of 7th graders will prove capable of discovering logical taxonomies and ontologies, and providing definitions which eliminate all ambiguity). The only holdup is our motivation.
But, we are all responsible -- it's hardly the fairy inventor's burden.
We all go along with it, perhaps in the false expectation that this failed system might grant us wide ranging liberty to define our own fairy inventions. I hate to deliver more bad news, but this pleasant fiction has not been our history.
The fairy community has (on rare occasions) imposed their own rules, and their own naming conventions (quite unequally, upon the inventions of disfavored individuals, it would seem).
The inventor of Platzwechsel Circe, for whatever odd reason, adamantly insisted his invention was not a form of circe, despite the fact that it is undeniably so for anyone who accepts that circe constitutes a superset, under which all such variants -- which rebirth captured units (according to whatever distinct rules govern said rebirth) -- should be logically classified.
People protested his invention, and demanded it be renamed "exchange circe" (contrary to the inventor's wishes).
Eventually, the inventor settled on calling it PWC (insisting this constitutes no acronym for Platz Wechsel Circe).
He was satisfied to have removed the word "circe," I suppose.
Fast forward a few decades...
I happened to notice that the "Take & Make" condition is actually a subset of anti-circe.
The capturing unit is reborn according to some distinct rules.
From this new realization, I offered one interpretation of an analogous circe form of Take & Make (from which a vast array of popular conditions were born).
The inventor of Take & Make, quite naturally, insisted that his invention should not be renamed (he argued, as the PWC argued before him, that the rules governing the rebirth of capturing units in his condition was sufficiently different to require a separate classification -- this could not be classified as a subset of anti-circe).
And what happened?
Well, before we could have a rational discussion, somebody came along and plagiarized my idea, and promptly renamed it (removing the anti-circe superset -- without which, none of these fairy possibilities would have ever been discovered).
Rather than follow precedent, the larger community favored the plagiarist (who later insisted that the inventor is irrelevant, he had secured the right to name the condition because he, and he alone, had made it popular).
This is what inventors are up against!
Ask yourself: what rational person would want to contribute in such a corrupted environment?
And the irony is, people here will actually question your contributions!
They will declare you responsible for the bad atmosphere!
It's so pathetic, it expands the borders of absurdity into madness.
(note: in fairness, there were a number of people who did recognize that I had it right all along; and, without their support, in private messages, I probably would never have returned to this community).
My sincere hope is that the fairy problem community will eventually find the motivation to get itself together.
That's why I originally posted in this thread. I came here to congratulate the group which had the courage to right this sinking ship.
When I discovered they had done the exact opposite, I admit, I was supremely disappointed to discover that their document delivered not a logical classification system, as promised, but yet another sham!
Somebody is going to get this right.
Probably it will be software development teams (who will have their own agenda, and this entire community will wish we had done more when their dominance, and their logical classification system, leaves us no claim to our own artform).
AI is not coming to help -- they are coming to dominate!
They aim to demonstrate artificial intelligence by measuring their machines against our failures to behave intelligently.
Logical classification (unambiguous taxonomies and ontologies, developing a knowledge base) is the key to AI development.
The first thing they are going to do is upend your entire classification system. Of this, I have no doubt.
This community has a narrowing window of opportunity remaining. If we don't get our classification system in order, we will surrender the framework of our artform to outside agents (which is going to cause a feeding frenzy -- computer scientists are going to come out of the woodwork to expose our collective failures).
I would be the first to welcome our AI overlords, except that I recognize their devotion is not to our artform.
Their agenda is to kick us all in the head, and declare themselves intelligent for exposing our gross history of malfeasance, illogic, and corruption.
But, if we are content to do nothing but mindlessly defend the status quo, hey, we're begging problemists to join with those software teams.
It wouldn't take much more of the same treatment to push fairy enthusiasts (like me) in the opposite direction.
It's like the year is 1535, and people want to map the known fairy universe.
Note that no rulers, no country name, and no human defined boundary from that period has endured.
What will endure? Simple: a logical, well-defined classification system does not change.
That should be our foundation, and we should provide it to every fairy inventor.
We should encourage their contributions (the way Alpha-Composer will welcome every contribution from Alpha-Fairy-Inventor -- both of which will rely upon the advantages that a superior classification system confers).
| | | (28) Posted by Kevin Begley [Wednesday, May 5, 2021 08:48] |
@Neal,
>"Kevin likes to use the word 'logical', what I'd like him to explain is what it actually means in this context."
Good question! I'm glad you noticed my penchant for overusing this word. LOL. The distinction is important.
In this context, I'm asking only for a classification which is based upon unambiguous definitions.
It turns out, this is the whole idea behind any classification (that it be logical).
If the group would provide you with a concise definition for fairy condition, and a concise definition for stipulation, it would enable you to:
1) distinguish between the two, and
2) determine for yourself how to classify new elements (based upon the logical definitions provided).
That's the FUNDAMENTAL task in any fairy chess classification project -- to provide readers with a LOGICAL basis to independently determine groupings of elements (both existing elements, and elements not yet discovered/invented).
It is NOT logical to group dinosaurs according to subjective traits.
The class of dinosaurs I consider "cool" (or "scary") has no logical utility (logical classification depends upon traits which can be objectively measured). Favoritism is not a logical form of classification!
I don't ask anything else from this group -- just unambiguous definitions! Should they provide a logical basis for their divisions, the elements will naturally fall into place (and though we may argue whether their divisions are optimal, or the most informative classification, nobody can argue that they will have provided us all with a logical classification system).
I was first introduced to the science of classification as a 7th grader (I have since learned this exercise was commonplace in American public schools).
You divide the class into a few small groups, and make each group responsible to establish and present their own logical categorization system (dividing dinosaurs, tools, and what have you -- turns out, you don't need a paleontologist to find interesting, valuable ways to categorize dinosaurs).
There is a WIDE variety of logical ways to divide (and sub-divide) elements of any group.
And every logical classification can provide us new perspectives.
It's not intended as an insult when I tell you that a small group of random 7th graders could do a better job of logically classifying fairy chess elements (though it is highly doubtful their presentation would match the graphics produced by this volunteer group of fairy chess enthusiasts).
That's just an observed fact.
Why can't our experts get it right? I submit to you: they are deliberately sabotaging their own mission.
If problemists were motivated to produce a logical classification system, we'd have had it decades ago! Instead, we all just want to go along with the illogical historical divisions (rooted in favoritism), to get along.
I think you know that, Neal.
Deep down, I think we all know this has always been the obstruction.
We are in a fog, but it's undeniably a deliberate fog (one which registers in the visual cortex).
Who among us is motivated to defy convention, knowing it will bring only reprisals (we become a party to the corruption of our own artform, because it works to our advantage).
Would you want to bring forth a logical classification system, knowing any deviation from the illogical status quo will gain you only ridicule and insults? Are you ready to invite respected problemists to question your motivations, and to doubt your every contribution?
This is ALWAYS what honesty gets you here. It's a recurring pattern. It doesn't matter who you are.
If you challenge a corrupted system, the corrupted system will fight back in the only way it can (and it's not going to be with logic).
2nd question:
"Does it mean that we need to have 'logical' relationships between the various elements?
[and if so] What is the nature of these relationships?
Excellent question!
Discovering logical relationships between members is one major advantage of logical classification.
Classification helps us to discover that dinosaurs which walk on four legs tend to be herbivories, whereas dinosaurs which walk on two legs tend to be carnivores. There is much we can learn from ANY logical classification!
Indeed, a more logical classification enabled me to discover that Take & Make is actually a form of the anti-circe class.
Otherwise, I might never have noticed this relationship.
And only from that recognition, which emerged from logical classification, was I able to discern that an analogous circe version might also exist.
There is much we can learn of the logical relationship between elements of any group (or sub-group), just from the process of logical classification (EVERY logical classification conceivable has the potential to provide us new insights).
That's why I don't want to join this group.
Long ago, I developed my own logical classification system (little of which can be considered original, in truth, though I think I did contribute in some small part, by recognizing that stipulations could be considered aims within larger stipulations -- and this can be effective not only to model help-selfmates, and to implement a help-selfmate search function, but also to predict new stipulations that might be invented in the future).
ASIDE: shortly after that discovery, I was quite surprised to discover (in the Win Chloe database), that some of these "futuristic" stipulations my classification had predicted were already in some use (though far from common)!
If I join this group, I'm likely to wind up advocating for my own vision.
I stand to learn much more if I stay out of it, and allow this team to produce their own logical classification system.
The more they differ from my own logical classification system, the more novel will be the perspective gained (from which I may gain new insights on the inter-relationships between elements).
I welcome any originality.
I have no bias which elements fall into which bucket (so long as they as they are divided by unambiguous definitions).
Should series-movers land in the stipulation bucket (along with helpmates, reflexmates, and even progressive chess itself), according to a logical definition (based upon objective traits), I will have no case against this grouping. None. In fact, I will be very impressed if they can achieve this with a concise definition, and I will be very interested to discover what relationship such elements may share.
I could learn something from that grouping, if it were based upon a logical classification.
We learn nothing when the groupings are based upon favoritism and inherent bias.
Question #3 and #4:
"How do we know whether the parts of the system are logically consistent or not?
Could he give us an inkling of what his logical classification would look like?"
Hopefully, you can already answer theses questions for yourself (from the answers above).
There are many logically consistent ways to group dinosaurs, or tools, or fairy chess elements.
You know a grouping is logically consistent when the definitions for each division (and sub-division) are sufficient to inform you what goes where (when you need nothing more to know where a new element belongs).
Consider the document produced.
Suppose I invent a new fairy element, where black gets three moves per every two white moves.
Ask yourself: in which group does this new element logically belong?
If it were a logically consistent classification system, the definitions of their four main groups (which are not provided) would make that answer clear. You'd know immediately whether this new invention should be classified as a new stipulation, or a new fairy condition!
Ask yourself: how do you know a vertical cylinder is a board, and not merely a fairy condition?
What distinguishes these two possibilities?
If your aim is to classify fairy chess elements, you need to provide a logical answer to this important question (otherwise, you must admit that you have failed in your mission).
As I see it, this is (always) an arbitrary decision.
There are advantages and disadvantages to classifying a vertical cylinder as a variant board/fairy condition.
Both can be logically consistent classifications (if you establish adequate definitions to accommodate them).
How do you know which choice is best?
There's no good answer.
I tend to favor the classification which leads to the most concise implementation (in software).
In my experience, that tends to be the better choice.
But, I can't claim it's the only logically consistent choice.
If you want to take an alternative approach, that's fine. I welcome that. I welcome ANY logical classification (the more original your perspective, the more it may teach me of relationships I might never have imagined).
Have at it. Knock my socks off! You'll get nothing but gratitude from me (so long as I can discern, from your concise definitions, what element goes where, I am overjoyed).
>Comment #1:
"I ask these questions because in Chess we have the paradox that, although everything about it - the board, the pieces, the rules - is well defined, the game itself resists any and all attacks from the Logicians.
This is one of the very things that attracts people to it!"
I don't know how you perceive the chess game as resistant to "attacks" (definition?) from logicians.
It seems to me logicians attacked the endgame tables, and they came back from that expedition with plenty of big fish!
If you mean the game is so complex, that logicians lack the resources to fully solve it, that complexity is primarily a feature of board size.
I don't think complexity itself is what attracts people to chess. I think it's the balance between pattern familiarity (simplicity), and the recognition that almost every familiar pattern will have surprising exceptions (complexity).
Chess falls perfectly in the Goldilocks Zone of simplicity/complexity.
The game itself is not resistant to logic, it is only the complexity that is resistant to our capacity to solve it.
Play any chess engine on your phone, and you'll soon realize that the attacks from Mathematicians and Logicians (following in the footsteps of Claud Shannon) are not to be underestimated.
We will reach the point where an unbeatable AI will emerge.
Logicians needn't solve the game entirely, down to a 64-man EGTB, to effectively conquer it.
That's why Caissa saw fit to give us fairy chess. :)
| | | (29) Posted by Kevin Begley [Wednesday, May 5, 2021 14:38] |
Speaking of arbitrary classification decisions...
It's interesting that one option is to decide that rules governing units (rather than rules governing the overall game) may provide you a vehicle to describe the larger class of fairy conditions.
If you establish a Madrasi version of each unit (Pawn, Knight, Grasshopper, etc), then you can describe the Madrasi condition (or Rex Exclusive Madrasi) by applying it universally to all units in the problem (or to all except the royal units).
So, a fairy condition might be modelled as nothing more than a characteristic which applies universally to the units, and the characteristics of the units become your fundamental building block to describe all fairy conditions.
There are some advantages to this.
You can then describe a problem which contains a Vertical Cylinder Rook on the same board with a Horizontal Cylinder Bishop (note: you can not describe such a problem if you insist these must be considered fairy boards).
These are the considerations you want to investigate, if your aim is to honestly classify fairy chess.
You want not merely map the known world of fairy chess, but to create a classification which will cover combinations not yet discovered.
That, by the way, was the beauty of the Periodic Table of the Elements -- it didn't merely map the known elements, it correctly predicted what elements were yet to be discovered. That's why that particular classification of elements (among many competing classifications) has endured!
The best classification will be capable of describing fairy combinations not yet seen (and might even predict what we've been missing).
But, even more interesting is the fact that there exists a arbitrary duality in how you might model a Circe Unit (say a Circe Pawn, or a Circe Queen).
Suppose my Circe-Pawn captures your Circe-Queen.
One way is to model rebirth as a function of the capturing unit.
This is one valid method to allow for instant rebirth of your Circe-Queen (on its game array square, if unoccupied) as a consequence of the Circe-Pawn's established rules.
But, if a Circe-Pawn captures an Orthodox-Queen, the Orthodox-Queen is reborn!
The alternative is to model rebirth as a function of the captured unit.
This is an equally valid method to allow for rebirth of your Circe-Queen, except it's now a consequence of the established rules for the Circe-Queen.
But now, if an Orthodox-Pawn captures a Circe-Queen, the Circe-Queen is reborn.
A third option to model rebirth is to insist that it may only occur when both the capturing unit and the captured unit have the same Circe characteristic.
If the unit characteristics are your fundamental means to describe all fairy conditions (and perhaps all fairy boards), alternative methods to model this duality may require careful consideration.
But, once you honestly embark upon this mission to find an optimal methodology to classify fairy chess, you'd better buckle your seatbelt, because Kansas is going bye bye.
Everything you may think you know about fairy conditions, and fairy boards, and neutral units, and turncoats -- well, let's just say this: you're going to discover they are all a two-dimensional projection of a deeper reality (your understanding of these elements will be completely and irrevocably altered).
You have no idea where these arbitrary decisions may take you, until you are willing to take a few small steps in this process.
And when you try to relate your newly acquired understanding, almost NOBODY will understand a word you are saying.
Fairy chess classification is not a mission for the timid.
Mental astronauts are what you need to send on such a voyage (along with a good poet).
Anybody who gets easily queasy when solving a help-selfmate at sea, you may want to grab a shuttle and return to space dock.
Anybody lacking the courage to file an honest captain's log (for fear it will upset the landlocked problemists back on Earth), you are not Starfleet material. I hear the Ferengi are looking for alphabet mining entrepreneurs.
But, if you who have the right stuff to embark upon such a bold mission, into uncharted territory, well...
May you live long and prosper. May you go where no one has gone before!
ps: I almost forgot. Since you will not be taking any Romulan Ale...
| | | (30) Posted by James Malcom [Wednesday, May 5, 2021 16:00] |
Kevin, at this point, I must opine, instead of insightful remarks, you look as if you're wasting space with utterly long rants. You're making "stupendously" long posts in a thread that was meant to be a simple announcement with some discussion. Please, if you must, take these annoying scroller rants to your own thread. Let's try not to start a flame was here.
| | | (31) Posted by seetharaman kalyan [Wednesday, May 5, 2021 16:59] |
I agree with James Malcolm.
Instead of wasting his breath and writing 1000 sentences on general rant, Kevin can disclose his intended definitions and his own superior classification system (which he has been secretly preserving for so many years 😀), so that the team can accept or reject citing their reasons.
I suppose the reason why Julia started the thread is to receive such a constructive response.
| | | (32) Posted by Kevin Begley [Wednesday, May 5, 2021 17:35] |
LOL. There's nothing less courageous than the troll who both insults you, and advises you against a flame war.
You enjoin this thread stating NOTHING with any relevance to fairy chess classification, and you don't even realize the irony when it is you suggesting others do not belong here?!!
Need I say more?
When you challenge a corrupt system, this is all you get -- personal attacks from trolls who have no capacity to make a single counterpoint.
Is this the best we can expect from a problem chess forum?
You are projecting. Everything you just said about me is a manifestation of how you must feel about yourself.
You're not even involved in the discussion.
You needn't be here if you have nothing to contribute.
Unlike you, I have plenty to say which pertains to the subject at hand.
Is that a source of envy for you?
Why are you here?
I'm here to discuss fairy chess classification.
What in the hell brings you here?
Did somebody tell you we're giving away tribbles, or what?
Were you recently involved in a transporter malfunction?
Report to sick bay (and bring your twin with you -- maybe the Emergency Medical Hologram can make you whole again).
Get off the bridge, cadets.
Go classify Fizzbin (maybe you can manage a single definition there, before the Tuesday rule change -- just don't ask what the odds are against you).
| | | (33) Posted by Rosie Fay [Thursday, May 6, 2021 08:59] |
Kevin is right to challenge Julia's classification. But I am not sure Kevin has it right either.
How come series-play is play in a chess variant? How do the rules of this variant differ from those of chess? Series play is an alternative turn-passing protocol, rather than play of moves in a variant game. Series play is not progressive chess (or anything like it). In a series-mate in, say, 4, Black never has any hope of getting the turn, let alone one of 5 moves. Another difference from progressive chess is what is optimized. In progressive chess the rules specify the number of moves in each turn (except when check ends the turn early) and the player must optimize the position. By contrast, a series-mover specifies the goal (mate, or whatever) and the the task is to do this while optimizing the move-count.
"Aims may be detected without any alteration of game rules." Now I can see this point of view, but I don't see how it can be consistent with treating series-play as play of a variant game. If the laws of chess are altered by the game's objective being changed, isn't the result a variant game?
There's at least one phenomenon which shows that problem classification would benefit from re-negotiation of the boundary between stipulation and fairy condition: a phenomenon which is traditionally regarded as part of the stipulation, but which I think is really a fairy condition: "consequent". "Consequent" imposes a condition on the series-mover which affects the legality of each individual move in the series.
| | | (34) Posted by Hauke Reddmann [Thursday, May 6, 2021 10:17] |
@Kevin: The wording, the wording, the wording... (Subtle Blakes 7 pun)
You e.g. use "dishonest". I say, lets assume for teh lulz that the
classification of Julia et al. has scientifical shortcomings
(this is the worst critique I'm able of). This would imply that they
made a *honest* mistake. That even holds if they read your warnings
beforehand and chose to ignore them. Fairy classification is not
judging the trueness of 1+1=2. (If we could do it by feeding it into a
theorem prover, we wouldn't have needed it in the first place.)
Thus they have they right to think different on the matter.
I don't even deny that fur has been flying in scientific discussions
when the standpoints were incompatible, but please name me one
incident where it helped science.
You assert your system is the only logical. Hic Bajor, hic salta.
I worked with theorem provers like COQ (other exist too), they
can make (not on their own yet!) a formal proof of the truth of a
logical assertion. I'm sure that questions like "What is a stipulation"
can be completely formalized by defining axioms. (I'm very well of
Mr. Gödel, BTW - you can't prove the consistency of your system in
itself.) Mind trying this approach?
The burden of proving that the Fairy Chess Classification Project
is useful (to set the lowest hurdle) is on the makers. The burden
of proving that your system is better is on you.
And now I will do like Morn and be quiet :-)
| | | (35) Posted by Kevin Begley [Thursday, May 6, 2021 10:43] |
Let's say you start a sandwich/pizza chain, which intends to offer customers the widest choice of ingredients.
You provide so many choices, your marketing team insist your menu must "classify" these choices for the benefit of your customers.
Do they really want you to provide a "classification"?
So, you start by identifying the main classes.
Breads, Meats, Produce, Cheeses, Sauces, and other options.
You offer so many ingredients which fall under produce, your marketing team insists upon a subdivision of a) fruits and b) vegetables.
The problem is: where do you list tomatoes (a fruit commonly mistaken for a vegetable, along with squash, olives, cucumbers, avocados, eggplant, peppers, and okra)?
Botanically, tomatoes are clearly fruits because they form from a flower and contain seeds. The tomato is a fruit by the very logical definition of fruit, despite what the clueless lawmakers in the US Congress have declared. That said, there are chefs who will argue that tomatoes are often utilized as a vegetable in food products (they're not considered a proper addition in a fruit salad, for example).
A thing does not become a vegetable by virtue of the fact is may be utilized as a vegetable. Chefs lack the authority to redefine the term vegetable (which has been clearly defined). What should happen -- what should have happened decades before you were confronted by this dilemma -- is that chefs should have developed a word of their own (which clearly defines a subclass of fruits, sub-diving those which may be utilized as a vegetable in food products from those which are considered fruity-fruits, or perhaps qualifying their fruitiness).
What to do?
Maybe you decide that your intent is not so scholarly -- your aim is to enable your customers to locate a specific ingredient where they might expect to find it located. So, maybe you decide to list a tomato as both a fruit and a vegetable.
Does that really serve your customers?
Aren't you part of the problem, in perpetuating their ignorance?
Some in your marketing department might favor your decision.
Others might recognize it exposes you to a hidden liability: if you don't at least make clear your awareness that a tomato is actually a fruit, a competitor may advertise that your chain doesn't even know the difference between a fruit and a vegetable!
ASIDE: The Subway chain lists tomatoes as a "veggie." Technically, they did not say "vegetable." Maybe "veggie" would be a good word to describe both fruits and vegetables which can be utilized as a vegetable in food products (but, unless the public is made aware this is not merely a colloquial term for vegetable, it can be reasonably argued that their menu perpetuates an ignorance about food).
This problem is not unique to fairy chess elements.
Reflexmates and series-movers are commonly mistaken for a stipulation, despite the fact that there exists no logical definition of stipulation which can explain such a grouping (it's an historical error which has been perpetuated by unscholarly proponents).
The difference is: Subway (and others) do not purport to provide you a classification of ingredients on their menu!
When a group of volunteer fairy problemists gather, intent to produce a "classification" of fairy elements, it is expected they will produce something other than a mere listing which falls under false labels.
No classification of fruits and vegetables by botanists would get this wrong.
They will provide clear definitions, such that anyone may determine the difference simply by applying the criteria found in their definition.
If your aim is to provide a graphical "menu" of fairy elements, and you have no concern what profound ignorance your listing may help to perpetuate, you may consider this first draft a success.
It doesn't teach us much, but it does offer a substantial improvement over the popeye readme file (it's better for marketing).
If you aim to provide a classification (which is a scholarly endeavor), OK, you fell short in the first draft (and I sincerely hope you will regard my strong objection as a constructive criticism).
My hope is that your aim is the latter, and that you will succeed in that endeavor.
I also hope that Subway will reconsider their menu -- because it offends me to know that they would be willing to mislead their customers about anything they eat. Restaurants have a responsibility to inform their customers about food choices, and it should be considered a betrayal when they violate that trust in the slightest.
I feel even more strongly about that when it comes to fairy chess problems -- which I consider a food for the brain!
I hope a group of chefs will establish a term which communicates the concept of a "vegetable utility", without incorrectly applying the term "vegetable" (or its derivatives, such as "veggie") to that which is not a "vegetable."
That said, I think I can retire from this thread (unless another troll comes back at me, under false pretenses, begging for a cessation of this discussion in a manner which serves only as a cry for more).
Should anyone care to further discuss the scholarly matter of fairy chess classification, I'm happy to volunteer all I can by way of private messages (or email). There's nothing secret about the classification system I have developed over the years, for my own sake -- but, I am not going to produce scholarly documents for a group that has yet to demonstrate any capacity to appreciate their value (I regard that petition issued as laughable and desperately self-serving).
I do sincerely hope this group will manage to improve their standing dramatically, because a proper authoritative body could offer substantial improvements, and create an atmosphere which welcomes participation.
You may have initially proved less worthy than a group of complete rookies, but as Captain Jean-Luc Picard once said: "Things are only impossible until they're not."
| | | (36) Posted by Kevin Begley [Thursday, May 6, 2021 11:30] |
@Rosie,
Thank you. I appreciate your point of view, and the issue you raised.
You might define move alternation as a consequence of stipulation.
Then, of course, chess rules become a static -- they would describe a frozen state.
I do not deny that is one possible way to logically classify chess problems.
But, if that's the case, Progressive Chess would then share the static rules of chess (its only difference is move alternation).
If rules exist in a frozen state (with no move alternation associated with them), Progressive Chess will share those rules (entirely).
Therefore, such a classification would render Progressive Chess as an orthodox form, but the game itself is based upon an alternative stipulation.
There are such games. For example, consider the game where one player is required to force selfmate (not checkmate) upon their opponent.
That could be considered orthodox chess (not a variant, not a fairy condition), but the desired aim (or the "play" motivation) of one player has changed.
If move alternation is not inherently a part of the rules, Progressive Chess might also be classified as such a game.
Also, consider the "game" (for beginners) where the aim is to be first to get your King onto the promotion rank.
Again, this is not a fairy condition (if you define a fairy condition to be a rule change) -- it's merely a game with a variant aim.
There are arbitrary interpretations possible, and I do not deny there may be good value in considering different forms of classification.
I have made no claim that my interpretation is in any way superior (contrary to false attributions a hostile party has attempted to cast unto me in this thread).
Depending how you chose to arbitrarily define move alternation, the Maharajah Game (where you have only one King, but may make two moves -- with the possibility of moving through check) might also be logically classified as an orthodox form.
On the other hand, if you come to this classification project with a specific, biased motivation (to make complex definitions in such a way that a particular favored element be deemed orthodox, or stipulation rather than a fairy conditions, that is possible).
If you can do this without failing in your duty to provide concise definitions, it might not be such a bad thing to favor historical precedent.
But, your primary duty is to provide concise, logical definitions. If you violate that, in favor of achieving a biased outcome, you have betrayed the mission of classification.
For example, you could classify the rules in such an arbitrary way, so as to proclaim Chess-960 is orthodox (and reduce chess to a subset of that larger orthodoxy).
I don't think anyone can achieve such a classification without overtly betraying a nefarious motivation.
You can not produce an elegant definition which renders a Reflexmate as a stipulation, nor can you produce an elegant definition which renders a Series-Mover as a stipulation (unless you are willing to allow Progressive Chess to come along with it).
So long as the boundaries of classification are unambiguously defined, based upon objective characteristics, I expect we (myself included) stand to benefit (read: learn) from any logical classification proposed.
If somebody wants to labor to achieve a specific classification outcome, I'm so confident they will prove unable to realize their desire, I'm willing to welcome all nefarious attempts.
They are going to quickly discover that achieving their bias requires far too much work (especially when a small group of complete rookies could achieve a logical classification system with a few days work).
And, think of it... should they manage to achieve such a framework, they'll be forced to endure every new fairy element invention which is designed to exploit their own loopholes.
At least that would provide a level playing field (and I'd consider that enormous progress).
| | | (37) Posted by Kevin Begley [Thursday, May 6, 2021 11:51] |
@Hauke,
>"You assert your system is the only logical."
Please, stop making false attributions.
That is false. I have NEVER asserted my interpretation is the only one logical.
Quite the contrary!
I said that ANY logical classification system must provide unambiguous definitions.
If you doubt this proposition, I can provide you numerous quotes from experts in the field of classification science.
That's the whole point of classification.
Where has logical classification benefited science? Ha! EVERYWHERE.
Taxonomy is the basis of BIOLOGY.
The periodic tables of the elements (a classification of elements) is the foundation of CHEMISTRY (and from that organization alone, we have made considerable advances).
If you want to see what happens when your community fails to manage its own logical classification, I saw an excellent article in Scientific American, some time back, about Taxonomic Vandalism. Just google that term, and I'm sure you'll quickly gain an appreciation for what can go wrong if your community is unwilling to invest itself in establishing an honest framework and establishing some mechanism to logically govern their conventions (including even naming conventions).
You can not possibly argue that Julia's group has produced a classification of chess problems.
A menu, perhaps. A list of options placed under undefined (read: meaningless) labels does not constitute a classification!
It's not even close.
Do you really require assistance in understanding the value of classification in the fields of mathematics or logic?
If so, you're inviting me to have a field day...
Not only is classification a cornerstone in human learning, it is fundamental for learning machines!
By contrast, you'll learn almost nothing interesting from lists organized to reflect a subjective favoritism -- there's a distinct difference.
A list of your favorite things which fall under an undefined label tells me nothing about the relationship of those elements (it only tells me something about you).
| | | (38) Posted by Kevin Begley [Thursday, May 6, 2021 12:11] |
@Hauke,
>"The burden of proving that your system is better is on you."
Another false assertion.
I have no such burden, and I have made no such claim.
You obviously are having difficulties comprehending what I have clearly stated, countless times, in this thread.
I developed my own classification (an actual classification -- not a menu, not a list which falls under meaningless labels, but a system which is rooted in unambiguously defined terms) to serve my own purpose.
I have no zero responsibility to prove it has value to anyone except myself (and I have already proven to myself that it has substantial predictive power).
But, I have never claimed to possess a "superior" classification.
I have a classification. Julia's team does not. Period. Full stop.
They have provided you a menu, without ANY definition.
According to the philosopher, John Stuart Mill, "classification is based on 'Definition' which states the essential
attributes or features of a class."
The procedure he describes is very simple: discover common and essential characteristics of elements, and then classify accordingly.
He goes on to suggest that scientific classification is always based on (wait for it....) 'Definition'!!!
Logical classification requires definition for two reasons:
1) So that independent agents may correctly identify the proper categorization of an element (including elements not yet known), and
2) So that relationships of elements contained therein may be illuminated.
As I have previously stated: should you be presented with a new fairy element, Julia's work can provide you no basis to determine which group the new element should be located.
That's not a classification. Classifications are required to provide you a parsing mechanism (generally definitional).
They failed. What they produced is an undeniable failure, if their intent was to provide you a classification of fairy chess elements.
What I recognized is that you can classify fairy elements using unambiguous definitions (and mine is neither the ONLY way to classify, nor have I ever claimed what I use is superior).
I only claim that I know it is possible to provide an actual classification of these elements, if you are willing to invest some time to produce some coherent definitions.
Not axioms, definitions. Don't throw that insane Godel crap at me, for the sake of all reason!
You can't prove there exists a set of rules for chess, based upon axioms. Yeah, yeah, yeah. We get it... nothing is possible.
I'll let you deliver this terrible news to Ian Nepomniachtchi. He's going to be heartbroken, I'm sure.
But, uh, I hope for your sake he doesn't know any Russian mathematicians, because they'll know you have raised Godel to the power of the Xeno Paradox.
It's a waste of keystrokes to go there with me, my friend.
Save that delicious trolling gravy for the mathematically illiterate. They'll eat it up.
I have been through this absurd debate already. Don't make me relive it.
To suggest no definition is logically (or axiomatically) possible constitutes a profound misunderstanding of Godel.
More importantly, this takes us entirely off the subject of a fairy chess classification.
If you want to argue it's impossible to achieve any axiomatic foundation for the rules of chess (nor for any fairy element), and therefore no logical definition (and no logical classification system) is possible, then you should present your case to the group which currently endeavors to violate the laws of logic.
All I am saying is: provide unambiguous definitions, so that people may independently discern what element belongs in what group.
That's what a classification provides.
I never said my (arbitrary) classification system is the only logical possibility, nor have I ever suggested mine is superior.
So, kindly retract the false attributions you have made to me. I would very much appreciate that.
In my view, the best measure of any classification system would be to compare implementations.
And if that's the measure, I believe Christian Poisson probably has the best case to make such a claim.
No other implementation (to my knowledge) has proved nearly as reliable when two (or more) fairy elements are combined.
And, no other implementation offers nearly as many options.
That's not proof his underlying (and certainly proprietary) classification system is superior, but to date, it would seem the best implementation (which suggests he has a very strong grasp on what belongs where in his implicit classification hierarchy).
He (and some Popeye programmers, too) probably has a stronger appreciation for establishing a well-defined classification of fairy elements.
He has been confronted with the implementation of, well, virtually every known element, whereas I have a focused only on a few (and of those, I have implemented nothing comparable).
Ask them whether a reflexmate can (or should) be classified as a stipulation, or as a fairy condition.
I found a consistent methodology to classify fairy elements, and to model a complex stipulation search function (wherein lesser stipulations may serve as an aim). It predicted a number of (uncommon) stipulations I only later discovered in the Win Chloe database.
To me, that demonstrates my methodology has value (to me).
I am not here to advocate my vision (though, I have freely shared the foundations of my approach, in post after post, for years).
I'm certainly not here to compete for the best classification system, nor to see my name on atop a community standard.
I'm here only to advocate for a logical classification, so that we might all enjoy an equal playing field, so that we might provide future enthusiasts with an honest framework which values innovations, so that new inventors may consult a set of community preferred conventions (which make it easier to navigate the already considerable, and constantly growing, set of possibilities).
This is hardly impossible, and would be a good thing for everyone.
I've been advocating for this community to come together and develop such a classification for at least a decade.
I came to this thread overjoyed by the discovery that Julia had managed to get that ball rolling (and she deserves tremendous credit for it)!!
That's not easy thing to achieve, believe me!!!!!
When I discovered this classification potential had been squandered in favor of bowing to the status quo, well, I decided the best thing I could do was to tell the truth about what they have (thus far) produced -- in the sincere hope that they might want to deliver on the promise made.
If they don't come through on that promise, well, I doubt anyone else could assemble such potential (not in our lifetime).
It's quite sad if we stand here on the verge of such a beneficial thing for the fairy problem community, but we lack the courage to insist that our divisions be clearly defined.
I am confident, if they find the courage to take that one small step, they can not help but make a giant leap.
I'm trying to motivate them to get this right, and to appreciate why it's not yet in the ballpark.
What are you trying to do?
You want to tell them it's axiomatically impossible?
You want to tell them it's pointless?
You want to discourage them by misattributing to me these false claims of a superior product (sowing divisions)?
You must love chaos (and I don't mean in the mathematical sense).
And I'm the guy who is accused of not being constructive here???
I'm the guy who is accused of not contributing enough to this effort???
I can contribute far more as a non-participant, because I have established a basis for comparison (note: I did not say for competition).
I am not waiting in the wings to pounce with an alternative vision. Far from it!
This product can't belong to me, not to any one problemist -- it must come from the larger community (there must be a buy-in).
I have no territorial desire to put my name on it.
I have no desire to steamroll anyone with an advocacy for my own vision.
I'm happy to share what works for me, if any of that might help (I'm not concealing anything).
You're free to take what you want -- again, most of my vision came from discussions with other problemists (who had a differing perspective).
I took what worked for me, and I made some small improvements (for my own sake).
I'm not sitting on a secret formula.
But this I do know: John Stuart Mill has it right.
If you do not provide definitions, then future problemists will be provided no logical means to continue using a classification which fails to accommodate new discoveries.
When taxonomy is based upon unambiguous definitions, it tends to endure new discoveries.
When a list, by contrast, is based upon bias and favoritism (with no discernable boundaries), it either fast becomes obsolete, or it systematically poisons all potential for new discovery.
Oh, and have I mentioned... the AI is coming for us?
Yeah... we don't have an eternity to get this right.
Ironically, the only thing that discourages AI is the fact that our award system is more subjective than our classification system.
| | | (39) Posted by Neal Turner [Thursday, May 6, 2021 21:28] |
I gave my own views at the very start of this project, expressing my doubts about its feasibility.
This thread has driven me to crystallize my thoughts...
Open up a magazine and we see problems of different types - we will call these different types (including orthodox) 'variants'.
We don't single out boards, pieces, conditions or stipulations - they're all in the mix together, combining to produce the variant.
Each variant has its own distinctive character, its own peculiarities, giving each one a unique flavour differentiating it from all the others.
Now what can we do with these variants?
We can list them:
#3
s#6
h#4
#2 Breton
#3 Circe
h#2 Anti-circe Grasshoppers
h#4 Madrasi Half-neutral pieces
hs#3 Horizontal Cylinder board
Ser-h#12 Chinese pieces
s#2 SAT royalgrasshoppers (recommended!)
r#3 Gridchess
h=1.5 Patrouille Supercirce Madrasi EinsteinChess Sentinelles PionAdvers AndernachChess/AntiAndernachChess (yac #330481!)
etc, etc, etc...
What else can we do with them? - nothing!
But don't we want to explore the nature of the different stipulations, conditions, pieces and examine the relationships between them to build up a logical structure?
No! Because there are no stipulations, there are no conditions, there are no pieces - there's only the variants.
Each variant is unique and stands on it's own, and any attempt to classify them is both meaningless and futile.
That's it - you can feel free to insult me.
| | | (40) Posted by Kevin Begley [Friday, May 7, 2021 08:51] |
@Neal,
I do not share your perspective (as I happen to believe it worthwhile to explore the nature of these elements, and attempt to discover a concise model which describes them), but I can certainly appreciate your position.
You have articulated a cogent way to view the fairy universe.
While I can provide no direct evidence that my perspective is any more correct (read: I have no intent to persuade you otherwise), I will tell you what has persuaded me to adopt a view which is diametrically opposed to your own.
First, if it's worthwhile to model the physical universe as a collection of fundamental matter/anti-matter particles (the fermions), which are acted upon by a limited set of force particles (the bosons), in such a way that force-matter particles can exist in combinations, then why wouldn't the analogous effort (to achieve a more fundamental understanding) have benefits in the (human invented) fairy universe?
Obviously, I can't prove this process is necessarily beneficial. Moreover, it's fair to presume that a fictional universe may be less inclined to exhibit any logical structure (it would be absurd to expect any logical utility could be derived by studying Tinkerbell's relationship to fairy dust in Peter Pan's fictional universe).
Frankly, the most remarkable feature of our own universe is that it actually does exhibit logical (and mathematical) structure.
But, the fairy chess universe lies somewhere between Peter Pan's fantasy universe and our physical universe.
The key difference in our polar opposite perspectives may be rooted in our differing estimations for which of these two universes (Peter Pan's fictional universe, or our physical universe) is the nearest neighbor to the fairy chess universe.
I believe fairy chess is much closer to our physical universe, where the underpinnings of logic will provide us with useful patterns and relationships.
Therein lies the key point -- we can actually measure whether such a modelling process has value.
We can demonstrate that the study of science in our own universe has value, because scientific models have the power to make incredibly accurate predictions about the future.
A science lacking all predictive capacity is approximately equivalent to a very rigorous religion (it might provide us with beautiful patterns, but it unleashes for us no predictive utility).
I can not prove to you that a deeper understanding fairy elements will have predictive utility.
However, I can tell you that it has demonstrated something very close to this same power (nearly analogous to prediction), in my experience.
For example, simply by realizing "Take & Make" should be logically classified as an anticirce form, I was able to predict that there should exist (at least one) circe interpretation.
That's not exactly a prediction of a natural phenomenon.
It's not like I predicted when a comet would be appear in a specific location of the sky -- I predicted only that a human invention might be expected to occur in a specific place (and it might be of interest to fans of the related "Take & Make").
It's worth noting that I certainly can not claim this process will only predict fairy conditions which make logical sense (I also tried to discover an anticirce parrain, and let me tell you, the most elaborate attempts to achieve this have resulted in some hideous abominations).
But, from my perspective, in my experience, classification has helped me to imagine fairy elements I might never otherwise have dreamed.
Earlier in this thread, I made reference to a second example.
When I first realized that some complex elements could be modeled as if a sub-stipulation were the aim (within a larger stipulation), it immediately opened my eyes to vast new possibilities.
Instantly, it hit me that a "help-self-helpmate" stipulation might be possible in the future (moreover, I knew exactly what this might mean -- which I'd have never guessed previously).
In looking for a concise way to model complex fairy elements, I had suddenly unlocked an entirely new appreciation for untold treasures. And then, to my great surprise, I discovered some of these "futuristic" stipulations I had predicted were already in use (very uncommon use, to be sure, but in use nevertheless)!
Did my model predict them?
I can't exactly demonstrate that my stipulation model predicted these stipulations would "someday" (past, present, or future) be discovered, but from my standpoint, I have no doubt that modelling stipulations proved to be a very beneficial endeavor (nearly to the degree of demonstrating a predictive capacity), because they enabled me to imagine fairy elements I would otherwise have never dreamed possible.
I could say much more about the value of appreciating relationships between fairy elements, as well.
One simple example:
I was browsing a database of economical circe problems one day, when I happened upon something quite extraordinary.
At first I couldn't believe such a theme was possible with so few units (8 in total).
After a short time considering the mechanism, I began to wonder: why is this not more efficient (why couldn't the authors eliminate the one additional unit, and get this into miniature form)?
So, after a few hours motivated to find the more economical version I was sure could be realized, I gained a greater appreciation for these particular composers -- they had it right.
Maybe I should have expected that from these authors, but trusting gains you nothing.
I thought I'd put the whole thing out of my mind, when suddenly it hit me: there does exist a sister-condition!
Sure, everything would be inverted, but the mechanism should still hold, and maybe there I would find the miniature...
And I did find a 7-man version (I never published it, of course, because it was hardly original, and hardly offered an improvement upon the original version). Also, in fairness, I should mention that the sister condition did not exist when these authors published their 8-man problem.
But the principle remains.
Logical Classification can help make you aware of fairy element relationships.
And, should you want to achieve a specific idea, knowing these relationships can potentially benefit composers.
These relationships are also more welcoming (you can "speak" more fairy elements in a faster period by making such associations).
If you want to achieve a mastery of fairies, you can expect to benefit from a less globular appreciation of your paillette wheel.
| |
Read more... | Page: [Previous] [Next] 1 2 3
MatPlus.Net Forum General The Fairy Chess Classification Project @ Julia's Fairies |
|
|
|
ISC 2024
