Henrik Juel: Completing the key isn't legal, 1.'Rd1'/'-bPc5'/'e8S'+, but omitting the promotion in the mating move is, 1.Sxe7+ Bxe7 2.'f8'#. Excellent joke. (2003-09-19)
James Malcom: A very witty joke Valladao! (2020-09-24)
A.Buchanan: Really like this joke (2020-09-25)
Henrik Juel: To see the illegality of completing 1.fxe8=S+?, note the captures:
Black captured fxDe, so [Pg7,h7] were captured on their files, while the other six missing black men were captured by white pawns
The illegality of completing 1.0-0-0,bxc6ep? is rather obvious (2020-09-26)
James Malcom: And for those you don't find it obvious: If the White king hasn't moved, then where did wRa5, and wBb8 if bPb5 has just done a double-step, come from? Neither can be promoted pieces, as White still has all eight pawns. Trying to finish castling and en passant therefore both produce illegal positions and thus cannot be the solution. (2020-09-26)
Henrik Juel: Continuing beating the dead horse...
How does the white player actually perform an entire move?
1.0-0-0+: Ke1-c1 and Ta1-d1
1.bxc6ep+: Pb5-c6 and remove sPc5
1.fxe8=S+: remove sYe8 and Pf7-e8 and replace wPe8 with wSe8 (three fractional actions)
2.f8=Y#: Pf7-f8 and replace wPf8 with wYf8
So a marginally better stipulation might be: 'White to move mates in less than 2 moves' (2020-09-26)
A.Buchanan: Hurray I've got the animation working! I agree with Henrik's stipulation. There is a dummy pawn on f8, but not by the Dummy Pawn rule. Instead it's the joke that does it. If Dummy Rule applied, the move would be full length! :) (2022-02-09)
A.Buchanan: All three tries are full #1.5 not as the solution currently has it #0.5. Really fine joke! (2020-09-25, hidden)
Henrik Juel: The legal mate could be written better:
1. Sxe7+! Lxe7 2. f8=B# (Omitting the promotion) (2020-09-26, hidden)
A.Buchanan: I've tried to set up some intermediate diagrams to allow us to show the animation, but the pieces seem to move even for words which are not moves (and marked italic/red). It's quite baffling. It's almost as if someone with privilege has gone in the background and assigned moves to those dead words. If so, that's extremely unhelpful. Gerd was quite active on this site for a month in German lockdown, for which I'm grateful, but one key task is still waiting: to empower the editors by documenting exactly how the animation decides what to do based on the diagram(s) and stipulation. (2020-09-27, hidden)
A.Buchanan: The PDB stipulation field should be non-text form to drive the animation. Where the stipulation is non-standard, there should be a Unicode representation of the original, and translations into German, English and French. But cannot overwrite any original non-standard stipulation, in my opinion, as it is part of the composer's work. (2020-09-27, hidden)
A.Buchanan: I no longer think this is Dummy Pawn: instead it's a truncated regular promotion move. (2022-01-26, hidden)
A.Buchanan: Hurray I've got the animation "working"! I know this *looks* like a dummy pawn promotion, but if the famous Dummy Pawn rule applied here, then W2 would constitute a full move, not a fractional one! :) This is a truncated orthodox promotion instead! (2022-02-09, hidden)
A.Buchanan: Hurray I've got the animation "working"! I agree with Henrik that this is #2. We end up with a dummy pawn on f8, but not by the Dummy Pawn rule. Instead it's the joke that does it. If Dummy Rule applied, the move would not be considered to be truncated! (2022-02-09, hidden) more ...
comment

BUD20 AS-2
A.Buchanan: While it's true that there is no legal last move for Black, does this make the problem a "Retro"? The reasoning is hyper-trivial. Was this intended as a joke? Or is there some context for the composition? If only I knew a Danish person who might have some knowledge of the culture at the time... :) (2021-10-07)
Henrik Juel: Ha-ha, here goes
Knud Hannemann was the best danish composer in his prime, but this is surely a joke
Possibly the try 1.g8=D#?? was intended
Skakbladet 1972 is not available for me, but it might throw some light on the problem (2021-10-07)
A.Buchanan: Given it's a joke, I think it's ok to class it as Retro genre. This is related to the new Uplex theme, which I will soon upload. "=2" also works. And "h#2 (3 solutions)" must be well known. (2021-10-07, hidden) more ...
comment

James Malcom: Wrong source and author!

C. Planck merely got it from a friend and it was reprinted in The British Chess Magazine in his letter to the editor in the March 23, 1901 issue-https://www.google.com/books/edition/The_British_Chess_Magazine/mZlJAAAAYAAJ?hl

Quote: "This afternoon the following curiosity was submitted for my examination. I unfortunately omitted to learn where my friend got it from."

The 1910 books, is indeed in there, gives the wrong information0https://www.google.com/books/edition/150_Schachkuriosit%C3%A4ten/NVzzAAAAMAAJ?hl

Luckily, in a short afternote on May 7, Planck mentions that the solution has recently been given in The Pleasure Hour. It can be found here-https://hdl.handle.net/2027/mdp.39015068398471

I found the solution on page 544 in April, not surprising that it was mentioned in May due the the "slowness" of the day. It gives the solutions for the February problems.

Jumping back to page 362, it appears as an original problem by B. G. Laws.

Mystery resolved! (2021-01-02)
comment

Originalforderung: White mates in two - with the Knight at KKt6.
Motto: "A La Napier"
A.Buchanan: It's easy to make a miniature ideal mate version of this cheeky problem, e.g. 8/6pk/3N1PN1/6P1/7K/8/8/8 but the charm of Marshall's original is the apparently pointless wSb1 (2020-10-21)
comment

Mario Richter: Cook: Schwarz hat keinen legalen letzten Zug (2014-05-25)
A.Buchanan: This is not a retro problem. If we grant Black first move, then 0. ... Txb7! refutes. It's just a technical oversight. I can find no correction in PDB or WinChloe. Observe that sBb6 serves no purpose. If we remove it, then wLb5 serves little purpose. Removing both units, and shifting wD to e1,e5,e6,g6 or h5 retains soundness, but gives Black a last move. All variations persist, including the minor dual after 1. ... Lxb7, where both mates are separately forced. I prefer wDe6, because the key defending Sc8 is otherwise too strong. Also note the try 1. Dg8? In summary therefore: correction krN5/pN6/b2KQ3/8/8/8/R7/8 #2. (2021-10-09)
A.Buchanan: WinChloe gives a different original source, Wiener Schach Zeitung, with the same year 1932, towards the end of Keeble's chess problem career. (2021-10-09)
Mario Richter: My guess is, that the publication in 'Wiener Schach-Zeitung' is only a reprint as part of what I call "a real solving competition" (since amongst the problems to be solved there were also [intentionally] some with severals cooks or with seemingly or real illegal positions):
"Aus dem diesjährigen internationalen Lösungswettkampf.
Diesem Turnier ... lagen zum Teil sehr knifflige Probleme zugrunde. Es sei für unsere Löser vorausgeschickt, daß die Aufgaben zum Teil mehrfach nebenlösig sind, mitunter eine partieunmögliche Stellung vortäuschen oder auch haben. Die Illegalität der Stellung ist zu beweisen." (2021-10-09)
A.Buchanan: Wow, thanks Mario! This is a great example then of how one can't understand a problem sometimes, without knowing the context in which it's composed. How should we classify this one here in PDB? (2021-10-09)
A.Buchanan: Since illegality is the intention, it's not cooked. And it's a joke. bBb6 is a weasel. So much is clear, but I am uncertain whether to use the descriptive keyword "illegal position" or "no legal last move for black" or something else. The latter is actually a retro-stalemate (defined as a one-sided illegal position). However, 'retro-stalemate' as a keyword it has the sense more of applying to retro compositions where a retro-stalemate is sought in the history of the game, not Article 15 cases where the diagram is already one-sidedly illegal. So I think we should keep the keyword 'no legal last move for black'. (2021-10-10)
comment

Originalforderung: How is it possible for White to effect mate in two moves by strict chess?

'Recreationist': "A CHESS PUZZLE. We give below an interesting position which is to be looked upon as an end game occuring in actual play. It is calculated to test the ingenuity of those of our subscribers who delight in unravelling mysteries."

Erinnert von der Machart her an P1323153 und ähnlich gelagerte Probleme.

Interessanter Leserbrief von F. Dunne Okt. 1873 S.121: "It appears to me (albeit I amone of the solvers), that Mr. Barber's 'logical' solution of his Chess Puzzle is 'il'-logical, and I believe the following will clearly prove it so. Mr. Barbier says, 'What was Black's last move? Clearly not a pawn or bishop'; and he proceeds to prove that Black 'must' have made a false move with his King, 'because' he could not legally have moved it or any other piece into the position it or they now occupy.
Now surely that is not 'logically' proving that he 'must' have moved his King! For by the same 'logic' I can prove that he must have made a false move with, say, the Knight or any other piece.
'What was Black's last move? Clearly not the King, for all the squares from he could have reached the one he now occupies, are already occupied, either by his own, or by his opponet's pieces, or under the range of the latter's pieces; and he could not have Castled, as the square over which he would have to pass is commanded by one of the adverse pieces.
It is quite evident the pawns have not been moved, and, therefore also, not the bishop. The Rook cannot have been move last, or the White King would not occupy the square he now does. The he 'must' have moved the Kt, and as he could not legally have moved this piece to the square it now occupies, he must have made a false move with it.
Now, one conclusion is as logically arrived at as the other, and by parity of reasoning any other of Black's pieces can be proved to be te one last moved, as Castling across check is no more strict Chess, than wrongly moving a piece."

Gedanken dieser Art haben vermutlich Roberto Osorio & Co. zur Einführung des Begriffs MDR (Minimum Deviation from the Rules) inspiriert (s. Hinweis P0005584)

Antwort vom Autor im 'Recreationist 12/1873 S.154: "In reply to Mr. Dunne's letter in No.9 I beg to say that the solution of the Puzzle cannot be taken apart from the conditions attached to it. Any solution means the solution of something or other, and must be taken inconnection with that something.
The question is to find a way in which White mates in two moves. I agree that if, saying nothing about a mate in two moves as a condition, I endeavoured merely to prove Black's last move, Mr. Dunne's argument would be perfectly sound. But the fact of Black's castling is not an absolute fact; it is only relative. He castled because thereby I can mate him in two moves which Mr. Dunne's move does not do, and which is the very thing that is required to be done.
Henrik Juel: So even if one accepts the dubious retro reasoning intended, the problem is cooked by 0... Kxd8 1.bxc8=DT,Lf6+ (2017-09-30)
A.Buchanan: Dubious or not, retract-and-replay *is* one of the canonical chess problem jokes that our great-great-grandparents must have split their sides laughing over. I am pretty sure there is a typo in the diagram. sLc8 should be b8. This restores soundness and adds a retro try. wBb7 is still necessary to cover c8. (2017-09-30)
A.Buchanan: Have pushed sLc8 to b8, to remove the obvious typo. (2018-10-06)
A.Buchanan: I would classify these problems as jokes, not because the positions are illegal, but because imprecise reasoning is required to determine the illegal move. This is a particularly nice example though. Any thoughts? (2018-10-06)
SP: Your "obvious typo" is imaginary; the intended solution after Kd8: is
explicitly stated to be 1.bc8:=Q+ Kc8: 2.Qa8# (p. 91, viii/1873), quite
impossible with a bBb8. Please correct the position. (2023-02-18)
A.Buchanan: Hi SP. thanks for your research. Have you got a link to this material - that would be very helpful for checking similar situations. Otherwise our discussions here are often speculative. Bad news for the composer though: what I had thought was a typo turns out to be a cook. It would be good to know the version of the touch move rules applying in 1873, as the rook penalty move is clearly thematic. (2023-04-03)
A.Buchanan: The stuff about bS having just made an illegal move is basically Dawson & Hundsdorfer's distinction between plausibility & implausibility. This is an implausible composition, and having said that, one need agonize no longer, and just enjoy it for the best it can be at this silly idea (2023-04-03)
comment

Henrik Juel: This un-dummy-pawn is like a pawn created on c8 by the Einstein condition
The retroplay looks wrong, what about 1... d8-d6!? 2.d7-d8=sB!? c6xYb5 (2020-04-30)
James Malcom: That makes much more sense Henrik. I have edited it accordingly. (2020-04-30)
A.Buchanan: You don't like Adrian's (: haha? What about :) then? In heraldic terms, it would be described as "regardant". See 270 on https://www.heraldica.org/cgi-bin/atlas.pl?12. I don't know the word for "non-regardant", which is the default direction for the head, e.g. most of the others on that page.
We could replace wBa6a7 with wSa6, but alas having the 3rd knight is non-thematic & distracting. The other thing I would've liked to do is empty b2 so the promotion is forced to be to D. Such a tiny problem should have precision promotion. Needs more thinking. (2020-05-01)
James Malcom: UPDATE: The position has been edited to have one less piece and a precise queen promotion. (2020-10-12)
A.Buchanan: There is dialogue here which is now orphaned by a change of diagram with new matrix. Suggest we revert to the older diagram and mark it superseded by a new PDB entry here (2020-10-12)
James Malcom: I agree-it is now done. (2020-10-12)
James Malcom: Oh, it’s (:! I remembered it wrong!

Also, here is one idea to make it so that queen promotion is the only one: k7/Pp1N4/1PpK4/Rp6/1P6/8/8/8
1. dxc7ep b6, and 2. c8=Q# to cover the newly vacated b7 square. Retro play is the same. (2020-05-01, hidden)
A.Buchanan: I don't think that works. Bl can retract a6xb5, and more comically: b8-b7 or c8xb7 (2020-05-01, hidden)
James Malcom: You’re right- it doesn’t work at all. (2020-05-01, hidden)
A.Buchanan: Let's just stick with the existing humble but functional version (2020-05-01, hidden)
James Malcom: Agreed. (2020-05-02, hidden) more ...
comment

James Malcom: This is not Staugaard castling! It is another, far less known form of joke castling. I wrote about it here: https://www.ecforum.org.uk/viewtopic.php?t=10857

See also P1378161, P1378167, P1373226, P1378169, and P1378170 (2021-11-02)
comment

A.Buchanan: Hi James, tiny point. Does 1. ... La7 allow both 2. GKxf6xd4# and 2. Lxe5# ? Can this be fixed by shifting wBc7 to b8? (2023-07-01)
comment

Henrik Juel: I see no mate, 2... Kf8 (2020-10-15, hidden)
James Malcom: Ugh-wSh7 added in. (2020-10-15, hidden) more ...
comment

Henrik Juel: The ep key is unjustified (2020-10-24)
A.Buchanan: There should be no space after comma where there are multiple alternative moves listed. This is not a stylistic choice but a part of the syntax. This usually forces anything after the space to be marked as an error by the animation which is functioning correctly here, but it would be possible to compose a problem where the orphaned text is mistaken for a move by the other player! If someone wants to request that Gerd changes his rules, then please go ahead but adding random spaces in solutions is well-meant but muddle-headed. (2020-10-25, hidden)
A.Buchanan: There should be no space after comma where there are multiple alternative moves listed. This is not a stylistic choice but follows Gerd's syntax rules. Putting spaces usually forces anything after the space to be marked as an error by the animation, but it would be possible to compose a problem where the orphaned text is mistaken for a move by the other player! The notion of "comma-separated values" (CSV) is well-established in computing, although it is not usual in prose. (2020-10-25, hidden) more ...
comment

A.Buchanan: Nice way to combine an orthodox theme with Bosma castling (2020-11-13)
James Malcom: Thanks, Andrew. Castling with check is the easiest way to achieve it of course, not to mention the flight taking. I'll try out a version without check, here soon. (2020-11-13)
comment

'DSZ': "Diese Aufgabe regt den Gedanken an, daß die itaienische freie Rochade noch viel besser hätte ausgenützt werden können, als es (ehedem) tatsächlich der Fall war."

ähnelt stark der P1277302
A.Buchanan: @Henrik how should the dual here impact C+ status we assign? What's the general principle we curators should apply, to avoid us having to make subjective judgements? (2021-01-07)
A.Buchanan: My own simple-minded aesthetic says this is a major dual, because one mate achieves no individual expression elsewhere (2021-01-07)
Henrik Juel: The easiest principle is to say that C+ means completely correct: no cooks and no duals
I would permit short mates, because they occur as a result of suboptimal defense, and double threats, if they do not materialize as duals
But it seems fair also to permit minor duals, including promotion mate duals like TD og LD
So I would not label this problem C+ (2021-01-08)
A.Buchanan: Thank you Henrik, for your thoughtful response. A lot is bundled into one bit of information: the value of the C+? flag. I went back to the Codex to learn more. I think C+ should mean (1) it's been tested & (2) it is sound. Sound (which we don't have in PDB yet) for an antagonistic problem (i.e. d# or s#) means three things: (1) there is a solution (2) there's no cook & (3) there's a defence which forces the specified number of moves. Legality doesn't impact soundness, and for antagonistic problems, neither do duals.

The Codex says there is a lot of judgement involved in assessing the impact of duals in a d#. "C+?" is a result of computer analysis, and I don't think there should be any curator subjectivity in determining its value. I think the answer is to have "duals" as a new separate field so that your valuable analyses aren't lost in conversation, and people can understand the spectrum of quality. In the end, maybe people will be able to set up in their profile their personal preference for dual tolerance, but that's probably some way ahead. I think there are naturally core fields (like cooked?, legal?, tested?) and derived fields (like sound? C+?, HC+?) whose values depend on the core ones. I've sent some notes to Gerd along these lines to see what he thinks. Any thoughts yourself? (2021-01-08)
comment

Originalforderung:
This Problem's thus enunciated;
And in the statement is no flaw:
White, with first move, can get checkmated;
Black, moving first, can only draw.

Das 'Sussex Chess Magazine' merkt dazu an: "There is a touch of brimstone about this. Strict disciplinarians had better not try it, if they value their peace of mind."
Und da keine Lösungen eingingen, wurde in der Ausgabe vom 14.03.1883 ergänzt: "... we will hint that, in the first part, White takes advantage of the lax wording of the law of promotion by claiming a black piece, and thus 'qui perd gagne'. The second part is plain sailing ..."
Henrik Juel: In part b) the remis seems wrong, e.g.
5.Kxf3 Kg5 6.Thg4# (2021-01-03, hidden)
Henrik Juel: Black can avoid losing by playing
4... Df1+ 5.Kg3 Dh3+ and perpetual checking (or stalemate if White captures sD) (2021-01-03, hidden)
James Malcom: You are correct, that is the solution Henrik; I messed it up. Trying to decipher descriptive notation can often lead to wild confusion. (2021-01-03, hidden) more ...
comment

Henrik Juel: good joke (2021-04-30)
comment

'Los Angeles Times': "This being April Fool's Day the column is devoted pretty much to the fair-enuf sex; for the female of the species reads more readily than the male."

Zur Autorin, die auch eine aktive und erfolgreiche Löserin war, heißt es: "Clara Cameron, Pomona, Cal.", und zur Aufgabe: "A variation of Napier's setting."

vgl. Marshall's Chess "Swindle" P1182118
Henrik Juel: Strange joke
There is no mate after 1... h3!
Also, after 1... Kxg6? Sg6 no longer stands on g6 (2022-02-25, hidden)
Henrik Juel: I see now that the intended 'mate completion' 3.Sxg6 brings a knight back to g6
But 2.f8=S+ is not mate, 2... Kf7,Kg5,Kh5! (2022-02-25, hidden)
James Malcom: I screwed up the position. It is now fixed. (2022-02-25, hidden)
Henrik Juel: Thanks James
There is still the problem of no mate: 2.f8=S+ Kf7!
Maybe there should also stand a white pawn on e6 (2022-02-25, hidden)
James Malcom: Sigh... bumped up the wK. (2022-02-25, hidden)
Henrik Juel: Yes, letting wK guard f7 is the way to fix it, of course
Now we can say C+ Popeye 4.61, finally (2022-02-25, hidden) more ...
comment

Henrik Juel: Popeye 4.61 finds no solution (2022-09-11)
Henrik Juel: Joke problem
1.d8=sL?? thr. 2.g8=S#
1... Lc7 2.Sd7# (2022-09-12)
Mario Richter: Does Henrik's answer show that the problem is cooked? (2022-09-12)
Henrik Juel: Yes, Mario (2022-09-12)
A.Buchanan: Could it be that bRb6 is meant to be on a6? (2022-09-13)
A.Buchanan: The point being that with bRa6, 1. d8=sL? La5! (not Lc7/Lb6 2. Sd7#/Te6#) So bishop promotion becomes a thematic try. (2022-09-13)
Mario Richter: I have access to the original issue of 'Shakhmatna Misl', so I can confirm that the black rooks are on a7 and b6. I checked some of the following issues to see, if there was a mentioning a of the cook and/or a correction, but without success.
Changing the cook 1. d8=sL! into a try would be nice, but I think the purpose of black Tb6 is to prevent the defense 1. ... Db3!, so unless I'm missing something, Andrew's suggestion doesn't work. (2022-09-13)
A.Buchanan: Thanks Mario - I didn't check enough. If further one swaps bQb8 with bRa7, and pushes wK to h2 further out of harms way, then I think the =S solution & =L try are sound: 1r6/q2Pp1P1/r4k1N/2N3R1/5p1P/8/4R2K/8 (2022-09-13)
James Malcom: Nice find, Mario; sweet fix, Andrew. (2022-09-14)
A.Buchanan: Thanks James. r7/q2Pp1P1/r4k1N/2N3R1/7P/8/8/4RK2 further saves a pawn and has better positions for the remote units. I suggest attribution is MM correction HJ,MR,AB. (2022-09-15)
A.Buchanan: Alternatively r7/q2Pp1P1/r4k1N/2N3R1/5p1K/8/8/4R3 economizes a wP rather than a bP. I don't know which version my esteemed co-creators would prefer to record this momentous achievement for posterity. I don't see a way to reach Meredith (2022-09-17)
comment

Henrik Juel: Popeye 4.61 finds no solution (2022-09-11)
SCHRECKE: 1. a8=sS!
1. ... Sb6 2. Tc5#
1. ... Sc7 2. Sb8# (2022-09-12)
Henrik Juel: A joke problem, I see... (2022-09-12)
Mario Richter: Sorry, Henrik, I thought you've learned from the past that I sometimes can't resist to present a little trap. I was sure you would detect that popeye finds no solution to this seemingy simple position, but I expected that you - knowing me - would conclude that there must be something more about it (even more, since I presented it at Weekend). So I hope you can forgive me my little joke ... :-) (2022-09-12)
Henrik Juel: No problem, Mario
And Michael saw the trick (2022-09-12)
comment

Henrik Juel: Popeye 4.61 finds no solution (2022-09-11)
Henrik Juel: A third joke problem with two fake mates
No zugzwang, but the threat 2.gxh8=sT??# (2022-09-12)
A.Buchanan: A version r5nr/pR4Pk/7p/7P/8/8/K7/8 allows for a third fake mate (2022-09-13, hidden)
A.Buchanan: A more economical version 6nr/R5Pk/5p2/7K/8/b7/8/8 offers a third fake mate too (2022-09-13, hidden)
A.Buchanan: A version 6nr/R5Pk/5p1p/7K/8/b7/8/8 offers a third fake mate too (2022-09-13, hidden) more ...
comment

Henrik Juel: Try 1.d8=sL? thr. 2.g8=S#, 1... Lc7/Lb6 2.Sd7/Te6#, but 1... La5!
Solution 1.d8=sS thr. 2.g8=S#, 1... Sa7/Sc6/Se6/Sf7 2.Sd7/Te6/Se4/Sg4#
Thanks for your efforts, Andrew (2022-09-18)
comment

Chess problem by AI program ChatGPT

Q: Can you give an example of a chess retrograde analysis problem?

A: A retrograde analysis problem in chess is a problem where the goal is to determine the sequence of moves that led to a given position. For example:

White to move and mate in 2
[FEN "8/8/1p6/1P6/3k4/8/8/K7 w - - 0 1"]

In this problem, the goal is to determine the two moves (one for white and one for black) that led to the position shown. The solution is 1. bxa3+ Kd4 2. Kb2#
Henrik Juel: I do not understand this at all
It seems that ChatGPT is worthless for retro problems (2023-01-29)
A.Buchanan: Generally ChatGPT is amazing. We have entered a new age. Interacting with it reminds me of the first time I sent emails, or used a browser: we have crossed a frontier into engage with a function that will soon be commonplace. I already use it as a work tool for the last week. However it has limited skills and I was surprised it even attempted to construct a chess problem. It’s failure is humorous and kind of dreamlike (2023-01-30)
Henrik Juel: Thanks for your clarification, Andrew
ChatGPT is so amazing that it will give you a grade of B on some university exams; using it is obviously cheating, but do you enforce it?
In my youth calculators were developing fast; initially I banned the use of calculators in my OR exam, but the students were unhappy about it, saying that they were allowed in high school; I was forced to allow calculators, but then the poor students pointed out that the rich students had an unfair advantage; I then bought a few hundred cheap calculators and let the students borrow one for the exam (the rich students were still mad, but had to accept their lot in the social-democratic society of Danmark) (2023-01-30)
comment

Henrik Juel: a mate after 1... Kd4 needs moving Ta8 to d1, not d8 (2023-03-05)
James Malcom: Fixed. (2023-03-05)
Henrik Juel: Thanks, James (2023-03-05)
comment