Suggestions for WCA regulations 2008

[Ron]

Fellow cubers,

In the upcoming weeks WCA will be working on the new version of the WCA regulations, version 2008.

Here are some suggestions that we have gathered for this new version:
1) inspection time down to 10 seconds
2) new starting procedure (as proposed by Gilles beginning of this year)
3) format of Multiple Blindfolded event
4) procedure for making events official and retiring events
5) explicit brand neutralness
6) improvements on the scrambling of 3x3 and Megaminx
7) qualification rounds and seeding
8 ) improvements for the blindfolds (WCA is working on an universal hygienic blindfold)
9) clarification on the starting procedure of blindfolded solving
(current text is ambiguous)
10) standard notation
11) structure of decisions in WCA board
12) next phase of membership of WCA (elections, funding, membership responsibilities and rights)
13) improvements of the definition of the solved state

And there is also a list of textual changes as proposed by community. For this we received a list of 63 e-mails and suggestions from earlier messages on the WCA forum.

In this thread we welcome additional suggestions for the WCA regulations 2008.
For feedback on the suggestions above, please create a new thread on this forum with the title of the suggestion.

Thanks,

Ron van Bruchem

[cada]

Formatting: 1c5 isn't in the bullet list with the other 1c#s.

Scrambling: I support Stefan's new megaminx scrambling method and the new 3x3 scrambler suggested on the yahoo forum recently.

Puzzle defects: I don't know if these need be addressed in the regulations, but I have two questions.

If a competitor has a puzzle defect not caused by themself, are they still not given an extra attempt per 5c? At the Canadian open a cube had a corner twist in place during the scramble, the competitor had to fix the cube to finish the solve. Would they have been allowed to end that solve and be given another?

Similarly, what if the competitor notices the cube is defect before beginning the solve, but begins the solve anyway? In Canada, one competitor was given the incorrect cube during BLD, noticed this during memorization but chose to attempt the solve. After DNFing the solve, they were given an extra attempt.

Feet solving: This is the only event where a mean/average is an official format but not the preferred format. It seems odd to not have the more comprehensive format be the standard. I think mean of 3 should either be the preferred format or be eliminated.

Competition results: 9f2 seems to be regularly ignored.

One handed: For a pop/defect, can the table be used when repairing the puzzle?

[cada]

Seems there's no edit option.

Is this the new starting format you mention? I dislike the possibility of the judge having to stop a solve after the competitor has started, for example of cases where their hands were placed on the stackmat pads incorrectly. I'd like to take care of any problems like this before the competitor could possibly begin the solve.

Chris Krueger

[StefanPochmann]

Two more items for the list:

- Get rid of "B Finals". It's such a stupid thing. If you want to give more people another round, simply let more people get to the next regular round.

- The rules being easily accessible at the competition.

[BryanLogan]

Main judge for event - Get rid of it. The final rulings are always going to be handled by the WCA delegate. As long as there is an initial scrambler, and no one sees the scramble before they solve it, it's good.

2d - Allow them to prove their identity later. No need to make people bring their passports.

2e2 - How would NR's be handled if the competitor switched nationalities? More likely example is someone becoming a citizen and losing their old citizenship.

2t - add "or other means". For example, a competitor may be able to start and stop the timer, but need the timer sitting on their lap, without the mat. But just leave this regulation open for the odd situations that can be easily dealt with when they occur, but just can't be anticipated.

3a - Remove "like a " and all following text. They've already been defined.

Define solved states and +2's for other puzzles (Square-1, Megaminx, Pyraminx).

Get rid of closed competitions or at least discourage them. If the prizes should be restricted to certain citizenship, fine, but at least allow anyone to compete and have their times be official.

Leave the B-Final in. It's up to the organizer if they want to have it or not, and even if competitors aren't the fastest, they still want another chance to improve their PR's.

What's the 3x3x3 scrambling improvement?

[Edouard Chambon]

I agree to put down the inspection time. 7 seconds would be fine I think, because nobody solved it in less than 9 seconds.

If we decide that the judge do not cover the puzzle (which is a very good idea i think) and if the timer does not start, the competitor has to be blamed. So in this case we can put it at 10 seconds, because it include 1 or 2 seconds to take it and to put it on the table.

That's sure that 15 seconds are too much and has no really sense now.

[JChoi]

I talked to a few people at the Virginia Open and showed them this thread. They said:
"15 seconds is fine"
"Multiblind format doesn't matter because Matyas is going to win anyway"

Also, I talked to a couple people the night before and were talking about the solved state of the magic puzzles and how two tile height is a rather arbitrary value. How was this rule started? Could a change regarding this be considered (to 45 degrees instead or something else)?

[Clement Gallet]

Oh, I almost forgot :

About the Rubik's Clock : Why it has to be a mean of 3 ? Why not an average of 5 ?
It's a very fast puzzle to solve, and it's very easy to have a DNF on it.

Clément

[JChoi]

Could a standardized timesheet be made as well? That was talked about in Virginia too.

[BryanLogan]

Could a standardized timesheet be made as well? That was talked about in Virginia too.

A standard template that could be used (and generate the sheets from the spreadsheet registration) would be nice, but I don't think it necessarily needs to be in the regulations. It doesn't affect the competition in any way that would give a non-complying tournament any advantage.

[Edouard Chambon]

Oh, I almost forgot :

About the Rubik's Clock : Why it has to be a mean of 3 ? Why not an average of 5 ?
It's a very fast puzzle to solve, and it's very easy to have a DNF on it.

Clément

But not so easy to scramble !!! Grrr......

[BryanLogan]

4) procedure for making events official and retiring events
6) improvements on the scrambling of 3x3
7) qualification rounds and seeding
8 ) improvements for the blindfolds (WCA is working on an universal hygienic blindfold)
11) structure of decisions in WCA board
12) next phase of membership of WCA (elections, funding, membership responsibilities and rights)

Are there links for these discussions? I couldn't find them. Also, is there a draft of the 2008 that we can look at?

[Tim]

Oh, I almost forgot :

About the Rubik's Clock : Why it has to be a mean of 3 ? Why not an average of 5 ?
It's a very fast puzzle to solve, and it's very easy to have a DNF on it.

Clément

But not so easy to scramble !!! Grrr......

Is clock harder/slower to scramble than 5x5? I personally don't think so, and 5x5 is average of 5. Also, not a lot of cubers do clock, so there should be a lot of hands willing to scramble.

[Kenneth Gustavsson]

...the solved state of the magic puzzles and how two tile height is a rather arbitrary value. How was this rule started? Could a change regarding this be considered (to 45 degrees instead or something else)?

This is what I came here to write about and I was also thinking of propose 45 degrees for any angle as the solved state. That because there are far to many DNF's in Magic.

But that's not the only reason, more inportant is: "2 tiles thick". There are two types of Magic's, one with thick tiles and one with thinner ones. The thick ones was the common type but the thin type is more and more the one used by speedsolvers. At SCD we had a situation where, I think Matyas was mesuring Eriks sloppy solved magic and he was about to use his thin Magic for that. I told him to use a magic that had thick tiles intead, it was still a DNF. But if it's a close call, like the one Erik Johansson had at WC 2007, he got a DNF (mesured using Matyas Magic) if not he had won (I only heard about this so I don't know if it was a thick or a thin Magic he used then). I'm not asking to change a judgement, that's impossible I understand but the rules must be clear about this.

[BryanLogan]

For making events official, what about the idea of not acknowledging any records until the event has been official for one year?

That way you avoid the thrashing of the record while all the regions wait to have their first event.

[optakeover]

Nobody replied to Stefan's thread regarding a 0.02 second time on the stackmat at http://www.worldcubeassociation.org/forum/viewtopic.php?f=4&t=369. I think I have found something that would tally with this, and I feel that this should be addressed in the new regulations.

WSSA (World Sport Stacking Association) pioneered the use of Speed Stacks® StackMat timers and mats (well of course, they are THE world stacking organisation, right?) and we also use their timers. However, I found text in the WSSA Rule Book that shows much similarity to Stefan's description of the Stackmat behaviour. In the section Definition of Terms:

"Hiccup

An accidental slight movement of a hand or hands on the touch pads of the StackMat® causing the Timer to start seemingly on its own. Most often when this occurs the Timer will stop shortly after it starts. A Hiccup is not considered a “try”(a stacking term). Simply reset the Timer."

[Lucas]

B1E) ...In case the competitor solves under 10 minutes, then 10 minutes will be his result.

Can we take this out?

For Magic/Master Magic +2, I suggest this ruling:

G4a) The puzzles must be completely flat on the surface, with either of the two sides on top. Penalty: 2 seconds per column (of maximum size 1x3 tiles) not folded flat.
G5a) At most 2 columns (of maximum size 1x3 tiles) may be folded over from the solved state. Penalty: disqualification of the solve.

What happened to the 2-tile rule?



Clock should be average of 5 (not that I'll attend many competitions that host clock), I think. Accidents DNFs have even stopped Stefan and Mátyás from finishing averages...
Square-1 should also be avg5, but we don't have to torture our scrambles with it...

6x6x6 or 7x7x7: What if these come out? Are they not allowed until 2009?

Speed BLD: I think this should now become an official event; I'll create a separate thread with suggested regulations.

Could we also consider allowing 2x2x2 BLD? I personally don't care, but I know people who'd like to do it. It would take extra competition time, so perhaps few competitions would host it, anyhow. Also, WR's would be lucky.



Scrambling:
Our favorite topic; we've already concluded that current scrambling does not produce random scrambles. EO is 2.5 as likely to be solved after 25 random moves as it should be (by any common definition of orientation).

4e) Competitors must solve the same scrambles per round. At the main judge's discretion, scrambling algorithms in preliminary rounds may be randomly chosen from a pool of scrambles, for example to prevent cheating in large competitions.

What about combined finals? Or how about large events with no additional finals? Or events with few competitors that get stretched out over a long while (due to breaks, being a side event, or competitors being out for side events)?

Now, I suggest the following:

- Change the standard for 2x2x2 and 3x3x3 to "random-state scrambling".
4g) In order to ensure fidelity and fairness, the WCA encourages random-state scrambling. Random-state scrambling shall be defined by the following procedure (the following may be automated) for each scramble:

• 4ga1) A single state shall be selected randomly from all legal puzzle states.
  
• 4ga2) A scramble shall be generated that will allow the scramblers to scramble the puzzle into the selected state. The scramble is not required to be optimal.
  
• 4ga3) The main judge for an event shall have the discretion to remove unsatisfactorily "lucky" scrambles, but shall be reserved about such an action (to avoid unnecessary bias in selection). A 3x3x3 scramble shorter than 15 moves or a 2x2x2 scramble shorter than 6 moves shall be taken as lucky.
  

4h) Due to the high profile of 3x3x3 events and the nature of the 2x2x2 event, random-state scrambling is required for all 3x3x3 and 2x2x2 events. Random scrambles of 50 and 25 random moves, respectively, may be used as an alternative, though this is discouraged. Random-state scrambling is encouraged for all other puzzles, but is currently not required.

• Are 50 and 25 discouraging enough but acceptable as a backup? Scrambles could get lost, or maybe no one has access to a proper scrambler/computer and can only print WCA scrambles generated online...
• Should we allow requests to the WCA for scrambles? (Maybe even require? I don't think it's necessary...)
• So far we only have Kociemba's 3x3x3 solver, and he could probably make a 2x2x2 scramble addition feature (the functionality is already built in). We have a few Java solvers and ACube, but I'm not sure we want to use those yet.
• Should scrambling programs need approval? Should the WCA create a list or provide download of official versions? (Would submissions be taken in this forum?)
• Clock is already random, I understand, but I don't think anything else is necessary to force. We want to generate states that are not any faster (easier) to solve than random ones, and for other puzzles it's not as important. It's possible for Pyraminx and Square-1, though, and there's a 70-move slice-turn solver for 4x4x4 out there. If the WCA begs me, I could learn Java for a random-state pyraminx scrambles, or create something that works with the solver Jaap provides for Square-1...
• Can we also formally allow scrambles to be published 24 hours after the final event?

[BryanLogan]

B1E) ...In case the competitor solves under 10 minutes, then 10 minutes will be his result.

Can we take this out?

If you're not sure, then just take the B1d clause.

For Magic/Master Magic +2, I suggest this ruling:

G4a) The puzzles must be completely flat on the surface, with either of the two sides on top. Penalty: 2 seconds per column (of maximum size 1x3 tiles) not folded flat.
G5a) At most 2 columns (of maximum size 1x3 tiles) may be folded over from the solved state. Penalty: disqualification of the solve.

What happened to the 2-tile rule?

Can you post some pictures showing the situations? Rather than the 2-tile rule, I think that it should just be some standard height. Then the judge can just see if a some object of that height (a plastic sheet cut narrow) fits under. In fact, there could be a template used to judge +2 for the cubes also.

Clock should be average of 5 (not that I'll attend many competitions that host clock), I think. Accidents DNFs have even stopped Stefan and Mátyás from finishing averages...
Square-1 should also be avg5, but we don't have to torture our scrambles with it...

Why should clock and Square-1 be treated differently?

6x6x6 or 7x7x7: What if these come out? Are they not allowed until 2009?

Ron talks about the procedures for adding new events. Also, in 2006 there was a v2 of the regulations, so it's not like things couldn't change before the end of the year.

4h) Due to the high profile of 3x3x3 events and the nature of the 2x2x2 event, random-state scrambling is required for all 3x3x3 and 2x2x2 events. Random scrambles of 50 and 25 random moves, respectively, may be used as an alternative, though this is discouraged. Random-state scrambling is encouraged for all other puzzles, but is currently not required.

[*]Are 50 and 25 discouraging enough but acceptable as a backup? Scrambles could get lost, or maybe no one has access to a proper scrambler/computer and can only print WCA scrambles generated online...

Well, you can save the current WCA scramblers to your local disk. If you don't have a computer at the competition, are you calculating the averages by hand? If you've lost the scrambles, things are probably highly disorganized and chaos should break out soon. But really, I would never want to see "Random scrambles of 50", because I bet that would very greatly by the scrambler. People will unconsciously scramble a certain way.

[*]Clock is already random, I understand, but I don't think anything else is necessary to force. We want to generate states that are not any faster (easier) to solve than random ones, and for other puzzles it's not as important.

Why isn't it important for the other puzzles? Just because they're not "as popular", I think the WCA should be treating all events equal.

[*]Can we also formally allow scrambles to be published 24 hours after the final event?

I think you mean "require". I think it would be nice.

And other stuff as I think about it:

Regarding 3n, not the "bring you own puzzle part", but the "be ready when called". Move that out to another rule:

"When a competitor is called by an official, they must bring the puzzle to the official. If the competitor is currently competing in another event, the official should try to accommodate the competitor and call him at a later time."

You may want to add a statement that says competitors competing in the long events (FMC, big cubes blind, multi-blind), might miss out on an event if they busy doing the long event (you don't want to delay an event forever).

[Lucas]

B1E) ...In case the competitor solves under 10 minutes, then 10 minutes will be his result.

Can we take this out?

If you're not sure, then just take the B1d clause.

Well, right now it's technically not allowed. Also, rarely is anything judged to require more than 10 minutes (I've never heard of a judge deciding on this).
I remember Chris Hardwick on 5x5x5 BLD getting near one hour and then a 28 (US Nationals '06). What if the time limit was 40 minutes instead of 10, and the judge took the first as an indication for the second? The same could happen to a 4x4x4 BLD cuber who gets a 30 minute-success by being really careful and then gets confident enough on the second for a sub-10?
Why was this even put in the regulations?

For Magic/Master Magic +2, I suggest this ruling:

G4a) The puzzles must be completely flat on the surface, with either of the two sides on top. Penalty: 2 seconds per column (of maximum size 1x3 tiles) not folded flat.
G5a) At most 2 columns (of maximum size 1x3 tiles) may be folded over from the solved state. Penalty: disqualification of the solve.

What happened to the 2-tile rule?

Can you post some pictures showing the situations? Rather than the 2-tile rule, I think that it should just be some standard height. Then the judge can just see if a some object of that height (a plastic sheet cut narrow) fits under. In fact, there could be a template used to judge +2 for the cubes also.

Templates? At college clubs they don't even have money for covers.
Anyhow, almost every magic competitor yearns for a +2. Magics (unless you got really lucky or wore one in for a while without breaking it) tend to unsnap the last flip beyond control, and so completing an average is a matter of chance. This doesn't give a very accurate ranking at a competition, either...

Clock should be average of 5 (not that I'll attend many competitions that host clock), I think. Accidents DNFs have even stopped Stefan and Mátyás from finishing averages...
Square-1 should also be avg5, but we don't have to torture our scrambles with it...

Why should clock and Square-1 be treated differently?

Why should Megaminx, then? (Or 5x5x5 BLD? ) Square-1 is very difficult to scramble, and it's easy to make an incorrigible mistake -I think most scramblers have rarely played with one. It also takes longer to solve (and to fix if misscrambled).
I'm in favor of avg5 for Square-1, but good luck getting enough support...

4h) Due to the high profile of 3x3x3 events and the nature of the 2x2x2 event, random-state scrambling is required for all 3x3x3 and 2x2x2 events. Random scrambles of 50 and 25 random moves, respectively, may be used as an alternative, though this is discouraged. Random-state scrambling is encouraged for all other puzzles, but is currently not required.

[*]Are 50 and 25 discouraging enough but acceptable as a backup? Scrambles could get lost, or maybe no one has access to a proper scrambler/computer and can only print WCA scrambles generated online...

Well, you can save the current WCA scramblers to your local disk. If you don't have a computer at the competition, are you calculating the averages by hand? If you've lost the scrambles, things are probably highly disorganized and chaos should break out soon. But really, I would never want to see "Random scrambles of 50", because I bet that would very greatly by the scrambler. People will unconsciously scramble a certain way.

No...
Currently, the only way to easily generate WCA-style random-state-scrambles is CubeExplorer. That requires a Windows computer.
And I say "50-move scrambles," not scramblings (4b is still in place). You can use the current random scramble generator, but only with enough moves to ensure randomness beyond a good little doubt. Everybody still gets the same scrambles (except in a large round, etc.)...
Maybe 40 moves is fine for 3x3x3? And is "random-state scramble" a good term? (RSS )

[*]Clock is already random, I understand, but I don't think anything else is necessary to force. We want to generate states that are not any faster (easier) to solve than random ones, and for other puzzles it's not as important.

Why isn't it important for the other puzzles? Just because they're not "as popular", I think the WCA should be treating all events equal.

Sure! Okay, you go write a good random-state scrambler for Megaminx and 5x5x5, and the WCA will use it.
Popular also means more competitors and more chance of lucky scrambles (and more lucky solves, but that's not possible to control). I also prefer much having the confidence of of being able to say "Our 3x3x3 scrambles are completely random."

[*]Can we also formally allow scrambles to be published 24 hours after the final event?

I think you mean "require". I think it would be nice.

No, I don't. But 4b) is technically never lifted. Occasionally, official scrambles have been released, anyhow (but mainly of notable solves).

[Ron]

Hi fellow cube fans,

Thank you all for your suggestions.

An accidental slight movement of a hand or hands on the touch pads of the StackMat® causing the Timer to start seemingly on its own. Most often when this occurs the Timer will stop shortly after it starts. A Hiccup is not considered a “try”(a stacking term). Simply reset the Timer.

We have something like that already. See 11a3) Equipment malfunctioning.

B1E) ...In case the competitor solves under 10 minutes, then 10 minutes will be his result. Can we take this out?

Yes, we should take this out.
Solution:
1) for events: 4x4 blindfolded, 5x5 blindfolded, 3x3 multiple blindfolded:
- we use stopwatch
- if asked by the competitor we use Stackmat in addition to stopwatch, valid Stackmat times preceed stopwatch times
2) for events: 3x3 blindfolded
- we use Stackmat and stopwatch
3) for events: 3x3 fewest moves:
- we use stopwatch
4) for all other events:
- we use Stackmat
If Stackmat and stopwatch are both used: if time < 10 min we take Stackmat time. Otherwise we take stopwatch time.

For Magic/Master Magic +2, I suggest this ruling:
G4a) The puzzles must be completely flat on the surface, with either of the two sides on top. Penalty: 2 seconds per column (of maximum size 1x3 tiles) not folded flat.
G5a) At most 2 columns (of maximum size 1x3 tiles) may be folded over from the solved state. Penalty: disqualification of the solve. What happened to the 2-tile rule?

OK, I would not mind to introduce penalties for Magic. But prefer to have one penalty for solved state.
So:

10g) For Magic (and similar puzzles) no tiles of the puzzle must be elevated higher than 2 tiles.
10g1) Penalty if 1 tile elevated higher: 2 seconds.
10g2) Penalty if more than one tile elevated higher: disqualification of the solve.

Clock should be average of 5 (not that I'll attend many competitions that host clock)

My theoretical reaction is: yes of course.
My practical reaction is: no, it is fine like this and the more time a Clock event will take, the fewer competitions will host it.
I think we need a difference between 'main events', 'side events', 'no event'. For 'main events' the circumstances should be optimal, for 'side events' we have to bargain against available time. Side events can have fewer attempts, be on a side stage or in a separate room, be in parallel with other events et cetera. In that case it is more important to HAVE the event, so we can make some puzzle fans more happy.

6x6x6 or 7x7x7: What if these come out? Are they not allowed until 2009?

As said earlier, we need a procedure to decide which events are official and which aren't.
I would love to have 6x6x6 and 7x7x7 events, but at the moment I see 3 issues:
1) availability of the puzzle
2) time the event will consume in a competition (scrambling, average solving time)
3) timing of the solves, because the solving will often be around the limit of the Stackmat, with some competitors well below 10 minutes, but many competitors around or above 10 minutes.

Speed BLD: I think this should now become an official event; I'll create a separate thread with suggested regulations.

Feel free to do this and we will evaluate it.

Could we also consider allowing 2x2x2 BLD?

Same answer: we need a procedure.

Our favorite topic; we've already concluded that current scrambling does not produce random scrambles.

For 2x2 and 3x3 I will propose to use the new feature of Cube Explorer.
If members inside or outside our community would build programs for random state scrambling of other puzzles, then we would gladly evaluate them.

4e) Competitors must solve the same scrambles per round. At the main judge's discretion, scrambling algorithms in preliminary rounds may be randomly chosen from a pool of scrambles, for example to prevent cheating in large competitions.
What about combined finals? Or how about large events with no additional finals? Or events with few competitors that get stretched out over a long while (due to breaks, being a side event, or competitors being out for side events)?
...

I think we need a separated area where all competitors wait who have delivered their puzzle for scrambling.
From there I think we have to separate three situations:
1) a round where all competitors are waiting in the separated area: all competitors have the same scrambles.
2) a round where groups of competitors are waiting in the separated area and groups take at most 30 minutes: all competitors per group have the same scrambles, groups have different scrambles.
3) other situations: there are different sets of scrambles. Competitors get scrambles from any of the sets. Sets are changed at least every 30 minutes.
In all cases no communication is allowed about the scrambles until the end of the round.

A 3x3x3 scramble shorter than 15 moves or a 2x2x2 scramble shorter than 6 moves shall be taken as lucky.

No, I think we should not do that. And btw. my WR for 2x2 was optimally solved in 8 moves.

Can we also formally allow scrambles to be published 24 hours after the final event?

Scrambles can be made public at the end of the round.
We should not make it a requirement to do it. Again this is a practical opinion.

"When a competitor is called by an official, they must bring the puzzle to the official. If the competitor is currently competing in another event, the official should try to accommodate the competitor and call him at a later time."

I am against this again for a practical reason.
If you are called for, you have to come immediately. Otherwise you are delaying the event.
OF COURSE officials should accommodate competitors if they need to go to the toilet or have parallel events. But I think we should not make this a right of the competitor.

More feedback is welcome.

Thanks,

Ron