The 2013 edition of my Valentine’s Day puzzle contest has ended.  You can go to this very basic page to access the puzzles and, perhaps more importantly, the answers.  (I have made a few comments along the way in the answers as well, so you may want to read them anyway.)

The important business: giving out money!  Fifty-two people sent me a correct answer, and I had random.org shuffle the names around.  The $50 went to Yossi Fendel, who took a gift certificate to ThinkGeek, and $25 to Steve Williams, who took a gift certificate to Amazon.  (These will be filled in with whatever info the winners are willing to share.)  (If you’re curious as to where you might have landed, you can email me and I’ll look you up.)  Also I just wrote a check (well, not actually, but that sounds better than “Paypalled”) for $955 (and a little bit of change) to Transitions.  Adding in the UK donations to Refuge, we are somewhere north of $1000 in money raised.  So THANK YOU very much — such generosity coming in is really humbling.  (And those of you who know me know how hard that is.)

Please use the comment section for questions or remarks or grousing — we’ll consider that the spoiler zone.  NOTE: I don’t get a lot of spam, but it’s still about as much as real comments, so your comment may have to get approved before posting.  I’ll be trying to keep up as much as I can.

There have been some new puzzlehunts, and a couple that are explicitly designed to bring new people in, which is excellent.  So here we go:

1. Roy Leban’s Kickstarter puzzle hunt, which I mean in more ways than one: not only is it a kickstarter funding for a puzzle-hunt project, but the pitch page itself is also a puzzle hunt, so you can do one right there to see what’s up.  So: no excuses for not at least doing the one on the pitch page.  The pitch states it is designed to be accessible to all, for the purpose of bringing in fresh blood to the scene, so you should be up for it.  Roy Leban also has a puzzle company Puzzazz, which specializes in puzzle books for iThings, but since I am not an iFanBoi I don’t necessarily pay a lot of attention.

2. Trip Payne’s Extravaganza II, also available through Kickstarter.  Given that it’s Trip, we’re looking at all word puzzles, probably many crosswords and cryptics; the meta- elements will probably be explicit and/or easy, although the puzzles themselves (I’m betting) will have a wider gamut of difficulty.  You can also get some extra variety crosswords if you chip in more money.

3. I intend to do another charity puzzle contest for Valentine’s Day this year, same as last.  I’m about one-fifth of the way done with making it, which means I’ll need to step it up for the rest of the break here.

4. Not designed for beginners, I don’t think: One of the Manic Sages has released a puzzle-hunt-training puzzle hint that was written to prepare their team for last year’s MIT.  You can find it here, and for that matter if you want to join my “team” you can use the team ID “tabstop” when you register.

For live puzzle events, there’s MIT Mystery Hunt of course, although that’s certainly not for beginners. DASH may well be coming to a city near you in May. Maybe this time I’ll be there for that.

If you are just looking for plain old crosswords, there is of course the ACPT, as well as a couple of subscription venues: the don’t-call-it-the-New-York-Sun Fireball Crosswords, and the don’t-call-it-the-Onion-AV-Club American Values Crosswords.

Remind me of what I left out in the comments.

Now that this tournament is underway, I thought I’d give some rationales for some decisions, say what I’m hoping to learn out of this tournament, and open up a place for comments about the above.  So, here goes:

Pools

The pools were not randomly drawn, but were seeded: the top four players (according to Pro Rank) were #1 seeds, the next four were #2 seeds, et cetera.  (I used the Pro Rank as of when people joined, which may not be exactly the same as it is now, but I would imagine the seeding itself would not have changed much.)  I don’t really have any information about people’s Instant performance (other than maybe their high score), so Pro Rank was the best available substitute.  I think this will give a bit more evenness to the pools; there will probably still be a Pool of Death due to someone being rather better at Instant than they are at Regular, but I minimized that as much as I could.

Qualification and Tiebreaks

There are two schools of thought about prelims->playoffs: (1) most or all of the players qualify for a knockout playoffs, and (2) a smaller portion of the field is in contention for the trophy, with (often) another round-robin (or similar) format.  The question then becomes “what are the prelims good for?”  In case 1, the prelims provide seeding for the knockout phase, so a good performance in the prelims can give you byes and/or easier opponents.  In case 2, the prelims are more important, as you need to do well just to survive (and often any advantage from the preliminary phase is either eliminated or reduced to one game).  Perhaps a knockout would have been more in keeping with the “quick” nature of Instant, but I strongly prefer round-robin to knockout (e.g. everybody gets their chance to knock the favorite off their perch, you can avenge your loss to player B by beating player C who had beaten player B to level the records…) so I went with option 2.  In terms of how many qualify, to me about 1/4 to 1/3 feels “right” (in this case I did 6/20).  I will be looking at how closely bunched the wild card race turns out to be.

I’m using total raw score, instead of plain total points, as the tiebreaker.  I’d like to reward, at least in some small way, those players whose good 7s get beaten by great 8s.  However, since the main criteria is still “winning the game”, I feel that aggressiveness will still have its place.  (If this were Regular, I’d have to worry about the numbers game, but that’s not a problem in Instant.)  I don’t get to see the words people don’t declare when I’m spectating a game, so I can’t tell how often people chicken out of good words; I’ll leave this up to you players in the comments whether it changed your playing style (or would have).

The other caveat I worry about with both total points and raw score is the variation in points available.  I have an idea of a (rather more complicated) statistic to normalize that variation; I’ll be checking in this tournament how closely those results track.  If I get the same ranking as TRP, then I probably won’t bother with it, since it’s extra complication for no benefit.  If the ranking changes, though, then I’ll consider it for future use.

Software

For this tournament, I am creating essentially no web pages by hand; I am entering the games into the database as they happen (or at least as I find out about them) and all the cross-tables and standings and the like are generated on-demand when people visit the page.  I did some testing locally, and for the most part everything that I’ve put in has been working as expected (so far, knock on wood).  I’ve also added a few features as I’ve gone (links from one crosstab page to another, and from the overall results to individual pools).  Please, if you have an idea for something that would make this more useful from your perspective, put it in the comments and I’ll see if it can happen.  I don’t know whether the world needs a tournament administration suite in PHP, but it’s going to get one by the time we’re done.

I think v1 of my tournament-running software is good to go.  I’ve tested it on my local setup—we’ll find out how well it works when the real tournament starts on the 21st.  We should have:

• web-form additions of people and game scores
• automatically updating cross-tables in the round robins
• automatically updating “playoff picture” page showing who can still make the next round

What we don’t have but probably need:

• an automatic “generate the playoffs” button [waiting to see how my scheme for handling playoffs in the database works when doing it by hand]
• Making things pretty, or at least mediocre.

All will become clear over the next two weeks.

The blog has moved to the new website (which, hey, you’re here, so you already knew that).  But update bookmarks appropriately:  http://blog.tabstop.us is where all new posts will be.

I’d tell you what it is, but I’m not sure what it’s called.  It’s very similar to backsolving, and somewhat similar to guessing, and probably called “figuring out the answer after having done only half the puzzle.”

For example, in the most recent issue of P&A Magazine, this is what I actually have for the puzzles (I think there’s one or two sort-of spoilers here depending on if you’ve gotten the trick, so be warned):

• Imperialism: 12/14 rows
• Redistribution of Wealth: actually complete
• Like Making Sausage: 33/48 total bubbles/boxes; 5/8 courses
• King of the Jungle: actually complete
• Country Is King: 9/12 identifications
• Let Them Eat Gateau: actually complete
• The Royal Touch: 5 letters
• United We Stand: 20/25 circles (I think)
• Class President: 6/7 quotations
• Conspiracy: actually complete
• The Emperor’s New Clothes: 6/7 clues
• Redaction: crossword complete; answer backsolved from meta

I can’t really discuss the solving process for artifact five, as I did not solve the artifact so much as endure it, counting the days until the hints, and then answers, were sent to me (except, of course, for the one answer that was deliberately not put in the answer document–even in the lab writeup that you only get after submitting all the answers, they blacked out one of the answers, preventing you from finding it out unless you know the arcane “highlight text in a document and copy it”).

So I thought instead I’d talk about the event as a whole.  (This notion was probably aided by the recent post here and here about trust and story in puzzle events.)  This was certainly an ambitious project; five “real-world” artifacts with puzzles embedded in them.  The claim of round-the-clock GC interaction, even if aided by automatic correct answer recognition and immediately gainsaid by telling you when you wouldn’t get a response and when that response would take a while, required a great deal of manpower to maintain.  I have no idea how many people were behind this event, either in terms of writing puzzles or manning the e-mail queue or wrapping up mugs or anything else; I can say they made a good run of it.  If my e-mail filtering is working properly, I sent off in the vicinity of 200 responses (some of which made very little sense outside of my own head) and they sent back the correct response nearly 98% of the time.  That sentence looks very snarky, but actually I think it’s very very good, especially since the responses that are more important to us solvers (confirmation of methods, intermediate results, nearly but not right) are the ones that can’t be automated.  (I know this is the thing that gave me the most indigestion while running the Valentine’s Day contest: “What if someone sends me something while I’m asleep!  They won’t get an answer for a long time!  Oh no!”  It never made a difference, but I wish I had had the time to get a decent answer checker thing hosted somewhere.  I should probably pester Foggy to find out what he uses for P&A.)

I got the sense at the time, I’m not sure how, that the event sort of tripped over the finish line, went splat, and was content to lay face-down without the will to get up and acknowledge the crowd.  Looking back, I don’t think there was very much going on during the last puzzle to give me that impression, just that I found the last artifact’s puzzles to not make a great deal of sense.  However, now that it has been over a month since I “finished” (and it took me a month to do so as well) with no real expression of “hey this is over now” (such as congratulating winners, or coming out with the promised solution manual, or anything else), the sense is getting stronger.  I suppose people may still be working on things, but I fear the top ten is probably set.

What about the puzzles?  This is the part where I welcome the wisdom of my readers (both of you), but in my opinion the puzzles started out well done but tapered off towards the end.  I didn’t have any real complaints about anything on the first three; the paperback was marred (IMHO) by using OCR text; yes it is easily available, but deliberately choosing “filler” that has more (both less and more subtle) “anomalies” than the “anomalous artifact” we are supposed to be examining can only lead to heartbreak.  And again IMHO, the puzzles on the map started out well, but they all seemed to take a left turn at Albuquerque (difficult to spot, since it wasn’t on the map) that I didn’t necessarily see any reason to take (this is probably just me not being on the designers’ wavelength).  The thing that got on my nerves the most, though, was the “story”.  During all the interaction I had with the BLG team, they stayed in character (and believe me I know how hard that is).  There are only two problems with that: the puzzles didn’t see any need to “stay in character”, unless you think that a puzzleartifact created in 1930 would reference cities (and worse yet, a feature-answer fictional character) that didn’t even exist yet (you could argue that the library card created in 1943 replaced some previous physical accompaniment), a book created in 1943 would have a web address on the back, as well as postcodes that didn’t exist at the time either (you could, I suppose, argue that this is a reprint; after all, this is not an organization that would feel obliged much to keep fidelity I suppose, explaining why a book printed in 1943 contains a chapter of Jurassic Park).  At the time I spotted an anachronism in the receipt puzzle, but I’ve forgotten what it was (maybe something about when the Whitney acquired the painting in question?).   The other problem was that the story greatly restricted what the team could do and/or say; when I complained (okay, whined) about the OCR text in the paperback, they were pretty much stuck with the (paraphrased) response of “We had to solve them ourselves when we intercepted them from the Seventeen, so we know just how badly they were constructed.”  Perhaps they saw the point, perhaps they didn’t; I’ll never know.

I’ve never had to deal with running a puzzle event where I was relying on other people to make the objects (I’m pretty hands-on with my PDFs), so I cannot complain about typos cropping up during the process, as I suspect that’s a guarantee.  I would have preferred an errata list, rather than the errors either being ignored or being mentioned in the automated response to the correct answer to the puzzle they were in.  Congratulations to BLG for owning up to the castle errata, even if only via Facebook.

All in all, with the bit of perspective that a month can bring, I think the whole thing was well done, overall.  There were a lot of puzzles, with only a few clunkers; the organization was mostly well-done and I know what to avoid, in terms of dealing with the organization, if there is a future.  I would likely do it again; I like to think that the next iteration would be smoother, but of course I have no way of knowing if the second iteration would even have the same people in it, let alone whether they’d learned anything.  Still, one has to have faith in something in this world.

I did a lot of working in the book itself (although it did take a while to get over my internal taboo of writing in a book), so this time we’ll go through my submissions to BLL.

back ten mins: Why not start at the cover?  I had to guess whether it was type on the given keyboard and see what would happen on the normal one, or the other way around; and of course I guessed wrong at first.

very loud annoying bohemian: BLL do love their acrostics.

Alice in Wonderland: Kudos to the writers here; when I first read that very first page, it hung together as a single excerpt.  I didn’t recognize any of the first sentences, although the “misgiving” line did tickle a very faint memory.  Only when looking up that paragraph and realizing the sentences came from two different works did the basic idea click into place.  I didn’t actually look most of them up–the first five gave me “Alice”, I counted that there were the right number of lines to get “in Wonderland” and ran with it.  (Only much later did I find yet another acrostic in the short story titles sending me here.)

Balto: I got here by combining two clues I (apparently) wasn’t supposed to combine and quite frankly in retrospect I’m not sure why I combined them: the pangrammatic lipogram clue (I knew what “pangrammatic” went, but not how it combined with “lipogram”; I knew of Avoid and the idea but not the name) and the one-page story “clue”, both from the author bio.  In any event, I went to that story, recognized that it was written in nearly-pangrammatic paragraphs, and obtained “Balto”.  I didn’t recognize the name, but Wikipedia did, and so I submitted.  I got the first “rejection letter” for my troubles, which is when I wrote “BASTARDS” in my solving notes (those of you following along on Twitter may remember that).  I recognized it as a parody of a rejection letter, and had at this point already noticed the URL under the UPC code, and assumed (I don’t know why) that I was being pointed to that.  I had no idea how to relate a ship to Balto, though, so I put it aside.

Montgomery Burns: Worked out LEPTON WARP first, then GLOW of DEATH.  That was enough to figure out who we were talking about there.

hint alpha: Eventually I succumbed to temptation.  Oh, you crazy people and your library cards.

Harry Shearer: An attempt (a bit meta) to apply the “whodunit” clue to Montgomery Burns.  At least they were ready for it.

More than pulp fiction It invades your cranium And can’t be un-read: I went back (just for fun) to the Amazon page for the book, and noticed the haiku.  I had not noticed it previously (not sure whether it was added later, or I’m just inattentive) so I thought I’d try it out anyway.  I got a rejection letter for it.  (I went back to Amazon again just now; someone has created a store “Overpriced Used Books Inc” and is offering a copy for sale at only $1499.99 instead of the list price of $1700.  Also, I was hoping that the review comments would become a bit more like a sign-in in geocaching or whatever; but only a few of us left a comment.  Shame.)

Matt Groening:  Still a bit meta, but at least I recognized that the name was anagrammed on the page.

Maggie Simpson: Now we’ve got it.

BALTO error: Somewhere around this time, the BLL people noticed the BASTARDS tweet and sent me a message as to what was up with that, and I explained the reasoning behind the submission.  At this point, they decided that that was actually an answer to a puzzle after all, so I got a replacement message that BALTO was on track for where.

binary->hexadecimal: My attempt to circumvent the “only two words” restriction on method submissions.  It appears to have worked, as I got the “interesting” email.

Red Olson Trail Monument: A very bizarre answer, it looks like.  I had successfully converted the true/false in the author bio to binary, then to ASCII, and gotten … DOG.  I currently had a dog answer (see “Balto”, above) that was supposed to lead to where, and so this happened.

Royal Observatory: This was a combination of looking up the postcode given for Greenwich Press, plus the idea of GMT (clued by the title).

HINT one came at this point.

very loud annoying bohemian dog: I was completely, utterly, and totally flabbergasted to earn points for this answer.  I was still stuck on what to do with “dog”, and since the other puzzle on that page had also given an incomplete-looking answer I stuck them together.  I didn’t really know of any bohemian breeds off the top of my head, but whatever.

solfege: An attempt to do something involving notes.  I had noticed a few solfege notes in the titles, and one of the review hints had mentioned notes, so I tried this.  It failed.

Indian Q Hunter: I had finally worked out more of the “enumerations” — the only ones I knew off the top of my head where prime, abundant, and deficient (I knew of Ulam but not that he had numbers).  I found some references for more of the numbers and got the pangrammatic business sorted out.  However, one of them I failed at–I missed the Q in a paragraph and got the answer above instead of the desired answer.  They didn’t recognize it at the other end.

Cleopatra’s Needle: I had looked up my answers from the pangrams and found the common location, but initially misread what I was looking at.

Central Park:  I quickly figured out what I actually had though, and corrected myself.

Alice in Wonderland statue: And now we combine with one of the much earlier answers to earn some points.

Braille double bar: I had looked at the tic-tac-toe grids, and noticed that most of them would be valid Braille characters.  So I looked them up and got SHARPSHARPSHARP.  One of them, however, was oversize; looking it up (and especially looking up musical symbols, since we’re dealing with “notes” + “sharp” + #) I found it was the symbol used to represent a double bar in Braille.  This submission got a rejection letter.

Science and fiction Together will be the end We see what’s coming: Finally noticed that all the capitalization gave clues.  The giveaway, of course, was “A FEMININE PRONOUNcement”.  Took a while for me to look at that page, I guess.

missing punctuation: I had spent a lunch hour (which naturally is not a whole hour) circling missing/extra quotation marks and commas and the like, particularly in “The Happy Ending”.  That was a lunch wasted.  (Well, I did eat a sandwich.)  As a note I wrote in afterwards, saying something resembling “you appear to have picked some stories that didn’t OCR well.  That was a bother to work through.”  The response was something resembling “Don’t blame us–that was what the Seventeen created; we spent a lot of time on it too.”  I’m happy they’re staying in character, but I was rather annoyed by the response anyway.

ROYAL OBSERVATORY error: At this point, BLL decided that Royal Observatory wasn’t complete nonsense, so I was told that its location might be useful.

23:50:  I had “back ten mins” still.  I thought that zero GMT might well refer to midnight, and took ten minutes off it to get the answer.  This is when I discovered that wrong time submissions got a different e-mail than general wrong answers.

prime meridian: Just to sort of confirm the GMT + Royal Observatory “response”, except it got a rejection e-mail.

04:30: I had a bright idea that “notes” might refer to currency notes, so I used the price on the back cover as a time.  That got a slightly insulting e-mail back, but at least it was hand-done and not automated!  (As a bonus, that e-mail had “SOrry” in it, which led me to go around again about solfege for a while.)

10:55: I don’t remember any more where that came from.

01:30: Finally realized to use the numbers on the keyboard being pointed to.  (Of course the keyboard goes up to 12!  Wait….)  This still didn’t earn me any points until I submitted…

1:30 GMT: Because every other attempt to use GMT failed.  (I have no idea whether 13:30 would have been accepted.)

forward ten: I finally noticed that the ISBN on the back didn’t match the ISBN listed for the book; I looked that up, then downloaded a barcode-reader app for my phone and read the barcode which was for another different book.  This got an “are you sure you have that in the right order” response, so…

ten forward: I submitted this and got it right, so then I had to figure out what it meant.  Apparently this is something in the Star Trek universe.  This is right above the URL with a “ship” in it, so I replaced “ship” with “enterprise” and got a picture of …

pinky ball: Whatever that is.

THREE hints came and went with me not doing anything.  (I know for a while I was in St. Louis.)  So I didn’t actually have to solve the back cover puzzle since eventually the answer came in a hint.  The rest of this is going to be a bit unfortunate, but here goes:

Musical staff: Eventually I had it beat into my head that each of the marginal notes represented a note, somehow.  I was originally going to try to write it down on a staff, but: no.

Bass clef: I was trying to make the initial sharps line up on a key signature and I thought I had a way to do that on the bass clef.  But: no.  However, there was a hint to look on the copyright page.  I finally sounded out the other author there the way I was supposed to and got

Piano keyboard:  Which is where I was supposed to be.

Key signature: Still trying to interpret the sharps as a key signature.  Hint: no.

cfabaggbd#fecccafcbccccebcd#cbbabeg#bbabef: Yes I really submitted that.  I was a bit desperate.  The Lab asked for clarification, which I gave, and “admired my tenacity”, and then for whatever reason tried to hint me back to “piano keyboard” (despite my having just told them in the clarification that I was using a piano keyboard).  I think I still have somewhere on my computer (maybe I’ll link to it if I can find a way how) the sound file that I generated from this.

d#g#b: This submission elicited the information that the Lab would not respond to submissions of musical notes (apparently they learned from the last one).  But also that lining up the book with a keyboard was a good idea.

Westminster chimes: This was seeing some sort of a pattern in the chord structure at the beginning.  It isn’t all that close to the Westminster chime progression, but I was feeling the answer e-mail breathing down my neck….

August 16: I submitted this, based on number of symbols between # signs.  Obviously it had nothing to do with the answer, but I wanted to at least submit something before I was told that the answer was…

November 22, 1943: which email went to my spam folder somehow despite my having set up a filter in Gmail.  (The reason cfa etc wasn’t right?  I was holding the book “upside down” and had the higher notes on the top instead of on the bottom.  Or maybe it’s the other way around, I’ve forgotten.  No I somehow did not recognize an “upside-down” version of Happy Birthday To You, even with the right rhythm.)

So: not a very “clean” experience, for me anyway; it got to the point where you couldn’t trust “no”–if it was just method-check I could live with that (since verifying a path still feels like a “bonus” to me), but somehow not knowing the answers to the puzzles left a funny taste in my mouth.  I heard rumors that one of the automated e-mails gave an answer to another puzzle, but either it was edited before I got it, or I never submitted the answer, or something.  I imagine that the BLL team is listening to criticism (or, in my case, whining), but by their own conceit (that they didn’t write the puzzles) aren’t allowed to acknowledge any of it, which also leaves a funny taste.

Well, really a live “blog” anyway, as I have (almost) all the notes I took as I solved the puzzle here next to me as I write this.  (The first page I kept online, and then it got eaten in a surprise reboot, but I have recreated most of it.)  Inspiration, tenacity, or sheer stubbornness?  Read on.  (Note: this took a week in real time, and I haven’t really distinguished things like days changing much.)

First Impressions

Now Is The Time We Work

I Need a Hint

Another Day

Finish It Off

Hey, What About….