2011 UK Puzzle Championship (25-26 June 2011)

[rob]

How do you deduce this? Why would

Code: Select all

333444
....4.
....2.
....2.

not work?

OK, I skipped a mental step. How could the 2 only achieve 2 neighbours now? (But I think you saw that eventually)

Yes. I still don't see a good argument for deducing that it's

Code: Select all

444433
....3.
....3.
....3.

and not something else, though. Saying that no other combinations work out does seem to boil down to a sort of trial and error.

[PuzzleScot]

Score Corrections

I've finished going through every single answer key that was submitted(!)

A list of answer keys, and points allocated thereto can be downloaded from here, if you want to check for fairness and consistency.

I felt that some leniency should be allowed for some puzzles, as well as the normal recognition of obvious typos.

The way I did it was:
1) I ran a database query that selected all distinct answer keys for each puzzle with a count of how many times each was used.
2) Compared every entered variation with the correct answer, and with the puzzle grid.
3) In 2 cases, people entered the wrong row or column. I awarded 1/2 marks for those two, since it was an obvious accident.
4) For puzzles where excessive counting was required (1,8 & 9), if there was a SINGLE counting error ONLY, a small penalty was incurred [#1: 3 points awarded; #8/#9: 10/15 points awarded]
5) In cases where the puzzle has been solved 'beyond reasonable doubt', and it is clear a single typo has been made (or answer key misinterpreted), I've awarded 2/3 marks (rounded up to next integer). Those cases are clear in the above PDF.

I did this all completely blind. I believe this is completely fair. I have no idea who I affected, so apologies if your score has gone down, though likely upwards. Nor in the end did I take any heed of any appeals. To do so would distort the fairness of the above systematic analysis.

Boy, am I going to make the answer keys more simplistic, and I won't be so forgiving, next time!
Puzzle Cruise is 2-3 July!!! See the main Contests page

The results table looks slightly different from usual - more LMI format. You should be able to click through to see each participants entries.
(red answer keys are too wrong to gain credit, green is correct, orange is partial credit)
Results will automatically appear on the contest page at 3am (just in case I ran late here)

[ronald]

I note earlier discussion suggested that 1 mark would be deducted where the puzzle was solved beyond reasonable doubt.
Now it's the case that all but 2/3 marks have been deducted (a much harsher penalty on the high point-value puzzles?).

This seems wildly different, although I haven't been personally affected. Was there a particular motivation for that change?

Perhaps there should be a forum thread to discuss how to fairly assign penalties to mistakes in future (or, alternatively, to ensure that the answer keys are foolproof).

[PuzzleScot]

I note earlier discussion suggested that 1 mark would be deducted where the puzzle was solved beyond reasonable doubt.

No. I actually said: "... in these cases I will award points, but no more than 1 point less than the total available for that puzzle."
That is exactly what has happened, in the fairest way possible.

[ronald]

"So, rather than half-marks (or none), I've decided my new strategy for these things is to deduct 'some' points from the full tally."

Despite your quote, the spirit of the discussion was that you suggested several times (and I agreed) that deducting a fixed number of points was better than awarding a fixed number of points.

This is exactly *not* what has happened, two days later - in fact you have reversed some previously published decisions - what caused you to change your mind?

Some discussion on best practice is desirable here:

It would be my suggestion that answer keys should be trivial for competitors to input (some discussion on how this would be possible is clearly merited, with respect to the answer keys of this contest)
but that any errors (beyond syntactic) cause a deduction of points: as they do in the USPC.

This level of clarity would avoid any need for a subjective judgement. I don't feel that this was done as blindly as stated, considering that there was open discussion of several answer keys linked to named users.

[PuzzleScot]

what caused you to change your mind?

In a word, fairness.

Both to those who took the time to make sure their answers were correct, and to those who had made a genuine mistake. There is, and never has been a 'declared policy' on this. I was simply stating before that you would not get full marks. I've not experienced having to make judgement calls on incorrect answer keys, so had to make some 'judges decisions' - or is this the first time in human existence this has ever had to happen?

You can see everyone's individually submitted answers, and there's a PDF of answer key variations with explanations. I didn't HAVE to do any of that.
In fact, I am deeply offended that you suggest there was any kind of non-blind favouritism going on!!
(By the way. Entering a wrong answer key row/col got 1/2 marks. The 'solved beyond reasonable doubt with a single typo' got 2/3 marks, all rounded up to the next integer. Not random.) I don't think it's possible to be more open and fair, and I think you are alone if you disagree with that.

You clearly haven't a clue what effort goes into making a high level contest work as smoothly as it did, nor comprehend the 1000s of lines of code, late nights, database configuration, testing, communicating, publicising etc etc, that go on behind the scenes. Thankfully, I've had dozens of congratulatory messages from people who do appreciate these things and how problems were resolved, and a couple of constructive criticisms, which are also appreciated. Of course we all know the answer keys were far from perfect for some puzzles. Some things could have been done better, and will be better for future contests.

[Janix]

Wow Ronald

Don't know where you're coming from but at the end of the day solving puzzles is a purely pleasurable pursuit - fun, hard work, recreation, pain, mental gymnastics, etc. For each contest there can only be one winner, some close runners up, then a plethora of results all the way down - each of whom had their own goal or milestone to achieve, and as the adjustments to certain competitors scores is not of concern of the winners - who loses?

I really don't know how best to say all this as I consider everything below to be important, but first, an event like the UKPC can only be the great success it was if one individual owns it - lives it, breathes it, and makes considerable personal and financial sacrifices for it.

At the UKPA gathering, I was sceptical that the UKPA could achieve a test this year, partly because of the need to create a puzzle set but mainly because one person had to come forward and be THAT person.

So, publicly, as well as privately, not just well done, but blooming marvelously well done, Alan. You achieved the almost impossible. Amazing! What a first birthday present for the UKPA.

Alan also addressed that bit of black and whiteness that given that solving puzzles is meant to be for personal satisfaction, it is incredibly frustrating to spend an amount of time on one (or more) puzzles, then lose all the credit for a slip up that can be proven to be trivial.

Yes - in the 'Grand Slam' events, maybe tighter rules can be applied, and one year soon, maybe the UKPC will be elevated to that category (I know that this 'sport' is in its infancy, however, the explosion of on line events this year has been nothing short of [please add your own] superlative-ous.

Given that there aren't standard answer keys (yet) and there are thousands of puzzle variants yet to be created, we are living in a dynamic environment where familiarity can be a help as well as a hindrance.

Alan has shared the data and started a debate that can only be to the good. Openness is key and the user community is the best placed to reach consensus on such a disparate area. There may never be a 'perfect' solution - thank goodness - as healthy debate keeps one focused.

And coming back to the puzzles. A perfect set producing the perfect range of scores over a perfect sized field (for year one, that is).

I have competed in the Mental Calculation World Cup (MCWC) on the last two occasions and the great success of that is, likewise, down to one man - Ralf Laue. He is looking for bids to hold the 2012 (biannual championship), having organised the first four (who will step up to his plate?). Currently, there is only one Ralf Laue, and there is only one Alan O'Donnell. Many competitions rely on the drive and energy of one person, not forgetting, of course, those individuals that support the figurehead with all the other important and essential bits.

However, if you go to the Mental Calculation forum, you will see bickering over scoring, the nature of the contest included, bicker, bicker, bicker over trivialities, that are nothing in comparison to having had the opportunity to take part in the first place. Yes, some competitors (and one who didn't even turn up - twice!) complained publicly about the organiser in that immediate post competition period when energy is spent and emotions are high. This non competitor was extremely insulting about another competitor (completely unwarranted, incidentally) in a way that makes me partly ashamed to be a part of that community - and this went on long after the most recent competition ended. I really don't want any of that to happen here.

The corollary of this is the Mind Sports Olympiad (MSO) which started off 15 years ago with great success, but has lacked ownership in recent years (although recently revived under a new 'owner'), and is a shadow of it's former self. I joined a 'committee' to help breathe life into it, but given that this set of face to face tournaments will always be expensive to run, all the goodwill in the world doesn't mean a thing without someone with the passion and who has deep pockets.

Sadly, the MCWC non competitor believes that they 'own' a comparative event at the MSO, and has also managed to sully the name of this organisation in the process. The 'sports of the mind' world is too small for such clashes (not that they would be justified if it was bigger) and I hope that people can only write things with a positive emphasis in the future.

I have either set up or run Chess and Bridge clubs from when a was a teenager (do you know how many chess teams you can squeeze into a Beetle?!) amongst other 'sports of the mind' things, and have done that because I wanted to - no more no less. I got pleasure from the pleasure it gave to others as well as the pleasure it gave to myself.

I hope that Alan has by now had enough sleep recovery that he will begun to have felt some pleasure from his achievement, and that having to respond to emails when 'punchdrunk' is not conducive to maintaining composure - well done Alan for your considered response under the circumstances.

As I knew both parties in the example I gave above, I too got some brickbats for sticking my oar in, but I would, and now have, done it again when the situations arises.

Looking forward to the next tournament.

[ronald]

I am somewhat surprised you express such deep offence. It's my goal to improve things for the future and I made no personal comment (I didn't suggest there was favouritism here, only that the process cannot be described as blind if you already know who has made which submission): the fact that you're taking offence means that I clearly haven't expressed myself well. Sorry.

I think a good job was done overall through some amazing voluntary work.

I had wanted to reflect openly on what has happened in an attempt to improve future experiences. I also wanted to help improve the process to avoid the need for such a large amount of the work being dependent on one individual's kind and hard efforts - but clearly I'm going about this the wrong way here. Thanks for the responses.

[david mcneill]

Wow Ronald.

I love your latest message.