In Loop (Shaded), the instructions only ask for the existence of a loop, but not its unicity.
Is this intentional or should the unicity of the loop be added to the constraints ?
Permalink Submitted by onigame on Thu, 08/12/2021 - 20:10.
Uniqueness of solution is not a rule for ANY of the puzzles. For example, if you look at puzzle 1, Easy As, the instructions do not mention whether the solution is unique. Or, if you look at the standard rules for Classic Sudoku, they also do not specify that the solution is unique.
However, it is generally considered bad style to write a puzzle that does not have a unique solution, and so it is often safe for solvers to assume that the solution is unique when solving and use that assumption as a shortcut. But it *is* an assumption, not a rule, and solvers do so at their own peril.
In this specific case, the puzzle uses the wording "You must be able to draw..." instead of "you must draw..." (or "Then, draw...") because the Answer extraction rule doesn't require you to actually draw the loop, only identify the location of shaded cells. It has nothing to do with uniqueness of the loop.
(I do notice, after careful reading, that there *is* a mistake in those instructions; the last part of the answer should say "shaded cell" instead of "ship piece". I will get that fixed.)
Permalink Submitted by Gotroch on Sat, 08/14/2021 - 12:27.
Uniqueness of solution is not a rule for ANY of the puzzles. For example, if you look at puzzle 1, Easy As, the instructions do not mention whether the solution is unique. Or, if you look at the standard rules for Classic Sudoku, they also do not specify that the solution is unique.
However, it is generally considered bad style to write a puzzle that does not have a unique solution, and so it is often safe for solvers to assume that the solution is unique when solving and use that assumption as a shortcut. But it *is* an assumption, not a rule, and solvers do so at their own peril.
Wei-Hwa, can you assure us, that at Puzzle GP competition (and i hope in Sudoku GP as well), all puzzles are thoroughly tested and players won't see any of such bad written puzzles with multiple solutions? So in fact we can safely use uniqueness when solving WPF GP puzzles, right? :-)
Permalink Submitted by onigame on Tue, 08/17/2021 - 09:56.
I will not assure you that all Puzzle GP puzzles have only one solution, because I want to leave open the possibility that a puzzle designer will have a very elegant puzzle that only works because it has more than one solution.
However, I will say that such a puzzle would have to be VERY impressive such that the violation of style would be worth the violation.
Uniqueness
Uniqueness of solution is not a rule for ANY of the puzzles. For example, if you look at puzzle 1, Easy As, the instructions do not mention whether the solution is unique. Or, if you look at the standard rules for Classic Sudoku, they also do not specify that the solution is unique.
However, it is generally considered bad style to write a puzzle that does not have a unique solution, and so it is often safe for solvers to assume that the solution is unique when solving and use that assumption as a shortcut. But it *is* an assumption, not a rule, and solvers do so at their own peril.
In this specific case, the puzzle uses the wording "You must be able to draw..." instead of "you must draw..." (or "Then, draw...") because the Answer extraction rule doesn't require you to actually draw the loop, only identify the location of shaded cells. It has nothing to do with uniqueness of the loop.
(I do notice, after careful reading, that there *is* a mistake in those instructions; the last part of the answer should say "shaded cell" instead of "ship piece". I will get that fixed.)
Uniqueness
Uniqueness of solution is not a rule for ANY of the puzzles. For example, if you look at puzzle 1, Easy As, the instructions do not mention whether the solution is unique. Or, if you look at the standard rules for Classic Sudoku, they also do not specify that the solution is unique.
However, it is generally considered bad style to write a puzzle that does not have a unique solution, and so it is often safe for solvers to assume that the solution is unique when solving and use that assumption as a shortcut. But it *is* an assumption, not a rule, and solvers do so at their own peril.
Wei-Hwa, can you assure us, that at Puzzle GP competition (and i hope in Sudoku GP as well), all puzzles are thoroughly tested and players won't see any of such bad written puzzles with multiple solutions? So in fact we can safely use uniqueness when solving WPF GP puzzles, right? :-)
No Assurance
I will not assure you that all Puzzle GP puzzles have only one solution, because I want to leave open the possibility that a puzzle designer will have a very elegant puzzle that only works because it has more than one solution.
However, I will say that such a puzzle would have to be VERY impressive such that the violation of style would be worth the violation.
Your answer makes perfect
Your answer makes perfect sense ! I'm try not to make a habit of using uniqueness while solving, but still it is an usual assumption.
I had noticed the "be able to" part and should have checked the answer key, that would have confirmed everything.
Thank you for this reminder :)
GP8 Instruction Booklet - new version
There has been published new version of the Instruction Booklet!
https://gp.worldpuzzle.org/content/instruction-booklet-2
Best,
Karel (Admin)