Mystery Hunt Warp-up?
We had a typo in our team's hunt email announcing the wrap-up, so I think I'll repeat it here and warp-up a quick summary of what I wrote for EfZ. When I find the actual links to the puzzles, I'll update them here, but I contributed the following as the main constructor (co-constructors if any listed):
The Orbital Nexus Round Concept and Meta - Solution
Harvoid -> Growth Involves Reconstructing Legos
Orbital -> Cross Something-or-Others (with Dan Katz)
Reverse -> The Energetic Man Sporting a Very Long Scarf (with Mike Selinker)
Reverse -> ... and His Alien Barbarian Girl and Her Robot Dog
I also assisted on:
Reverse -> ... and His Investigative Journalist (Dan Katz as main constructor)
Lazyr Zone -> Our Crew Is Replaceable. Your Package Isn't. (Greg Lohman as main constructor)
You'll notice I only contributed to Part 2 of our Hunt and that is sort of a sad side-effect of the specific timing of our construction steps. I was at the WSC in India in April when our meta team began and then finished Round 1 metas, so I arrived back from Goa with emails telling me I completely missed the window here. Determined to write at least one meta this time, I came up with a round concept that fit well for the moon counter amongst our more complicated Round 2 structures. This became the Orbital Nexus, a round where each puzzle had "quarters" that would look like an entire puzzle on the page to a team until a forced refresh put a slightly different puzzle up to give a different, but still thematic, answer. Just as the moons in EfZ had four quarters, the puzzles in this sector had, effectively, four quarters (although the solving time was meant to be about 1.5x over all 4 versions, and not 4x over all 4 versions, with only slight differences at the last stage where possible).
While we agreed the concept was interesting if we could implement it cleanly, I went through a couple different drafts for the meta itself. My first construction turned out to be too close to an existing meta in its mechanism and was thrown out leading us back to the drawing board. I eventually came up with the "planetary system" concept for the moons (lunchboy was helpful in figuring out some of the steps here in a brainstorming session). So, for several weeks, I went about building a good set of numbers for the rotation and revolution of moons and programmed and reprogrammed an applet to make the data collection as friendly as possible (initially, I thought taking screen caps of the round as it occurred might let you get the data, but we decided to be as friendly as possible and gave teams an orrery after half the round plus the 4 quarter concept had been solved).
It took some time to improve the initial set of answers and the rule I discovered is ~3 of the 4 can be very good with a fair bit of work, but getting all 4 to be great was oftentimes impossible. I got 3 US Open winners (tennis and golf) in the sports-based "Warren Moon" but had to settle for a relatively unknown Australian athlete to complete the answer set. I got 3 reasonably common chemicals/elements for the "Neutral Moon" but had to settle for a common medicine name for the 4th. I'm satisfied with the construction and final clue phrase, but I'd be interested to hear what people thought of the whole round and if the "gimmick" was more tedious than interesting.
Anyway, this long story is meant to tell you that 2 months of my hunt constructing time was spent specifically making sure this thing worked. It was our last meta to go final, and at least worried some on the team that it would be the most likely to stymie. (Trivia: it ended up being the first round 2 meta to be solved, and was not one of the "barrier" metas for any team looking to get their last meta or two finished at the end. On that end, I think I met my goal of a meta that 75% or maybe less of the answers perfectly suffices for, which was not always as true for our other tough metas). During those 2 months in which my only priority was Orbital Nexus, the answer words for part 1 were opened and almost all grabbed, so I did not claim any part of Round 1. And that is how I missed the boat on the "easy" part of the Hunt. Out of the country for Metas in 1 and obsessed with moons for puzzles in 1.
Still, when I got to constructing puzzles, I knocked out my two favorite ideas in phenomenal ways (and this is not just my opinion so I feel free to state it this way). Dan and I independently rethought of something we had discussed but not done in 2007 which was to write some "ambiguous to assign" grid-based logic puzzles and this time around, by suggesting a focus on Kakuro, I think we wrote a very solid set of variations with a friendly "assignment" step to get you started. I hoped I met my secondary goal of giving enough kakuro to teams with too many logic puzzle solvers that they actually got to work as 3-4 people on this puzzle and all enjoy a piece of it.
Growth Involves Reconstruction Legos (or Lego Skyscrapers) was another visual concept I thought natually had to be explored and the MIT hunt was a great place to do it. I never liked how Skyscrapers puzzles imply numbers are building heights but, to an actual observer who is not very far away, the clear identification of heights from visual data is not always there. If Clash of the Titans from 2007 was my first attempt at visually redefining "standard" skyscraper puzzles, then Growth Involves Reconstructing Legos was certainly my second, even better attempt. I loved hearing from at least one team that they were trying to actually build the cities from legos to see that they could see similar images at the end. This is probably a great way to get around the visual illusions I left as false clues for teams in some places. The only slight problem with the puzzle is that the answer extraction stymied a couple teams and I hate when this step is the problem when the clear meat and fun of the puzzle is doing the skyscrapers.
My other major independent construction turned out to be the easiest Reverse Dimension puzzle and got teams into the round concept generally. I wanted to write a fun but easy sports puzzle to fill a hole our Hunt had at the time (Cricket and NASCAR, while sports, were not the fun and easy sports puzzle need I had in mind). Mike and I converged on a "Doctor"-perspective to the puzzle with some playful remedites for different sports teams, and I whipped together a somewhat basic color PBN variation that would interface well with the puzzle.
I had intended not to write any sudoku for this hunt (my team shut down most all of my sudoku ideas last time around so I basically, to avoid further frustration with my team, will no longer ever propose any sudoku for an MIT hunt), but I filled the need when I was asked for help for another Reverse puzzle, and I also served a similar role in assisting a constructor on our challenging River Crossing puzzle variation. And that is basically my story of constructing the Hunt.
I'll likely not write about the actual running of Hunt HQ, for various reasons, so don't look for any real details outside of my contributions here. I'll close this entry with links to my two favorite other puzzles that I encountered during my testing of the Hunt; both turn out to be by tigupine. Reflections on a Milky Steed Who's Quite Amphibious Indeed is probably my favorite, with Resolutions a close second.
The Orbital Nexus Round Concept and Meta - Solution
Harvoid -> Growth Involves Reconstructing Legos
Orbital -> Cross Something-or-Others (with Dan Katz)
Reverse -> The Energetic Man Sporting a Very Long Scarf (with Mike Selinker)
Reverse -> ... and His Alien Barbarian Girl and Her Robot Dog
I also assisted on:
Reverse -> ... and His Investigative Journalist (Dan Katz as main constructor)
Lazyr Zone -> Our Crew Is Replaceable. Your Package Isn't. (Greg Lohman as main constructor)
You'll notice I only contributed to Part 2 of our Hunt and that is sort of a sad side-effect of the specific timing of our construction steps. I was at the WSC in India in April when our meta team began and then finished Round 1 metas, so I arrived back from Goa with emails telling me I completely missed the window here. Determined to write at least one meta this time, I came up with a round concept that fit well for the moon counter amongst our more complicated Round 2 structures. This became the Orbital Nexus, a round where each puzzle had "quarters" that would look like an entire puzzle on the page to a team until a forced refresh put a slightly different puzzle up to give a different, but still thematic, answer. Just as the moons in EfZ had four quarters, the puzzles in this sector had, effectively, four quarters (although the solving time was meant to be about 1.5x over all 4 versions, and not 4x over all 4 versions, with only slight differences at the last stage where possible).
While we agreed the concept was interesting if we could implement it cleanly, I went through a couple different drafts for the meta itself. My first construction turned out to be too close to an existing meta in its mechanism and was thrown out leading us back to the drawing board. I eventually came up with the "planetary system" concept for the moons (lunchboy was helpful in figuring out some of the steps here in a brainstorming session). So, for several weeks, I went about building a good set of numbers for the rotation and revolution of moons and programmed and reprogrammed an applet to make the data collection as friendly as possible (initially, I thought taking screen caps of the round as it occurred might let you get the data, but we decided to be as friendly as possible and gave teams an orrery after half the round plus the 4 quarter concept had been solved).
It took some time to improve the initial set of answers and the rule I discovered is ~3 of the 4 can be very good with a fair bit of work, but getting all 4 to be great was oftentimes impossible. I got 3 US Open winners (tennis and golf) in the sports-based "Warren Moon" but had to settle for a relatively unknown Australian athlete to complete the answer set. I got 3 reasonably common chemicals/elements for the "Neutral Moon" but had to settle for a common medicine name for the 4th. I'm satisfied with the construction and final clue phrase, but I'd be interested to hear what people thought of the whole round and if the "gimmick" was more tedious than interesting.
Anyway, this long story is meant to tell you that 2 months of my hunt constructing time was spent specifically making sure this thing worked. It was our last meta to go final, and at least worried some on the team that it would be the most likely to stymie. (Trivia: it ended up being the first round 2 meta to be solved, and was not one of the "barrier" metas for any team looking to get their last meta or two finished at the end. On that end, I think I met my goal of a meta that 75% or maybe less of the answers perfectly suffices for, which was not always as true for our other tough metas). During those 2 months in which my only priority was Orbital Nexus, the answer words for part 1 were opened and almost all grabbed, so I did not claim any part of Round 1. And that is how I missed the boat on the "easy" part of the Hunt. Out of the country for Metas in 1 and obsessed with moons for puzzles in 1.
Still, when I got to constructing puzzles, I knocked out my two favorite ideas in phenomenal ways (and this is not just my opinion so I feel free to state it this way). Dan and I independently rethought of something we had discussed but not done in 2007 which was to write some "ambiguous to assign" grid-based logic puzzles and this time around, by suggesting a focus on Kakuro, I think we wrote a very solid set of variations with a friendly "assignment" step to get you started. I hoped I met my secondary goal of giving enough kakuro to teams with too many logic puzzle solvers that they actually got to work as 3-4 people on this puzzle and all enjoy a piece of it.
Growth Involves Reconstruction Legos (or Lego Skyscrapers) was another visual concept I thought natually had to be explored and the MIT hunt was a great place to do it. I never liked how Skyscrapers puzzles imply numbers are building heights but, to an actual observer who is not very far away, the clear identification of heights from visual data is not always there. If Clash of the Titans from 2007 was my first attempt at visually redefining "standard" skyscraper puzzles, then Growth Involves Reconstructing Legos was certainly my second, even better attempt. I loved hearing from at least one team that they were trying to actually build the cities from legos to see that they could see similar images at the end. This is probably a great way to get around the visual illusions I left as false clues for teams in some places. The only slight problem with the puzzle is that the answer extraction stymied a couple teams and I hate when this step is the problem when the clear meat and fun of the puzzle is doing the skyscrapers.
My other major independent construction turned out to be the easiest Reverse Dimension puzzle and got teams into the round concept generally. I wanted to write a fun but easy sports puzzle to fill a hole our Hunt had at the time (Cricket and NASCAR, while sports, were not the fun and easy sports puzzle need I had in mind). Mike and I converged on a "Doctor"-perspective to the puzzle with some playful remedites for different sports teams, and I whipped together a somewhat basic color PBN variation that would interface well with the puzzle.
I had intended not to write any sudoku for this hunt (my team shut down most all of my sudoku ideas last time around so I basically, to avoid further frustration with my team, will no longer ever propose any sudoku for an MIT hunt), but I filled the need when I was asked for help for another Reverse puzzle, and I also served a similar role in assisting a constructor on our challenging River Crossing puzzle variation. And that is basically my story of constructing the Hunt.
I'll likely not write about the actual running of Hunt HQ, for various reasons, so don't look for any real details outside of my contributions here. I'll close this entry with links to my two favorite other puzzles that I encountered during my testing of the Hunt; both turn out to be by tigupine. Reflections on a Milky Steed Who's Quite Amphibious Indeed is probably my favorite, with Resolutions a close second.