﻿﻿ MatPlus.Net

Website founded by
Milan Velimirović
in 2006

5:06 UTC
 ISC 2021

Remember me

 CHESS SOLVINGTournamentsRating lists1-Jul-2021
 B P C F

MatPlus.Net Forum General What is Fairy Chess?

Page: [Previous] [Next] 1 2 3

### What is Fairy Chess?

A question asked has apparently drowned in a thread with a curious title. In order that the question, in my opinion important, gets the attention it deserves, it is here in its own thread. By repeating the question I rest my case.

- - - - - - - - -

Question concerning the definition of fairy chess in the Codex for Chess Composition (the essential parts, i.e. Chapter II and footnotes, copied below). In your opinion, can we from the text conclude what fairy chess is?

Chapter II - Types of Chess Composition [8]

Article 5 - Classification according to Stipulations

Chess compositions can be classified into several groups according to their stipulation. Besides the historically developed groups, viz studies, direct mates, selfmates and helpmates [9], further groups [10] have developed [11]

Article 6 - Special Types

Additionally, and independent from the classification according to Article 5, there are a number of special types, including:
(a) Retroanalytical chess compositions
(b) Mathematical chess compositions
(c) Constructional chess compositions.

Article 7 - Classification according to Rules

Furthermore, chess compositions can be classified into those which apply the FIDE-rules of the game of chess [12] and those which apply modified rules [13,14]

Footnotes

8 Articles 5 to 7 are not intended to be exhaustive. Other classifications are possible and also practised, for example according to the material used (miniature, minimal, Meredith etc.) or according to other criteria.

9 According to this classification, examples of frequently used stipulations are:
(a1) White to move and force a win, without restriction to a specified number of moves (studies).
(a2) White to move and force a draw, without restriction to a specified number of moves (studies).
(b) White to move and mate the black king in a specified maximum number of moves (direct mate).
(c) White to move and force Black to mate the white king in a specified number of moves (selfmate).
(d) Black to move and cooperate with White in order to obtain a mate of the black king in a specified number of moves (helpmate).

10 Further groups are, for example, stalemate or series stipulations etc.

11 Compositions other than studies are usually called problems.

12 Presently, the rules for the game of chess as agreed during the FIDE-congress 1996 in Yerevan are valid. Relevant for compositional chess are Articles 1 to 5.

13 In this context, the terms orthodox, heterodox, fairy and exo are used.

14 Modifications of the FIDE-rules may for example consist in:
(a) Rules (conditions) on which the composition is based (for example maximummer, circe, seriesmover).
(b) Pieces used in the composition (for example nightrider, grasshopper, chinese pieces).
(c) Chess space on which the composition is based (for example chess board with 10x10 squares, cylindrical chess board, multi-dimensional chess boards).

If I may ask- can you have a fairy chess study? I have never seen one, but I can't imagine why not.

There are several studies which employ fairy elements -- in fact, non-orthodox studies predate FIDE chess by a thousand years (so, the real deviation is orthodox Chess itself)!
I don't find many in databases -- the most common fairy condition employed in studies seems to be Giveaway Chess.

Probably, there is a favoritism for fairy elements which are highly playable in games -- for obvious reason: this yields some audience of chess variant game players, and helps to develop some basic endgame theory.

If you're really serious about pursuing this field, you might want to seriously consider creating a fairy EGTB.
This requires a large investment (not only in time!), but the benefits would be considerable.

As to the thread's question: I maintain that there is no valid definition of "fairy chess."
If you define it to be "not orthodox," then you have to address the evolving nature of what is accepted as orthodox.
There have been significant rule changes in orthodox chess -- even recent changes in the FIDE Rulebook, dealing with dead positions, have altered the intent of several orthodox chess studies (are these now to be considered fairies?).

Chess should be categorized according to the nature of the problem (read: what is opponent's reaction to the goal -- helping or opposing?), and sub-divided according to valid aims and valid stipulations.
Unfortunately, a number of problem editors have failed to define basic terms (especially: what constitutes a fundamentally valid aim/stipulation), and this has resulted in a number of invalid goals (e.g, reflex-, series-movers, parry-movers, cap-zug, h##n, etc).

There is a difference between problems which use a fairy board (vertical cylinder/haan/10x10 board/holes/etc), a fairy condition (circe/madrasi/series-movers/parry-movers/etc), a fairy constraint (maximummer /pawns on the 1st rank can move /extra-long castling is allowed/promotions limited to fairy elements/double-capturing en passant are legal/etc), a fairy piece (nightrider/grasshopper/etc), a fairy stipulation (helpmate/selfmate/ProofGame/A->B/etc), and a fairy aim (stalemate/capture/check/etc).

An intelligent fairy invention should appear as a singular form -- and any complex conglomeration should be listed by the most economic classification technique (that is, as a collection of its fundamental fairy elements)!
The conglomeration of fundamental elements should not be absorbed into a singular (and invalid) stipulation (as has been the recent practice of some ill informed fairy inventors).

Inventors should be discouraged from this practice by the CODEX.
The false economy which results constitutes an invalid method of categorization, and undermines any definition of fundamental terms (stipulation/fairy condition/etc).
The problem is, the CODEX has consistently failed to establish these fundamental terms -- therefore, what was implicitly known for the larger fairy community, is being undermined by misinformed newcomers.

This practice assures a future categorization failure -- as more inventors resort to this unmanageable technique, the false economy spirals toward catastrophe (instead of being able to seek the definition of a few fundamental fairy elements, you would be required to learn several redundant components of each complex invention -- as a result, the fairy community would lose all categorization by fairy subdivisions).

It is plainly obvious that this ill considered practice is destined to fail; yet, there are those who can not see beyond the short-term form economy of their new inventions (the inventor doesn't want to list 3 fairy elements, when they can obscure them all as if it were a single stipulation -- regardless that it fails to adhere to the fundamental definition of stipulation).

It's high time the CODEX took steps to avoid this practice, by defining the fundamental terms (for those who are content to pretend an inability to grasp them).

We were handed down an implicitly logical, universal classification system -- we can not afford to allow a few new inventors (and several unwise editors, unfamiliar with fairies) to undermine such a valuable inheritance.

To preserve our logical classification system, I suggest the following:
1) clearly define the universal terms already stated in the CODEX,
2) new inventions can be encouraged to conform to the categorization system, by providing an important endorsement (from the delegates), for only those fairy elements which adhere to our established, logical and universal form.

Everybody knows this is the right course, and it's not much to ask that the CODEX provide a clear definition of what terms it already uses!
I can only suppose that progress continues to stall, because the corrupting politics of awards/titles -- the deeper interests are invested in preserving a distorted award/title system.
What this community fails to realize is, this practice is fast eroding any value held in its own awards/titles.

Think of all the trees cut up & mailed as self-nominations to all of the FIDE Album judges, for each and every independent album, every 3 years... then, think how little we devote ourselves to the viability of our own codex, or in maintaining a logical categorization system, or in providing newcomers with basic definitions of the terms we use in our codex.

In a word, it is shameful.

What's worse than the alleged CODEX issues is the fact that all fairy discussions are dominated by one person who resorts to chastising, shaming, ridiculing, and "shouting down" anyone who dares to oppose him.

Some fairy endgames (not in giveaway chess):

P.Petkov K.Gandev, 1973
(= 4+5 )
win, Circe

This one from FIDE Album 1986-88:
Jürgen Tschöpe, 1986(= 2+2 )
win, Dynamo

N.Petrovitch, 1960
(= 2+10 )
win, white and black can play two moves in turn

No Dan -- you don't really believe that's how I discuss these things in this forum.

Furthermore, it is not my intent to single out your inventions.
We've been friends a long time -- I think you well know there is no personal animosity in my discussion here.
I would much prefer to use other examples, but two of your inventions offer the best illustration of how we (this community -- not you!) are categorically failing to preserve the logical classification system, which we implicitly inherited from the problemists who came before us.

That inherited system was not always perfect (for example, series-movers, and h##n are stipulations widely known to be mis-categorized) ...
I am happy to concede that your parry-movers have only followed the series- format (which I, and many others, consider to be a poor precedent).
Furthermore, I am happy to conceded that series-movers and parry-movers (along with several other fairy conditions, which affect move alteration) occupy a unique class of fairy conditions -- to the extent that they have some immunity to our standardized threat-notation.
And, I have voiced concerns about the misrepresentation of parry-movers (using threat-notation), in some databases (notably Win Chloe, and potentially others).

Furthermore, to the extent that you, a titled composer, create a precedent of unmanageable classification (and influence journals down the path of categorical decay -- where basic terms can no longer be defined), it becomes difficult to escape an honest examination of your two inventions (parry- and capzug).

It's nothing personal -- let's not go down the personal path of false accusations again -- this only serves to delay/avoid the open, honest & critical discussion surrounding a troubling issue in problem chess (How are Fairies defined?); and, we all have an interest in resolving this matter (sooner, rather than later).

You well know it is certainly not my intention to shame you (or anybody else).
I have done nothing but help you with your inventions, from the very beginning -- and, I can provide countless correspondence for anybody who doubts that fact (in every single one of your MANY inventions)!
I contributed suggestions, problems, and testing, for virtually every one of your fairy inventions -- even fairly recently, I composed a parry-mover (as you well know).

I am not opposed to your inventions -- I am merely troubled at the unmanageable FORM in which you have attempted to classify two specific inventions... the inventions themselves I have never doubted.
You well know that my intent is only to correct an error in the classification form; and, we both know that the CODEX needs to define the terms it uses -- including "Fairy Chess" (how we define that term is the fundamental question of this thread)!

So please, don't take this discussion into the personal territory of accusations, such that people can not contribute openly.
The last thing we need is a cloud of false allegations -- nobody is out to shame you personally, or to shout you down.
If you want to contribute, I certainly welcome your participation -- in fact, I can think of nobody with whom I would prefer to reach some agreement on these matters!

As a matter of fact, on numerous occasions, I asked you to provide your own definition of the term: Fairy Condition.
It is perhaps not surprising that it has proven a significant difficulty for you to provide any definition here.
While I myself managed to establish a potentially viable definition of this term, I must concede that my simplistic definition failed to address some recent problem trends.
For example, even some orthodox studies have been impacted by some seemingly minor changes recently made to the FIDE rulebook (e.g., the intent to show a number of stalemates can be dramatically curtailed by the dead reckoning rule -- begging the question: how do we best classify/correct such an intent?).
It is not sufficient to define a Fairy Condition as anything contrary to the orthodox chess rules (which still remains an evolving form, and thus, constitutes a moving target!).
The truth is, the term Fairy Condition (which should be relatively easy in relation to the larger question of Fairy Chess), is far more complex than my simplistic definition had previously suggested.

My intent here is simply to make the case that we (in the problem community) should not give up on re-establishing a universal, logical classification system, where our codex provides clear definitions for the terms it uses.
If we can not define Fairy Chess, it makes no sense to divide problems (whether in the FIDE Albums, or journals, or databases) into false categories, according to indefinable terms.
We might be better served by grouping problems according to their MOTIVATION-TYPE (does the opponent help/oppose), and sub-divide according to aims (#, =, +, x, 00, ep, etc)...

However, whatever categorization we decide upon, we are responsible to clearly define our divisions.
To the extent that Fairy Chess commonly serves as the major dividing line, it remains shameful that this community refuses to provide any definition of this term -- a term which impacts FIDE Album divisions, magazine & tournament divisions, and database developers (therefore, it directly affects titles & awards).

How can problem chess ask to become an Olympic Contest, if we can not clearly identify (and separate) our own set of unique events?
The longer we go without addressing this failure, the more difficult it becomes to correct the old inventions (which were not logically classified).
Let's not run out the clock on this issue -- it's never honest acceptance to sunset an idea into practice.
Let's address the issue, honestly, in this forum.

I'd be happy to hear suggestions -- from ANYBODY! -- how we might clearly define our fundamental terms, and reestablish a classification system based upon universal, logical divisions.
And, I have zero interest in how this may impact titles (which I do not seek)... though, I certainly accept that the influence of self-interests may make this cleansing effort considerably more challenging.

Personal accusations do not inspire contributions to this open discussion.
Let's set aside ego, and focus on the issue; we might yet find better solutions.
I'm only here in aid of that pursuit... I certainly hope others will care enough to join in it.

@Kevin: There's ample proof of your personal remarks, directed at me (and others), on these forums. And your latest 700-word essay proves my point: this thread will be dominated by one person.

@Per: Apologies for getting off-topic. To answer your question: 'fairy chess' must be defined by what it is NOT. Paragraph (9) seems an attempt to do so, but then paragraph (10) makes me wonder.

@Dan,

A thread is not "dominated" by word count -- all that really counts are the ideas which contribute to a desired harmony.
I have given these matters considerable thought, but, I hardly have a monopoly on harmonious ideas (if I had all the answers, I'd simply present my findings here).

I'm afraid my findings are incomplete -- I have made some discoveries where harmony is lacking, and I have spent some time investigating possible resolutions; however, as noted in my previous post, some of my resolutions have proven overly simplistic.

You can hardly claim that my ideas are dominating this discussion, when I have essentially abandoned all my efforts to define Fairy Chess!
I'm afraid the best I can offer is to help lead this discussion into territory where resolutions are required.

Furthermore, if wordiness is a dominant issue for you, I'd be happy to consider any suggestions you might have, how I might express my ideas with fewer words (please direct those suggestions to private emails).

Nobody is accusing you of anything, or shaming you for anything, or shouting you down.

Your sensitivity need not be so heightened -- there is no reason that it should prevent and open discussion about fairy chess.
Remember: nobody is evaluating you -- nobody is even evaluating your fairy inventions.

This discussion may examine the FORM of some of your fairy inventions (along with any other relevant fairy element inventions) -- but, that is only motivated by a desire to obtain an improved classification system (read: something logical & universal, with clearly defined terms/divisions -- including a definition of "Fairy Chess").
It is well known that the form of two of your fairy inventions cause difficulties in defining some fundamental problem terminology (e.g., Fairy Condition and Stipulation).
As such, it may be necessary to examine the form that your inventions were initially given.
This is nothing personal, I assure you... it just happens that an examination of such fairy elements may be fundamental to this discussion.

There is no need for you to perceive this discussion as inherently hostile -- you are not a target of this discussion, and should feel in no way victimized by a diverse, open, and sincere discussion, here.

If you require any further accommodations, please elaborate them here.
I feel highly confident that everyone interested in this thread (myself included -- in fact, myself especially!) will go out of their way to provide for your needs.

This forum can facilitate a discussion devoid of personal egos, where everyone has an opportunity to voice their ideas.
We certainly should be able to examine the form of your fairy inventions, without it constantly being perceived as an attack upon your person.
How can we overcome this obstacle?

@Kevin: Since you think this is about my ego, and I yours, there's an elegant solution. I will agree to have my MPF posting priveleges revoked, permanently, if you will do the same.

@Dan,

No thanks -- I have no interest in censoring you.
I don't believe this discussion should have anything to do with ego.

And, there's no reason to perceive any of this as an attack on your person.
But, if anything, it is this which continues to dominate (and obstruct) an open discussion.

@Kevin.

Could you please clarify your objection to the series-mover concept? I don't see how it is any way flawed.

@Kevin -- Your continued denial was more or less much what I expected.

@Ian -- Series play (including pser-*) is technically a fairy condition, but it's represented in the stipulation. Hence the dire prophecy: "This practice assures a future categorization failure -- as more inventors resort to this unmanageable technique, the false economy spirals toward catastrophe".

PS @Kevin: I think your expectation of mutual-exclusivity in regard to (Fairy) chess problem taxonomy is overreaching and unrealistic. Consider the following problem:

Ian Shanahan: "The Problemist" 1993 {F1386}
7th place Wenigsteiner of the Year
~ In memoriam Norman Macleod ~

(= 1+3 )

R#18, Circe. Black moves only to check

Hauptplan: 1.Kb8 2.Ka7 3.Kb6 4.Ka5 5.Kb4 6.Kc3? Se4+ ... 11.Kc8 Bf5+!
Vorplan: 1.Kb8 2.Ka7 3.Kb6 4.Ka5 5.Kb4 6.Ka3! (6.Kb3? Bc2+ ... 12.Kc8 Sd6+ 13.Kd7 Ba4+!) 7.Ka2!! Bb1+ 8.Kb2 9.Kc3 Se4+ 10.Kb4 11.Ka5 12.Kb6 13.Kc7 14.Kc8 Sd6+ 15.Kd7 Bf5+ 16.Kxd6[Sb8] 17.Ke7 Sc6+ 18.Ke8 Bd7# (not 18...Bg6+: illegal in R#).

This really, in its move-form, is a type of series-mover - with intermittent checking moves by the other side (hence it's a precursor to, and antiform of, Dan's parry-series idea). But it could also be classified as a Fairy reflexmate. Surely both classifications are correct, and that we choose one or the other depending on the context and what one wishes to emphasize?

So long as problemists are aware of exactly what stipulations and conditions mean - including all of their prevailing conventions - the Fairy sky will not cave in. Subeditors of Fairy columns, such as Stephen Emmerson in "The Problemist", do a sterling job in clarifying such things in order to avoid ambiguity and confusion.

@Dan,

I have no idea what "denial" you are referring to.

It seems to me that you continue to remain hostile towards any discussion of these issues.
This is unfortunate -- it creates serious difficulties for people who might wish to openly contribute.
I have asked you to elaborate how we might overcome this obstacle; but, you provide no opportunity.
What more can we do -- you can't expect people to remain permanently silent (nor should ask them to join you in a pact of self-censorship).

There's no need to suppress opinions, here.

However, your extreme sensitivities seem to be inhibiting members from engaging in an open dialog.
Nobody wants to offend you, Dan -- this discussion has no need for any personal component.
All anybody wanted here is an opportunity to discuss these matters, in a civil, non-confrontational fashion.
Why don't we just agree to that, and we'll all get along fine, without further back & forth...
Shall we?

Would you please give people permission to critically examine the form (and classifications) of your fairy inventions?

@Ian,

Look in the Codex -- you'll find a number of terms that are left undefined.
Look at your journals, your Albums (the source of all awards & titles) -- you'll find your divisions are not clearly defined.

Nobody claimed these things are necessary to hold up the sky...
However, your argument -- that life will go on in the face of a poor classification system -- can be used to justify any number of poor decisions.

You can not presently provide an adequate definition of Orthodox Chess, let alone Fairy Chess.
Think how life might be improved, if you had clear definitions for these fundamental terms.
Think how life might be improved, if you avoided the redundancy of poorly formed, overlapping stipulations.

Further, your example problem fails to provide any reasonable demonstration of your conclusion (that it is impossible to achieve a clearly defined classification system).
While, I certainly appreciate your sharing the example, you should be aware that it demonstrates nothing (other than a redundancy problem within the present set of derived stipulations).
As a matter of fact, I (along with others, including GM Petko Petkov) had warned Dan of this redundancy mess, but Dan was still arguing that his Parry-mover invention was NOT a fairy condition.

Lots of time was wasted on such arguments -- for Dan, I suppose it seemed he was put on trial (which was never the intent).

If you want to avoid the redundant, unclassified mess that has resulted, Ian, you start by clearly defining fundamental terms (What is Fairy Chess?).
Then, you maintain logical divisions.

This is not an issue where Kurt Gödel's theorems apply -- as somebody once claimed (without any proof).
This does not even require a degree in library science -- it's actually quite easy to create a logical classification for chess problems.
It requires little more than a sincere desire... that's exactly what I hope this thread will establish (starting from the definition of Fairy Chess).

Are you really so against trying, that you would insist (without any proof) that a logical classification is impossible to achieve?

C'mon, Ian -- you should know better than to make such an argument.
If you're going to make this case, you have a heavy burden of proof to show it's impossible.
In fact, you'd have to show that the present chaotic classification is beyond improvement -- good luck with that!

My friend, the sky does not fall from improvement, either!!

KB: "As a matter of fact, I (along with others, including GM Petko Petkov) had warned Dan of this redundancy mess, but Dan was still arguing that his Parry-mover invention was NOT a fairy condition."

Not true. In the first post on this thread...

http://www.matplus.net/pub/start.php?px=1336387910&app=forum&act=posts&fid=tt&tid=643

... I clearly state:

"Section A:

Parry-series movers of any type (see article in StrateGems, Jul-Sep 2009, p. 150). No fairy pieces

@Dan,

I hope you will reread my previous post, and consider what I've said.

The debate about pser- started with Petko -- not with me.
He was the first to claim that this was a Fairy Condition -- not me.
You responded (in Parry-Hub) to Petko's article, stating, essentially that you disagreed with his conclusion:
Roughly paraphrased, you said:
Parry movers are no more a fairy condition than series-movers [which are widely -- virtually universally -- considered to be a fairy condition, themselves]

From there, we've had quite some debate on this matter.
And, we've both made some concessions -- for example, you recently admitted to me that the fairy condition "W/B moves only to Parry Check" was necessary (which is exactly the simple fairy condition I had suggested to replace your stipulation) -- and one that I had correctly predicted would be necessary (even with your stipulation).
And, I have admitted (in this very thread) that a fairy condition does not adequately cover cases where a player may idle (because threat-notation is an unsuitable form to show such problems).

[note to Ian: the redundancy of this duplicitous fairy-condition/stipulation might have been avoided -- which would have saved the popeye programmers from having to implement two forms of a completely identical idea. This clearly demonstrates the kind of improvements we can achieve, if we only think ahead, define our fundamental terms, and reforge the universal, standard notation we were meant to inherit from our predecessors. It's not a case of the sky falling if we don't -- it's a case of the sky being the limit, if we do elect to make intelligent improvements.]

But, I'm not here to re-litigate the parry-series form.
Suffice it to say this is a fairy condition, which *MAY* require something more (perhaps even a stipulation -- depending how the Codex defines this term) -- in order to direct solvers to the solution's intended notation form.

I'm here to find/suggest an improved classification system, where the Codex can provide clear definitions for all of the fundamental terms it uses.

It aint personal, Dan.
This issue is much bigger than parry-series, or capzug.
Your inventions aren't going away; and, nobody is out to diminish either of them.

I've had some recent emails which revealed some shocking new truths -- things that might profoundly impact the definition of orthodox/fairy problems... things that might redefine duals.
Some very old disagreements might just finally get resolved (in a win-win scenario).

I'd appreciate your participation in the discussion -- but, that means putting aside the hostilities, and understanding that you are not the target of views which may run contrary to your own.

You have to give people permission to disagree, and understand it is not personal.
Otherwise, the only place people can confidently discuss this matter, would be behind your back.

I am convinced we can reforge a logical, universal classification system, with clear definitions of our fundamental terms; but, this will require considerable participation (and commitment).
All I'm asking you is to let the discussion proceed.

OK, assuming we will get Dan's permission to have this discussion...
Let's start over by attempting to define three very basic terms: Orthodox, Fairy, and Fairy-Condition.

sincere apologies for the wordiness of this post, but it contains mainly questions -- no strong opinions.

What is the purpose of defining these terms?
Perhaps, we might achieve some very desirable goals:
1) we would like to allow people (especially newcomers) to read the Codex, and understand the terms it uses.
2) we hope to obtain a standard, universal problem notation (whereby the task of a wide variety of chess problems is immediately clear to a wide audience), and
3) we hope to obtain a logical classification system for at least some *fundamental* chess rule options,

The last point is a major goal, which implies a number of features:
a) it should preserve invention data (inventor, date, precise definition, etc).
b) it should operate with minimal redundancy (which is why it should address only fundamental elements),
c) it should provide some scheme to evaluate invention precedent, and provide for alternative interpretations,
d) it should serve as a filter in database searches,
e) it should provide as a logical basis to (sub-)divide problems, for any subscribing journal, TT, etc.

If this doesn't already seem a tall order, wait until you try feeding in some seemingly obvious terms!

It is actually remarkably difficult to define any of three terms I've asked for.
Does anybody have a good definition?

The general definition of Fairy Chess has been anything which employs a non-Orthodox Element.
However, with no easy definition for an "Orthodox Element," it's difficult to define a non-Orthodox Element.

In fact, the term, "Fairy Condition" is surprisingly non-trivial.
The general definition has been to define a Fairy Condition as anything which alters (or constrains) any of the Rules of Orthodox Chess (presumably, where the Rules of Orthodox Chess are given by the FIDE Rulebook).
However, we must acknowledge that the rulebook is evolving (even recent minor changes to this rulebook, such as Dead Reckoning, may have a substantial impact upon this definition).

Thus, Orthodox winds up defined by a moving target -- and we never obtain the logical basis we desired from it.

There are some options to consider:

1) We might define Orthodox Chess Rules according to the most recent FIDE Chess rulebook.

This is highly unappealing, because our most fundamental term (Orthodox) would be an inconsistent definition.
This fails to provide a logical basis by which we can establish a universal, problem notation.
Furthermore, as rules evolve, we might be required to alter the classification of a substantial number of problems (which were previously classified as Orthodox).

2) We might develop our own definition for Orthodox (which can not be altered by an evolving rulebook).

Essentially, this is equivalent to freezing progress on today's (or some prior) FIDE Chess Rulebook.
As the rules of the actual game evolve, orthodox problem chess would remain fixed.

The benefit of this is obvious: problemists (rather than chess players) would control the destiny of Orthodoxy.

As we will soon see, the present rulebook may have some surprising impact upon even some highly evolved chess problem theory -- especially regarding duals in selfmate problems!
It might actually be necessary for problemists to define their own orthodoxy, as a basis to provide alternative stipulations (where duals in selfmates are resolved according to universal stipulation, rather than a disputed theory!).

I realize it is somewhat unappealing to consider that the "Orthodox Game" may evolve away from a direct correspondence with "Orthodox Problems."
Nevertheless, unless problemists are given some input in drafting the FIDE rulebook, this might be an inherent necessity.

3) We might consider Orthodox to be defined by the collection of ALL FIDE Rulebooks (for any given problem date).

This is an interesting possibility.
The drawback is obvious -- Orthodox itself becomes an evolving set, rather than a clearly defined area.
Thus, our basis may require an elaborate evolution of rules; and, we still lose a direct correspondence with the present FIDE rulebook.

Furthermore, many rulebooks had ambiguous rules -- which have led to a variety of created fairy conditions; and, in some cases, orthodox problems may already have been updated, for evolved rulebooks.
It may be unappealing to consider some rules, which were not intended to be ambiguous, as orthodox.
Plus, it's unappealing to be tasked with a rulebook search, based upon the construction date (which might vary from even the publication date) of each problem.

4) We might abandon entirely the effort to define Fairy Conditions, Fairy Chess, and Orthodox...

If we are unable to clearly define Orthodox/Fairy terms, it would seem a strong suggestion that we are not intelligently selecting a valid (sub-)division for chess problems.
Perhaps all sets of chess rules should be considered as just some variant.

If you want to filter problems according to the present rulebook, you simply query according to that variant.
If you want to filter all problems which were ever considered orthodox, you would want to query according to the orthodox variant family (whether ambiguous rules are included in this family, is something that would have to be resolved).

If you treat all rulebooks as variants (and families of variants), how do you obtain a logical classification?
We might consider dividing problems according to the fundamental nature of problems.
For example, divide problems according to the motive of the virtual opponent: Opposing-play versus Help-play.
We could always sub-divide according to the variant rulebooks, and we might further sub-divide according to the aim (#, =, x, +, etc).
Something similar to this is employed by feenschach.

5) We might have some options I haven't considered...

If anybody has alternative suggestions, let's hear them.
And, if anybody strongly favors a particular option, please share which one, and why.

All I'm asking for is 3 definitions: Orthodox? Fairy Chess? Fairy Condition?
How should the Codex define these fundamental terms (terms it already uses, without providing a definition)?

I am composing fairy problems for more than 20 years and personally I had never felt anything like a pressing need to define fairy problems using any of ways like stated above. Some broad definition with perhaps large grey zones is enough, possibly even understood differently by different people. Why bother? I am suspecting there are others with similar feelings, and some people with opposite views have already expressed them in this thread.
Let's have a poll on question: "Is it necessary to define precisely what is Fairy Chess?"

"Is it necessary to define precisely what is Fairy Chess?"

Indeed this is the right question, it makes me think to the Bourbaki team in math... If I correctly understand, Kevin would like to set up fairy chess, in the same way a scientific theory is based on a few axioms, the rest being constructed via logical deductions.

Indeed, it was possible to deal with infinity before Cantor's works. But it seems difficult to deny that having at hand well-established and clarified concepts is better, and also easier for newcomers.

To get a chance to ascend such a high mountain as the axiomatization of fairy chess, the "elementary particles" should in no way be the Fide rules, I guess. Some of those rules probably already have to be considered “fairy” (e.g. castling possibility, double step from a pawn, and so on), while some fairy elements look like “orthodox”.

So I would say axioms should deal with very elementary features such as what is a piece, a move, a capture (annihilation being a special case), a chessboard (with their admissible topologies), an aim, a stipulation, etc. And then constructing step after step the whole edifice, some of those steps looking “orthodox” and the remaining looking “fairy”, but without a close frontier between them.

@Nicolas,

That's exactly how I see it.

@Juraj,

I've listed some possible benefits.
If you think only of the work you might save programmers (such as the popeye team), or think only of the benefits to newcomers, or think only of the redundant database searches the future might avoid, I think it's well worth the effort.

What's the harm?