WCA Documents: Updated January 1, 2013

[Bob]

Assuming scramblers follow 4g as intended, there is no way that the competitor can blame a twisted corner during scrambling. I agree with Dene; it's the competitor's responsibility for having a vulnerable puzzle.

This said, is it feasible for scramblers to check every scrambled puzzle before sending it off to be solved, as far as time effectiveness/efficiency goes? With events like 2x2, 5x5-7x7, and 5BLD (and FMC ofc), puzzle "defects" like this are less bound to happen because of the physical capabilities of the puzzles. What then, if scramblers are told they are required to meticulously verify the scrambled state of every puzzle, despite the improbability of a defect like a corner twist from occurring? And then, what if the improbable through some twisted black magic manifests itself on a puzzle?

This isn't really feasible. When scramblers made mistakes scrambling Megaminx I told them to ignore them. It is not worth delaying the competition any further.

However I told scramblers that I'd they suspect they twisted a corner to just scramble randomly and give it back to the competitor to be solved so we can rescramble.

 

[uberCuber]

Well, I forgot to tell them about the video evidence, so it's my fault.

The fact that you would actually need video evidence is an annoying part of the problem, though hard to address without opening up an easy way to cheat. The same thing happened in one of my 4x4 solves at Nats '12, but again, since there was no way for me to prove to them that I didn't twist a corner during the solve, blame was put on me (and believe me, I know when I twist a corner on my 4x4. I had done it numerous times in the months leading up to that comp, and I very clearly felt the twist happen every single time, but no such thing happened during my solve). I wish there were an easy way to do something about it, but apparently the existence of 4g isn't enough.

 

[Ranzha]

This isn't really feasible. When scramblers made mistakes scrambling Megaminx I told them to ignore them. It is not worth delaying the competition any further.

However I told scramblers that I'd they suspect they twisted a corner to just scramble randomly and give it back to the competitor to be solved so we can rescramble.

Which has higher priority--time effectiveness or proper scrambling?

Perhaps having more scramblers would balance the regulatory deliberation 4g requires? Or should 4g be rethought?
I like the way 4g is written. It establishes responsibility and should help prevent cases like Antoine's.

 

[Dene]

"may" is not the same as "must" (and 5b3 btw speaks about doing that *after* repairing, not *to* repair).

Either you're playing dumb and hoping I won't notice, or you're over-thinking the regulation and have thus confused yourself... The word "may" is specifically in reference to the choice the competitor has of whether to repair the puzzle or not (i.e. fix it before stopping the timer, or taking a DNF). The word "may" is NOT in reference to the method of reparation (disassembling vs. not disassembling).

As to your note in brackets: ok perhaps my use of the word "repair" was inappropriate there, as the word is being used as a technical term in the regulations. But I clearly meant "repair" in respect to bringing the puzzle back into a solvable state from an unsolvable state. Once again either you're playing dumb or you're overthinking it. You won't catch me out that easily though

There's deliberate, and there's sloppy. Like you said, people aren't perfect. And corner twists can happen even with a good puzzle (happened with mine, and they're not at all horribly loose).

I agree that corner twists can happen to good, and even tight puzzles. My own Zhanchi can have corner twists, despite being fairly tight. However I am aware that this is a risk inherent in my choice of puzzle, and I accept that like a man.

I know it could be harsh, and in a perfect world we'd never have this problem due to scramblers checking every cube perfectly before submitting them for solving. However we live in a world with practical constraints in which oversights can occur. It's unfortunate, but that's the way it is for now.

Perhaps I should be more sympathetic in future, as a delegate, to allow people another chance if a corner twist shows itself at the end of a solve, but only if a trustworthy judge insists he didn't notice anything that might have indicated a corner twisting during solving. But there's a fine line, and if it gets crossed rules could get abused (in the same way the old POP rule could get abused). This is the tough part of a delegates role ^_^

Ooh, excellent point! Since "the scrambler must verify that he has scrambled the puzzle correctly" and "If the puzzle state is wrong, he must correct it ", even if you blame a corner twist on the puzzle and thus the competitor, not verifying+correcting it is 100% the scrambler's fault who is explicitly obligated to do that.

I'm going to repeat myself a bit but I'm tired and I already wrote the above stuff before I got to this and I can't be bothered to integrate my response...

You talk about "fault", but deciding who is at fault is a tricky situation. Obviously, if there is evidence the corner was twisted by the scrambler (such as video/photo evidence) then the answer is simple. But otherwise it is impossible to know; judgement calls have to be made; not everyone is going to be happy with the decision... etc. etc. refer above for more stuff.

(Look at that, I didn't really repeat myself after all ).

 

[Stefan]

Either you're playing dumb and hoping I won't notice, or you're over-thinking the regulation and have thus confused yourself... The word "may" is specifically in reference to the choice the competitor has of whether to repair the puzzle or not (i.e. fix it before stopping the timer, or taking a DNF). The word "may" is NOT in reference to the method of reparation (disassembling vs. not disassembling).

No, I'm actually serious and don't think I'm overthinking. Where did you get your "must" from then? I don't see anything like it in 5b3. And that's why I think Francois is right and corner twisting was already allowed in 2008.

As to your note in brackets:

Yah, I knew what you meant, was just a minor side note, that's why I put it in brackets and with a "btw"

However we live in a world with practical constraints in which oversights can occur.

Twisting corners is probably the only accident that leaves puzzles unsolvable and might go unnoticed. Maybe we should make it a guideline for scramblers that at least the corners should be checked? At least in finals, at least at major competitions? For cubes, the eight stickers on opposite sides would suffice. And might be a better check than what scramblers are doing right now (cause with our notation, every turn involves corners).

 

[Bob]

Which has higher priority--time effectiveness or proper scrambling?

Perhaps having more scramblers would balance the regulatory deliberation 4g requires? Or should 4g be rethought?
I like the way 4g is written. It establishes responsibility and should help prevent cases like Antoine's.

In the case of Megaminx, an incorrect move or two still yields a randomly scrambled puzzle so I say time effectiveness has a higher priority. I agree more scramblers would have helped (and a stricter cutoff would have helped even more) but that was not an option at the time as our staff was already stretched to capacity.

 

[Dene]

No, I'm actually serious and don't think I'm overthinking. Where did you get your "must" from then? I don't see anything like it in 5b3. And that's why I think Francois is right and corner twisting was already allowed in 2008.

The regulations only provided one choice for bringing an unsolvable cube into a solvable state (disassembling and reassembling). We are then left with two ways to interpret the regulations to apply to this particular situation which isn't directly specified (corner twist):
1) Disassembling and reassembling is the only option available to deal with unsolvable cubes
2) Disassembling and reassembling is just one method which can be used to deal with unsolvable cubes, among others

I am of the opinion that the most appropriate interpretation is option 1. When I read the regulation it comes across to me as saying "either disassemble and reassemble to fix the issue, or take the DNF". I think the only way to come to option 2 is if you are deliberately trying to look for a loophole. The first rule of any "close-reading" like this is to be charitable to the author; i.e. looking for loopholes is not cool (although helpful to point out for the sake of amendment).

Twisting corners is probably the only accident that leaves puzzles unsolvable and might go unnoticed. Maybe we should make it a guideline for scramblers that at least the corners should be checked? At least in finals, at least at major competitions? For cubes, the eight stickers on opposite sides would suffice. And might be a better check than what scramblers are doing right now (cause with our notation, every turn involves corners).

I think this would be a very good idea, and I will make sure that at all future Australian and New Zealand competitions that every scrambler checks corners before submitting a cube for solving.

 

[cubernya]

I actually think it would be a good idea to check the corners. Even the white/yellow (or U/D) color placement is enough, since you're only checking for twists. I think it would be good to add that into the regs (on the next change) or in the guidelines.

 

[Stefan]

We are then left with two ways to interpret the regulations to apply to this particular situation which isn't directly specified (corner twist):
1) Disassembling and reassembling is the only option available to deal with unsolvable cubes
2) Disassembling and reassembling is just one method which can be used to deal with unsolvable cubes, among others

Right, and I guess I'm more in camp 2. Or in camp 1 and I interpret the "disassemble and assemble max. 3 pieces" as including zero pieces. Anyway, how about we drop it cause we have a clear ruling now anyway?

I actually think it would be a good idea to check the corners. Even the white/yellow (or U/D) color placement is enough, since you're only checking for twists.

Might not catch as many misscrambles, though, and might be harder to do than checking fixed eight sticker positions on two opposite sides. But that's just my gut feeling, didn't try it or think about it much.

 

[Mike Hughey]

This isn't really feasible. When scramblers made mistakes scrambling Megaminx I told them to ignore them. It is not worth delaying the competition any further.

I know you did this and have done it in the past, but isn't it technically not only against the letter of the rules, but against even the spirit of the rules? After all, we have specific instructions that the delegate may, at his or her discretion, approve an improperly scrambled 6x6x6 or 7x7x7, but have not included megaminx in the list of such puzzles. If we wanted to allow misscrambled megaminxes, shouldn't we include them in the list of puzzles that may be misscrambled?

(This has always bothered me - I wonder if it shouldn't be added.)

 

[Bob]

I know you did this and have done it in the past, but isn't it technically not only against the letter of the rules, but against even the spirit of the rules? After all, we have specific instructions that the delegate may, at his or her discretion, approve an improperly scrambled 6x6x6 or 7x7x7, but have not included megaminx in the list of such puzzles. If we wanted to allow misscrambled megaminxes, shouldn't we include them in the list of puzzles that may be misscrambled?

(This has always bothered me - I wonder if it shouldn't be added.)

Yes it is. IMO Megaminx is the most difficult puzzle to verify the state and when that regulation was originally written, a picture wasn't provided for Megaminx to verify against. Does that regulation still exist?

 

[TheNextFeliks]

Yes it is. IMO Megaminx is the most difficult puzzle to verify the state and when that regulation was originally written, a picture wasn't provided for Megaminx to verify against. Does that regulation still exist?

It just says that 6x6 and 7x7 are the only you don't have to solve and re-scramble.

 

[Mike Hughey]

4g) After scrambling a puzzle, the scrambler must verify that he has scrambled the puzzle correctly. If the puzzle state is wrong, he must correct it (e.g. by solving the puzzle and applying the scramble sequence again).

4g1) Exception: For the 6x6x6 Cube and the 7x7x7 Cube, it is not necessary to correct the scramble, at the discretion of the WCA Delegate.

 

[Goosly]

4g) After scrambling a puzzle, the scrambler must verify that he has scrambled the puzzle correctly. If the puzzle state is wrong, he must correct it (e.g. by solving the puzzle and applying the scramble sequence again).

I wonder how many scramblers actually check the full state. Usually one would just check a certain pattern, like a row of yellows on the right side and a blue-green-blue pattern on top (random example). I doubt more than half of the scramblers would notice a cornertwist - unless of course, they felt something went wrong during the scrambling.

 

[Rune]

"And I want to add oldest solver and youngest solver.
Haruo Aoki, he is 77 years old, enjoyed looking these data and these data gave his power"

You must understand that there are big risks involved here for men of Aoki´s age. Maybe he has children - or friends - who with the best intentions urge him to try harder in order to accomplish some results. If he follows the "advice", he may ultimately come under stress, which can lead to a heart attack with death as a consequence.
We must by all means avoid such an outcome and therefore the first step will be to stop the ranking list for the oldest solvers.

 

[Kattenvriendin]

Youngest then too.. child labor! :rolleyes:

 

[TheNextFeliks]

I wonder how many scramblers actually check the full state. Usually one would just check a certain pattern, like a row of yellows on the right side and a blue-green-blue pattern on top (random example). I doubt more than half of the scramblers would notice a cornertwist - unless of course, they felt something went wrong during the scrambling.

Usually you can feel a corner twist. I checked it and I know others checked at my comp. Except one time. It was Sarah's fault.

 

[Dene]

On the topic of megaminx, if I recall correctly the scrambler we use doesn't create a random state, nor even come close to covering all possible cases for a megaminx. Am I remembering correctly? If so, is it still the same? Because if it is, I don't see how it can be *meaningfully* enforced to have the exact scramble result as provided.

 

[Bob]

On the topic of megaminx, if I recall correctly the scrambler we use doesn't create a random state, nor even come close to covering all possible cases for a megaminx. Am I remembering correctly? If so, is it still the same? Because if it is, I don't see how it can be *meaningfully* enforced to have the exact scramble result as provided.

You are correct.

 

[Stefan]

Because if it is, I don't see how it can be *meaningfully* enforced to have the exact scramble result as provided.

I don't understand what you mean here.