MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

12:33 UTC
ISC 2024
 
  Forum*
 
 
 
 

Username:

Password:

Remember me

 
Forgot your
password?
Click here!
SIGN IN
to create your account if you don't already have one.
CHESS
SOLVING

Tournaments
Rating lists
1-Jan-2024

B P C F





 
 
MatPlus.Net Forum General Reciprocal Stipulations
 
You can only view this page!
Page: [Previous] [Next] 1 2
(1) Posted by Kevin Begley [Wednesday, Mar 8, 2023 02:16]

Reciprocal Stipulations


I just read an article by Shankar Ram titled "Reciprocal Stipulations."

I don't recommend reading it, but if you want to print it out and start a campfire, here's a link: https://juliasfairies.com/wp-content/uploads/Reciprocal-Stipulations.pdf

These are not stipulations, and there's nothing reciprocal about them.
At this point, more problemists should really know the difference between a GOAL and a STIPULATION.
Even more importantly, they should appreciate the value of formalized standards for AIMs, GOALs, STIPULATIONS, and FAIRY Conditions.

Further, they should appreciate that it is a CATEGORICAL MISTAKE to propose a category called "FAIRIES" in the FIDE Album, when in fact, this category is neither exclusive to FAIRIES (it contains many ORTHODOX problems, including stalemates) nor does it encompass FAIRIES (Fairy PGs already exist in another category, and Fairy hs#n problems are proposed to be removed).

 QUOTE 

A category mistake, or category error, or categorical mistake, or mistake of category, is a semantic or ontological error in which things belonging to a particular category are presented as if they belong to a different category, or, alternatively, a property is ascribed to a thing that could not possibly have that property.
-- Wikipedia.


The proper term for this category (what you call "FAIRIES") is either "MISCELLANEOUS", or "POTPOURRI".
If you're going to propose a re-catogorization of the FIDE Album, at least get the names right.
I suspect you all know better, but nobody wants to admit it -- nobody wants to offend fairy enthusiasts by admitting there is little (or no) respect for the "FAIRIES" category.
I appreciate that trepidation, but it's better to be honest.

Back to the reciprocal stipulations...
In this article, Shankar Ram is clearly struggling to determine what is a GOAL and what is a STIPULATION.
This is what happens when you are unmoored by definitional integrity (if you can define the word GOAL as a fundamental element in a chess problem stipulation, you don't float off and get caught between a GOAL and a STIPULATION -- take heed Circe's warning to Odysseus).

The problem with Shankar Ram's methodology is simple: it lacks an appreciation of formalized terms/nomenclature, and the meaning of his terms are not anchored to definitions.
Every time he encounters a COMBINATORIAL GOAL he can't process, he wants to name a new stipulation.
This is self-harm, and it's tragic.

It's high time we put away the silly mythology about "equity of popularity" in album categories, and return to providing a more reliable (FORMALIZED) problem framework which does not burden every newcomer to learn a million new names (which describe nothing more than simple combinations of things already known).

What's the point? Has fairy chess run out of things to invent? Need we resort to inventing combinations of things already known?
This nonsense has gone on too long. It's time to call bullshit on this recklessness.

Check your moorings, please... don't get tangled in your own knots.
What is a formal chess problem STIPULATION? What are the ELEMENTS of a formal chess problem STIPULATION?

You have a PLAY-TYPE (help play or adversarial play), you have an GOAL (I'll come to that), and you have a move limitation (2 moves, 3 moves, infinite moves).

What is a GOAL?

1) GOAL can be a simple AIM describing features of a position (checkmate, stalemate, check, double-check, etc).
2) GOAL can be a simple AIM describing features of a specific move (capture, castle, ep, promote, etc).
3) GOAL can be a COMBINATORIAL of several AIMs (++ and #, # or =, # by ep, etc).
4) GOAL can be a COLLECTION of GOALs (e.g., help-Valladao-in-7).
5) GOAL can be a sub-stipulation (the best way to formalize this, without inventing tons of useless names, is to use what you already know and express this with RECURSIVE goals).

Yes, Johnny, that's correct. The immediate GOAL of a helpselfmate problem is to achieve a s#1.
It's equally possible to make s#2 the AIM of a HELP-PLAY stipulation (and this has been done).

note: a composer may further want to inform the solver that the solution of the SUB-GOAL (s#2) is not required to be solved itself (thereby avoiding any duals in that solution).

When you are anchored by definitions, you can't help but realize stipulations can be GOALs (which have sub-goals).
That's when you begin to appreciate how to create a formalized nomenclature that covers all kind of problem stipulations (including many not yet seen).
You'll also discover that reciprocal-reflex problems are really just self-helpmates (and, unlike reflexmates, they conceal no fairy condition) -- who knew?

This is not a place where you want to name every bloody combination of AIMs and GOALs imaginable (you are mapping sea monsters, because you are unmoored by definitional integrity on fundamental terminology -- if you're not going to adhere to definitions, stop bandying meaningless terms).


It should be noted that Shankar Ram admits he instinctively knew this was better viewed as a combinatorial GOAL (because that is exactly what it is).
And by some extraordinary bad logic, he later concludes that GOALs can't be combinatorial, because some combinations do not mix.


I mean no disrespect to Shankar Ram (in fact, I regard him in high esteem, as a composer, as a courageous voice calling for improvement in problem chess, and as a person who has thought deeply about some of these issues), but this is a tragic conclusion.

We already knew combinations of GOALs do not always mix. You'd have known that if you ever considered some SIMPLE GOALS do not mix.
For example, you can not ask for both check and stalemate.
From that, we do not conclude that check and stalemate can not be GOALs (or AIMs).
We do not then make a new stipulation every time we encounter these sea monsters.

Learn your fundamental knots, Captain, if you want to master the sea in which a gnat may drink and an elephant may bathe.

Formalization is what keeps us near the shore, and away from the Straits of Messina (where we're unmoored, and mapping imaginary sea monsters).
There is no new PLAY-TYPE in these silly "reciprocal stipulations" concocted.
There is no new GOAL.
There is NOTHING new here. It's not a new stipulation.
The only thing new is that Shankar Ram newly learned he doesn't understand the fundamentals (which is why he admits -- and I credit him for this honesty -- he is struggling to differentiate a GOAL from a STIPULATION).

I've been there, myself. It was a struggle for me to find a formalized methodology by which chess problems could be classified.
Breaking complex ideas into logical sub-classes is what helps us to understand.
It worked wonders for me -- try it some time.

I sincerely hope that helps somebody understand the fundamental foundation of this artform better.

I hope the FIDE Album will take note, because a child can see their categorization is dishonest (and the recent proposal to remedy this does less than nothing to address the elephant on the board).
If your titles mean what they should to you, if your Album means what it should to you, if this artform means anything at all to you, then I implore you to learn the fundamentals (so that you are never again caught as one of the six young apprentices who drifted between the Scylla and Charybdis).

I say none of this to be disrespectful.
I have enormous respect (admiration is a better word) for the problemists who propose changes to the FIDE Album categories (I happen to agree with them on many facets of their proposal, though it is flawed to group problems in a fashion which can only be viewed as self-serving -- problems should be grouped logically: don't exclude orthodox hs=n in a group including both orthodox- and fairy- hs#n problems).

And, I have enormous respect for Julia's leadership -- she insisted upon a Classification Project, when so many dissuaded her from taking the first step on a journey that may be endless (and, yes, some good benefits came of that effort -- a credit to her and her team).

But, we have a classification project that failed to provide a system by which problemists (especially newcomers and inventors) can determine what is an AIM, what is a GOAL, what is a STIPULATION, and what is a FAIRY ELEMENT.
That's the whole point of CLASSIFICATION (it's not a list of groupings for existing elements, it's a methodology by which grouping can be achieved even for elements not yet invented).

Similarly, we have categories in our Album which are categorically false.
And, we are seeing inventions which fail to respect the value of FORMALIZATION.
This is not freedom (I fully appreciate the concept of freedom), this is chaos (these errors are far worse than the error of Descriptive Chess Notation).
Reckless disorder is not beneficial to this artform; it is harmful.

These failures are all linked to a lack of appreciation for the fundamental terms (and the substantial value of formalizing a universal problem nomenclature).
Chess problemists are smart people, so it's a major displeasure to me to endure constant failures to express our collective intelligence.
I hope this verbose post will encourage some of you to think before inventing more chaos.

But, I know many of you are adamant about never reading more than 50 words (maybe we're not all so smart).
 
(Read Only)pid=24510
(2) Posted by seetharaman kalyan [Wednesday, Mar 8, 2023 19:26]

50 lines i mentioned not words or goals
 
(Read Only)pid=24515
(3) Posted by Juraj Lörinc [Wednesday, Mar 8, 2023 19:36]

For me Shankar Ram's article is ok.

I'd say that newcomers are more repelled by remoteness of fairy elements from the game (I remember how mere nightriders were difficult for me to grasp in my first steps in composition), sheer amount of fairy elements and obviously exorbitant formalism (why kids are not taught natural numbers using Peano axioms?).
 
 
(Read Only)pid=24516
(4) Posted by Vlaicu Crisan [Thursday, Mar 9, 2023 01:06]

As the editor of the fairies column of Problem Paradise, I received one original composition which doesn't fit well in Shankar Ram's article description.
The most accurate formal description would actually be (7.5h[ws] --> [s#1 & hs#1.5]) as the author himself acknowledges instead of the misleading term "reciprocal hs#9".

I would therefore prefer a more generic approach for handling such nested combinations of play by explicitly specifying the plies when the shift of play [type] occurs.
This specification might be more precise than the "reciprocal" current naming.
 
   
(Read Only)pid=24520
(5) Posted by shankar ram [Thursday, Mar 9, 2023 07:29]

 QUOTE 
These are not stipulations, and there's nothing reciprocal about them.
I agree that the "Reciprocal" prefix is misleading. I just extended the use of the existing terms "Reciprocal Helpmate" and "Reciprocal Selfmate", also known as "Grazer Helpmate" and "Grazer Selfmate". However, these ARE stipulations. They contain a play type - opposition (adversarial): for Reci-s# and Reci-semi-reflex and help: for Reci-h# and Reci-hs#. They contain two clear cut goals - #1 and h#1 by black (Reci-h#), #1 and s#1 by white (Reci-s# and Reci-hs#), and #1 and semi-r#1 (Reci-semi-r#). And obviously, they contain a move length. Note that the total move length between the two or more goals may vary, depending on the nature of the final goals. We have addressed this issue in the generalised notation explained below.

 QUOTE 
Even more importantly, they should appreciate the value of formalized standards for AIMs, GOALs, STIPULATIONS, and FAIRY Conditions.
We used the term "Goals" in the project as it was used in the BCPS Glossary, which was our base document. WinChloe uses the term "Aims". Both these are interchangeable.
We define stipulation as Play-type + Goal + Length. Here again, Popeye already uses a similar definition in its help file.

 QUOTE 
Every time he encounters a COMBINATORIAL GOAL he can't process, he wants to name a new stipulation.
I'm not clear how you arrive at that conclusion. However, a follow-up article is in the works. The term "Reciprocal Stipulations" will be dropped in favour of "Multiple Stipulations". In a multiple stipulation problem, a position has to be reached from which one side can execute 1, 2, 3... goals together. I emphasise the term "together" because there is a related type called "Quodlibet", in which the multiple goals can be executed individually, in separate variations. They need not be possible from the same position, as in the original case.
The way to differentiate these two types of multiple stipulations was first suggested by Thomas Maeder here: https://juliasfairies.com/fccp-march2022-update/ and more specifically in the diagram here: https://juliasfairies.com/wp-content/uploads/classification-stipulation-outline-revised-colour-coded.png
Thomas suggested the use of the boolean algebra terms AND, OR and NOT. The "regular" multiple stipulations we have been discussing fall into the AND category: ALL their goals are mandatory and should be executable from the same position. The "Quodlibet" type falls into the OR category: Each of their goals is optional and can be executed in different variations of the problem. Twins of the type a)#n b)=n or a)s#n b)r#n (Torsten has many problems with this type of twinning!) are different from Quodlibets. The third NOT category holds types like "White to play and NOT mate"!

 QUOTE 
What is a GOAL?

1) GOAL can be a simple AIM describing features of a position (checkmate, stalemate, check, double-check, etc).
2) GOAL can be a simple AIM describing features of a specific move (capture, castle, ep, promote, etc).
3) GOAL can be a COMBINATORIAL of several AIMs (++ and #, # or =, # by ep, etc).
4) GOAL can be a COLLECTION of GOALs (e.g., help-Valladao-in-7).
5) GOAL can be a sub-stipulation (the best way to formalize this, without inventing tons of useless names, is to use what you already know and express this with RECURSIVE goals).
Agreed. "Sub-stipulations" have also been called "Nested Goals", earlier on this forum by yourself and others.

 QUOTE 
This is not a place where you want to name every bloody combination of AIMs and GOALs imaginable...
I have no such ambition. In fact, I have discussed with Torsten Linss and Vlaicu Crisan and we have come up with a generalised notation to describe multi-stipulations. Vlaicu has already given a sneak peek in his post. Torsten introduced it first. A final version will look like: d|hm.nw|b -> w|b(A & B & ...) where d = opposition play, h = help play, m.n = move length to the position where the multiple goals can be executed, the first w|b = who makes the first move, the second w|b = who executes the multiple goals, A, B, C... = the different goals. This assumes all the goals are to be executed by the SAME side (W or B). But you can also have a stipulation where each goal is executed by a different side. This involves an illegal "out of turn" move by one side. For this case, the notation can be modified as: d|hm.nw|b -> w|b(A) & w|b(B) & ... A series mover version is also possible, for which the prefix "Ser-" can be used: Ser-mw|b -> w|b(A & B & ...). Here, m is the move length and either W or B can play the series and also execute the final goals. For Quodlibets, the notation can be slightly modified as: d|hm.nw|b -> w|b(A | B | ...)

 QUOTE 
...And by some extraordinary bad logic, he later concludes that GOALs can't be combinatorial, because some combinations do not mix.
Again, I'm not clear how and where I concluded as such.

 QUOTE 
For example, you can not ask for both check and stalemate.
Really? "Check" this: WKc2, WRe2, BKa2; h0.5b -> w(+1 & =1). That is, B makes a move, to reach a position, where W can both +1 and =1. 1.Ka2-a1 and 1...Re1+ & 1...Kc1=.
Or even the famous C.Tomlinson, Amusements in Chess, 1845: WKc6, WPc7, BKa7; #2 1.c8=Q? 1.c8=R! After moving the BK to a6, this can also be set as h0.5b -> w(+1 & =1 & #2). 1.Ka6-a7 and 1...c8=S+ & 1...c8=Q= & 1...c8=R 2.Ka6 Ra8#.
Perhaps you got confused by thinking of both check and stalemate by the SAME move, which is, obviously, impossible.
I was referring to cases like h#1 and s#1 by the same side in the same position. Obviously, the s#1 solution will also work for h#1.

 QUOTE 
We do not then make a new stipulation every time we encounter these sea monsters.
Well, we can, by using the above notation!

 QUOTE 
Learn your fundamental knots, Captain, if you want to master the sea in which a gnat may drink and an elephant may bathe.
Naughty, Knotty, Gnatty!

 QUOTE 
The only thing new is that Shankar Ram newly learned he doesn't understand the fundamentals (which is why he admits -- and I credit him for this honesty -- he is struggling to differentiate a GOAL from a STIPULATION).
Illuminance, Cognizance and Concordance dawn when you admit your ignorance and give up your arrogance and obstinance!

 QUOTE 
And, I have enormous respect for Julia's leadership -- she insisted upon a Classification Project, when so many dissuaded her from taking the first step on a journey that may be endless (and, yes, some good benefits came of that effort -- a credit to her and her team).
Thank you!

 QUOTE 
But, I know many of you are adamant about never reading more than 50 words (maybe we're not all so smart).
We try, We try!
 
   
(Read Only)pid=24521
(6) Posted by shankar ram [Thursday, Mar 9, 2023 08:32]

Of course, all these stipulation types will be only theoretical constructs without some interesting problems!
Here are a few:


Niels Hoeg, 1st Prize
Skakbladet 1907
(= 7+7 )
#2/s#2/=2/s=2 Quodlibet!
d1.0w->w(#1 | =1 | s#1 | s=1)
1.Qe1!
1...exf1=Q 2.Kxg3 Qxe1#
1...exf1=R 2.Qxg3#
1...exf1=B 2.Kxg3=
1...exf1=S 2.Qxf2+ gxf2=
1...g2 2.Bxe2#
AUW

Torsten Linß, 7th Prize
Strategems, 2020
(= 2+4 )
Reci-h#8
h7.0b->b(#1 & h#1)
1.Ra2 Kg1 2.Qb2 Kf1 3.Kc2 Ke2 4.Kc1+ Kd3 5.Qe2+ Kc3 6.Rd2 Kb3 7.Sc2 Ka2 8.Qc4# & 8.Qd1 Bb2#
Doubled, reciprocal and reversed Bristol, doubled Indian with two distinct line closings on the critical square c2, doubled switchbacks and a long march by the WK.

Christopher Jones, 4th Honourable Mention
Strategems, 2020
(= 6+10 )
Reci-h#3.5 2 Solutions
h2.5w->b(#1 & h#1) 2 Solutions
1…Be5 2.cxb5+ Kxd6 3.Rc7 Sc3 4.Bd2 Bg3#, 4.Bxe5#
1…Rc4 2.d5+ Kxc6 3.Bc7 Sg4 4.Re2 Rc1#, 4.Rxc4#
ODT in a reciprocal helpmate, with line closings, batteries, moves to the same square, line openings and self blocks.
 
 
(Read Only)pid=24522
(7) Posted by Kevin Begley [Thursday, Mar 9, 2023 10:48]

@Vlaicu,

I completely agree: the move issue posses a difficulty, particularly at the handoff.
It can be very confusing to make this work under the present standard (using moves rather than ply).

I further suspect this may be due to the fact that our current system holds no distinct meaning for a helpmate in fewer than 2 ply, a selfmate in fewer than 2 ply, or a help-selfmate in fewer than 3 ply -- that is, the difficulty may not only be the moves (versus ply) issue.

Perhaps we might consider a standard where the helpmate in 1 ply means black makes one helping move (with the goal of producing a subgoal: a # in 1 ply, for White).
Similarly, the selfmate in 1 ply would mean white plays 1 ply to reach the subgoal (what Petko calls a "compelled mate" in 1 ply, or c#1 ply, for Black).
The helpselfmate in 1 ply would mean black plays one helping ply to reach the subgoal (the aforementioned selfmate in 1 ply, which then reduces to the subgoal: a compelled mate in 1 ply for black).

The point here is, if you see a helpmate in 3 ply, under such a standard, you know immediately there are 3 ply to be spent in help-play, and that black starts (given the odd number) with a move helping to achieve the goal, white plays a move advancing the goal, and black plays the final move helping to achieve the goal.
This seems counter-intuitive at first, because we tend to associate the goal, even in a help-goal problem, as something white is tasked to achieve (because we view the aim as the goal, but the actual goal is to achieve a checkmate in 1 ply for white). It is actually Black, not White, who always plays the final helping move, which achieves the help-portion of the problem!
I suspect this may be a major source of our confusion, so perhaps the best remedy is to use ply, and also establish a stipulation standard that gives meaning to every play-type such that they may be executed in a single ply.

I have actually been thinking about this potential remedy since last we spoke about some of these issues, but I will admit this is not an easy change to propose.
This is no simple tweak to the past standard, and this would require an adjustment period. So, knowing the upheaval it's likely to cause, I was reluctant to suggest this as a remedy.

I remain hopeful a better solution will suggest itself (which is why I appeal to the many smart people in the problem community to find an easier way to resolve this issue).
I haven't yet found it, but that doesn't mean it's not possible -- I may well be overlooking an elegant solution.

Whatever the case, I remain determined to fight for this methodology, because I am certain this is a better path forward (naming more stipulations for every combination is, by far, the most inelegant option imaginable). We need to recognize goals may have sub-goals, and create a formal stipulation standard that covers a wider range of problem stipulations.



@Juraj,

Frankly, I'd encourage young math students to read "The Little Red Hen."

I can not speak to the merit of teaching highly abstract math to young children. I have my reservations, but I'm willing to test this theory.
I know there are teachers who strongly believe that this would benefit children, and there are those who insist it is folly.
Somebody is right, and somebody is wrong. I would be interested to see this tested -- not to find out who is right or wrong, but to learn why they are right or wrong.

I lean toward the practical math camp.
I suspect teachers who motivate good questions, and demonstrate how math is a tool to answer them, will tend to motivate students better (and I strongly believe motivation is the primary roadblock, not only for kids).

I also appreciate your analogy to problem chess, using abstract math.
I had no apprehension to math. I had the benefit of some good teachers.
I did have an extreme apprehension to problem chess composition. It was too abstract, and totally impractical (I knew only one goal in chess: to improve as a player).
I wouldn't even look at a diagram with an upside-down chess piece in the diagram. If you showed me a problem, I'd insist it had better not involve castling or en passant capture. No silly tricks! LOL.
I didn't overcome this disdain until I realized what a huge advantage problem enthusiasts had over me (when it came to problem solving, I could not fast enough unlearn everything I had known about chess).
What's the point achieving an opening advantage, if I can't solve problems in unfamiliar positions? It seemed the more I won, the less I understood. That's not what I wanted from chess.

So, I decided to do something about it.
Once I realized studying helpmates had a practical value -- to improve problem solving skills -- it opened the door to consider more unusual problems.
By the time I began considering fairy problems, I was already realizing my goal in chess need not be to win chess games.

I suspect many problemists can relate to this experience, and most wish we'd learned this lesson sooner.
I look back and see all the ways problem chess failed to connect, and I am inclined to blame those who knew what I did not know.
That's why I am passionate about improving the ways problem chess presents itself to chess players (there are plenty more who never considered problem chess, and we are a lesser community for not reaching more of them).

I can promise you absolute rookies are doing a better job of motivating problem enthusiasts than all our showcased brilliancies will ever achieve.
There are theories about why this doesn't work -- most are related to studies which show that a weak chess player is likely to have a stronger offspring than a strong player.
The hope of real victory is a strong motivator; second only, I suspect, to achieving it.
Better to show new enthusiasts those awarded problems you feel won undeserved acclaim, and tell them what you failed to achieve (that's a real motivator).

Disorganized chaos is not a great motivator, in any field of study.

There was a terrible period in history, when all appreciation for the value of mathematics was lost.
I sometimes wonder if that will be the fate of chess problems (when I'm not wondering how AI might erode a motivation for humans to solve, compose, and enjoy chess problems).
But, the immediate problem is that we (human problemists) are eroding our own artform faster than any Artificial Intelligence could dream possible.

I have yet to hear anyone admit it's a bad idea to rip fairy problems out of a category called "FAIRIES" (which contains problems which can not be considered fairies).
Should we change the category name (from "FAIRIES" to "MISCELLANEOUS"), or will we, at long last, turn to a logical categorization methodology?
Nobody will say boo. Everybody just wants to look away.

The message this sends is undeniable: when you look away, you teach newcomers to do the same; just look away.

I suspect everybody already agrees on the solution here (problems should be first classified based upon the PLAY TYPE, then the GOAL, then the AIM, then the number of moves, and finally, when necessary, based upon orthodox/fairy elements).

The helpmate section should be a HELP-GOAL section (containing h=n, h+n, hxn, etc). There is no doubt that's the best grouping for those problems.
The direct sections should be a DIRECT-GOAL section (containing =n, +n, xn, etc).

If you want to sub-divide those sections by the elements (orthodox vs fairy, or orthodox + constraining fairy conditions vs fairy condition, or whatever), go to town.

What you do not do is sub-classify based upon a particular combination of stipulation, goal, and aim (hs#), such that orthodox hs#3 is grouped with a Mars Circe hs#3 with kangaroos on a cylinder board, but the orthodox hs=3 is excluded.

There is one logical alternative: first sub-divide based upon fairy versus orthodox elements, then do the you proceed as previously prescribed (starting with PLAY TYPE, then GOAL, then AIM, etc).
I don't deny, there's an argument to be made for this approach, if you believe orthodox chess merits a special status.
And, guess what? I'd be very happy with whatever choice the problem community makes.
I'm not here to advocate on either side of those matters, I'm just here to seek a formal standard which logically classifies and conveys chess problems (in a logical, accessible manner).

That is the ever loving point of formalization and logical classification: to simplify complex endeavors, to break complex problems into simple sub-classes, to facilitate comprehension.
Formal structure simplifies chess terminology (and eliminates needless terms invented in what can only be considered a rite of territorial making).
I'm not pushing excessively rigid formalism, for the sake of formalism. That's a ridiculous claim.

What are presenting to the world by our categorical failures, our classification failures, and our definitional failures?
What are we saying about ourselves, when we flaunt proposals which do nothing to address any of these gross failures?
We appear to be a colony of grossly incompetent artists who can't even get our category names right, yet we're afraid avoiding any demonstrable, categorical dishonesty might impose an extreme formal rigidity.
This may come as a shock (brace yourself): math, like chess itself, is based upon formal rules (and both derive their value from those formal rules).

If you are concerned about rigidity, let me assure you: nothing could be more rigid than needlessly imposing a hundred new stipulation names upon every combinatorial pairing of iterative goals imaginable.
If that chaos is the future path of problem chess, we'd be better off memorizing Pi.

Would you be motivated to join a group of mathematicians who refuse to group Pi in the group of real numbers?
Maybe, if you're bent on imposing more chaos.

I do agree that the collective penchant (in problem chess) for needlessly extraneous terminology (read: chaos) is hardly something newcomers will consciously abhor; few will ever expect that so many smart people could be so wrong about their own field. Nonetheless, subconsciously, they will abhor the chaos (which nobody can pretend is perceived as anything less tangible than prison bars).


@seetharaman kalyan

The article I referenced has 68 lines. Call the cops.
 
   
(Read Only)pid=24524
(8) Posted by Kevin Begley [Thursday, Mar 9, 2023 10:49]

@Shankar Ram,

GOALS and AIMs are not interchangeable terms.

Frodo's ultimate AIM is to reach Mount Doom.
Frodo's immediate GOAL may be to reach Rivendale.

Your GOAL may be simple (read: equal to your final AIM), or it may by the first step on a journey with many SUB-GOALs (necessary to reach your final AIM).

Tell you what, let's not use the word AIM, since that may be throwing you off. Instead, let's use the word TARGET.

You may have a simple GOAL: hit the target (checkmate, capture some enemy unit, give check, etc).

You may have a GOAL with a combination of two (or more) TARGETS: double check AND Mate, checkmate OR stalemate, Vallado-in-7, etc.

You may have a GOAL which is a stipulation: the helpselfmate is a problem which requires HELP-PLAY, to reach the first GOAL (achieve a selfmate in 1 move, for white), then requires 1 ply of DIRECT-PLAY to reach the sub-GOAL (achieve a compelled-mate in 1 ply), and this ultimately reaches the TARGET (the white King is checkmated).

You may have a GOAL which is a combination of two (or more) stipulations.

Not all targets are consistent. You can not have a goal calling for "check and stalemate."
note: this does not alter the fact that check and stalemate are both TARGETS (the ultimate success in solving the problem).
Similarly, if the combination of two stipulations, in the GOAL, can not function as a viable GOAL, it doesn't change the fact that the stipulations are stipulated as GOALs.

You want to write a new article calling this multiple stipulations, because you're still having trouble identifying what is a GOAL and what is a STIPULATION.

Get this straight: There is one stipulation in each phase of a chess problem. Just one. That stipulation may be made of complex sub-GOALs, which contain stipulations, but there is ONE stipulation.
If there are two stipulations, it's a second phase, or a second chess problem. There is never more than one stipulation in a single phase of a chess problem. NEVER.

This is NOT multiple stipulations. Everyone knows this. I urge you to rethink your new article.


In a help-self-mate-n problem, the primary type of play is HELP-PLAY (it's primarily a scheduling problem), and the goal is to achieve a position where white has a selfmate in 1 move (read: in 2 ply).
The sub-GOAL is to solve the selfmate in 1 problem.
The TARGET is checkmate (#) of the white King.

You could argue, as Petko does, that the sub-GOAL of a selfmate in 1 problem is to achieve a "compelled mate in 1 ply" (a position where black has no legal moves except those which checkmate the white King (which is our ultimate TARGET), and that the primary PLAY-TYPE is just a DIRECT-PLAY problem, with a different GOAL.
I have no problem with Petko's interpretation.

But, what you do not do is claim that a combinatorial GOAL, made up of two stipulations (even if those involve more than one TARGET), conveys multiple stipulations in a single phase of one problem. That's false. Everybody already knows that's false.
Nor do you create new names, and declare this a new type of stipulation -- there is NOTHING new about it, and it is NOT a stipulation.
 
   
(Read Only)pid=24525
(9) Posted by Kevin Begley [Thursday, Mar 9, 2023 11:44]

@Shankar Ram,

We agree on this point:

 QUOTE 
We define stipulation as Play-type + Goal + Length.


Play-type = direct or help (I am not aware of any alternative).

GOAL can be a TARGET, a stipulation (s#1 is the GOAL of hs#n), an iterative stipulation containing more sub-GOALs, or a combination of TARGETS, STIPULATIONS, and ITERATIVE STIPULAIONS.

[edit: I also agree with the term "NESTED GOALS" in place of recursive or iterative GOALs]

But, there is ONE stipulation per phase of any problem. Just one. There are not multiple stipulations tasked for the solver (even when there are sub-stipulations contained within the GOAL).
There may be a combination of GOALs, there may be multiple TARGETS, but there is ONE stipulation.
 
   
(Read Only)pid=24526
(10) Posted by Kevin Begley [Thursday, Mar 9, 2023 11:59]

Here is why Thomas Maeder's Boolean logic proposal is insufficient:

This can formally describe a number of stipulations, but there are many instances (particularly sequential stipulations) where this will fail.

The purpose of formal stipulation is to reduce a world problem to a universally understood format.

What Thomas suggests will collapse the following word problems:
White gives double check or captures in 4.
White gives double check and does not capture in 5.
etc.

Boolean logic will fail when you stipulate a sequence of events:
White first castles, then captures en passant, and lastly promotes in 7.
White promotes Queen, then Rook, then Bishop, and lastly Knight in 5.
etc

A more universal formal stipulation would also allow for sequencing.

Furthermore, it must be possible to stipulate (in the nomenclature) when the GOAL containing a stipulation must be solved iteratively, versus when achieving the sub-GOAL is the end of the problem.
This is essential for composers who wish only to achieve the GOAL itself, and avoid iteration (e.g., to avoid duals in their solution), despite the fact that the GOAL is a stipulation.
 
   
(Read Only)pid=24527
(11) Posted by Kevin Begley [Thursday, Mar 9, 2023 12:26]

@Shankar Ram,

When I said, "For example, you can not ask for both check and stalemate," it is clear I am referring to combinatorial GOAL (e.g., help-check and stalemate-in 1 ply) where both occur simultaneously. You can NOT do both simultaneously.

You replied, "Really? "Check" this: WKc2, WRe2, BKa2; h0.5b -> w(+1 & =1)."

You can not execute both a check and stalemate simultaneously (as you can with ++ and #).
If that was not clearly conveyed to you, as my intent, one of us is to blame for the miscommunication, and I'll just accept responsibility for it.
I'm sorry I didn't convey this better.

However, What you are conveying here is an XOR (exclusive OR), not an AND.

White can + XOR = (one or the other, but not both) in one move.

That stipulation, with the XOR is the sub-GOAL of your intended stipulation. It's not an AND, such as h(++ and #)1.

What is the PLAY TYPE of your problem: Help.
What is the GOAL of your problem: Direct(+ XOR =)1 ply
How many ply of help play: 1 ply

Help ( Direct(+ XOR =) in 1 ply ) in 1 ply

This reduces to HD(+ XOR =)1 (note: using ply notation: see my response above to Vlaicu, also note that the 1 is the implied default when no length is provided).


ps: Thomas doesn't need AND, OR, and NOT -- you need only NAND (not AND) gates to achieve all three gates.

warning: chess problem stipulation may soon reduce to an iterative consensus problem. :-)
 
 
(Read Only)pid=24528
(12) Posted by shankar ram [Thursday, Mar 9, 2023 12:28]

 QUOTE 
Frodo's ultimate AIM is to reach Mount Doom.
Frodo's immediate GOAL may be to reach Rivendale.
Got it, Gandalf!

 QUOTE 
Play-type = direct or help (I am not aware of any alternative).
Well, there is the series play type. But you don't accept it! ;-)

 QUOTE 
GOAL can be a TARGET, a stipulation (s#1 is the GOAL of hs#n), an iterative stipulation containing more sub-GOALs, or a combination of TARGETS, STIPULATIONS, and ITERATIVE STIPULAIONS.
Agreed. You want to use the term GOAL like a basket, with aims, targets, nested-goals, multi-goals, etc. being the goodies inside!

 QUOTE 
This is NOT multiple stipulations.
...
I urge you to rethink your new article.
Maybe the term "Multi-Goal Stipulations" is better?

 QUOTE 
Here is why Thomas Maeder's Boolean logic proposal is insufficient:
Well, I suppose he meant it to be used for specific cases and not universally.
 
   
(Read Only)pid=24529
(13) Posted by Kevin Begley [Thursday, Mar 9, 2023 12:36]

Series play is not a PLAY TYPE, it is a FAIRY CONDITION.

Everybody knows this. You will know this too if you can provide a definition of the term "STIPULATION" versus "FAIRY CONDITION."
If you want to claim otherwise, then tons of fairy conditions will qualify as PLAY-TYPES in new stipulations.
I'll spare you the long list of games.

Is the Maharaja Variant a stipulation or a fairy condition?
Is Progressive chess a stipulation or a fairy condition?

If you really want to argue that point, I wish you'd have a go. :)
 
   
(Read Only)pid=24530
(14) Posted by Kevin Begley [Thursday, Mar 9, 2023 12:39]

 QUOTE 

Maybe the term "Multi-Goal Stipulations" is better?



We have reached consensus. I totally support that title for your article, and I'd love to read it. :)

Now, we are getting somewhere.
Maybe tomorrow, we can solve some global problems.
It'll be like the new adventures of Gandalf and Frodo.
 
 
(Read Only)pid=24531
(15) Posted by shankar ram [Thursday, Mar 9, 2023 12:44]

 QUOTE 
If you really want to argue that point, I wish you'd have a go. :)
God No! Not unless Dan Meinking's spirit occupied my body! ;-)
 
   
(Read Only)pid=24532
(16) Posted by Kevin Begley [Thursday, Mar 9, 2023 13:00]

I miss Dan. He was a great guy.
I don't miss our arguments, lol.

I believe he came to realize parry-series movers (and series movers) had to be a fairy condition -- he made a comment to me about it, which strongly implied he had changed his view.
Long story...

So, if he was here, I don't believe he'd have argued this point.
I do wonder if he'd have taken over StrateGems. I suspect so.


In all seriousness, thank you. I've enjoyed the conversation (and even the spirited banter).
It seems we agree on more than I had expected. I'm glad to see that.
And, I think we've both gained a better understanding from this conversation.

I need to run for a while...
But, do take note of the point I made to Vlaicu about a possible way to deal with the ply handoff issue.
I think it's an interesting idea, and who knows, maybe it will give you some ideas how to improve upon it.
 
   
(Read Only)pid=24533
(17) Posted by seetharaman kalyan [Thursday, Mar 9, 2023 14:17]

When you try to differentiate Aim and Goal you cleverly use the different adjectives 'ultimate' and "immediate". You don't seem to realise or refuse to accept that it is these adjectives which give them different meanings. Without the qualifying adjectives Aim and Goal mean the same thing
 
   
(Read Only)pid=24534
(18) Posted by Kevin Begley [Thursday, Mar 9, 2023 15:18]

No, Shankar, it's actually my clear definitions for these terms which differentiate them.
I used adjectives to help you understand their meaning.

These words do not have identical meanings, though I will admit many people may ascribe different meanings to each.

For example, see: https://www.projectsmart.co.uk/smart-goals/whats-the-difference-between-goals-aims-and-objectives.php

Under my definition, I agree with the article I referenced, that "AIM is a single statement of purpose," for example, but I do not agree that the aim "is a stepping stone towards your overall goal." Under my definition, for problem chess, GOALs are the stepping stone toward reaching your singular and ultimate AIM.

I have since decided that TARGET is a more clear term (it better represents the definition I established for AIM), as it better conveys a singular and tangible purpose (though you may have more than one TARGET in the GOAL, and each TARGET describes a singular and tangible purpose).

I developed unambiguous definitions for these fundamental terms (many years ago). That's why I never have trouble differentiating between a TARGET (formerly what I called AIM), a GOAL, a STIPULATION, and a FAIRY CONDITION.

Definitions are your anchor. You are free to define terms as you like, so long as you establish (for your audience) their meaning, and you remain consistent with your terms, I have no problem understanding the need to better define terms for chess problems.

So, no, don't get this tangled: my terms are not defined by adjectives. They are anchored in carefully considered definitions.
I encourage you to develop similar definitions for your terms, and provide them in your article.

And let me stipulate that we both share the same GOAL, AIM, and TARGET: we both agree that formal stipulations are needed which can convey a greater possibility of stipulations (including with NESTED GOALS).

And, I welcome input from anyone how this might best be achieved (here in this thread, in articles, anywhere -- please provide us a link here, if possible).
I am not here to impose my standards on anyone. I'm only here to argue the need for more intelligent formal standards of stipulation (and to avoid needless additional stipulations, fairy condition, etc).

I'm delighted to see that you, Vlaicu, and others (Thomas) are thinking about this issue. That's music to my ears.
If you want to correspond on your article, or any of these issues, I'd be glad to share my thoughts (just drop me a message me here, or by email).

The more we can simplify problem chess, the more we can logically classify and categorize it, the better... for EVERYONE.

I just want to be sure that whatever new standard, it has considered mechanisms to address unexpected complexities.

For example, if you have a combination of two goals (which are sub-stipulations), what do you do if a composer wants one goal iteratively played forward, but the other goal to be terminal (note: this is a BAD idea, of course, and the solution will be as ugly as the stipulation, but I'd like to see a stipulation capable of handling even some possibilities we may find aesthetically grotesque -- because you never know).

And, I'd be delighted to see you -- or anyone else -- find the best way to express a form for a better stipulation.
There's an art to that, and you never know who might think of the best way to convey these things.
I would invite everyone to contribute -- if you have a good (or better) idea, I'm delighted to consider it.
If you foresee a problem with one suggestion, let's hear it here. Don't hesitate. If we're going to consider improvements on this scale, we want to get this right the first time.

The best idea will cover the most possible stipulations, will be universal, will have no ambiguity, and will most elegantly express (in the fewest characters, on average) the stipulation. Furthermore, it will provide for the possibility to cover all existing stipulations (so that we might later transition entirely to this one standard), and it will will have flexibility to extend beyond what we might imagine possible on a chess board.

I'm not looking for any credit. I am not looking to impose my standards. I'm not here to tell anyone how this goes (I don't yet know -- nobody does).
I'm just here to contribute some ideas, to applaud the good ideas, and to criticize the bad ideas -- in the hopes it may be of benefit.
Problem chess nomenclature, and categorization, and definitions will inevitably go through a major upheaval (this can't be avoided).
That means databases, solving tools, old problem pages -- name it.

The questions are many...
Are we going to meet this challenge, or are we going to leave this mess for the next generation?
Are we going to preserve our history?
Are we going to get it right (thinking long term, so this upheaval doesn't happen as often going forward)?
Are we going to entrench ourselves in self-interest, and delay all forward progress?
Are we going to disagree in a civil manner (there will be disagreements -- nobody's going to get 100% what they want)?

It will be interesting to see how we meet these challenges.
I'm more optimistic than I've been in two decades of arguing for changes.
 
   
(Read Only)pid=24535
(19) Posted by shankar ram [Thursday, Mar 9, 2023 15:26]

Ahem, You should be addressing Seetharaman!
 
   
(Read Only)pid=24536
(20) Posted by Kevin Begley [Thursday, Mar 9, 2023 15:37]

My apologies, to you and to Seetharaman.

It's been a long day, and it's not yet 7am here. :)
 
   
(Read Only)pid=24537

Read more...
Page: [Previous] [Next] 1 2

MatPlus.Net Forum General Reciprocal Stipulations