osmie: (Default)
Last night I dreamed I had found some notes, dated 1980, toward the creation of a UBC Mystery Hunt to rival that of MIT. They were in my cousin's apartment, where for reasons obscure I was tiptoeing about and disconnecting electrical devices at night; and do note how conveniently I'm avoiding the questions of why my cousin had a UBC document from a year she was still in high school, of why I was in my cousin's apartment in the first place, and for that matter of what my cousin was doing living in an apartment.

I didn't recognize the names on the documents, which had been printed in dark purple ink on aging white office paper, about 8"x10.5", using a ditto machine. One of them might have been Donald something. Both were male.

The hunt was themed around special relativity. Some of the puzzles had a grid which was too small, a timeline that was too short, or (in some fashion the authors did not describe in their brief) not enough mass. Others had extra time, space or mass. You could make one puzzle bigger or smaller at the expense of another by calculating your own velocity relative to each puzzle. Since each puzzle came with a set of coordinates, this process would eventually lead to a spacetime diagram of lots of solutions flying around the cosmos at high relative velocities, which in turn would lead to the final set of puzzles.

The authors also provided two sample puzzles, in which one had to transfer length from one set of answers (which was clearly departing at relativistic speed and thus redshifted, so that its answers appeared longer than they actually were) to another (which was clearly arriving at relativistic speed and thus blueshifted, with the reverse effect). I don't recall anything about the puzzles themselves.

It sounded like a really neat idea for a ridiculously complex scheme of puzzles, but the major problem is that only university-level math & physics geeks would enjoy it. That's all well and good for a demesne like MIT, but I haven't even been a UBC student for many years. I can't think where to market it.
osmie: (Default)
Last night I dreamed I had found some notes, dated 1980, toward the creation of a UBC Mystery Hunt to rival that of MIT. They were in my cousin's apartment, where for reasons obscure I was tiptoeing about and disconnecting electrical devices at night; and do note how conveniently I'm avoiding the questions of why my cousin had a UBC document from a year she was still in high school, of why I was in my cousin's apartment in the first place, and for that matter of what my cousin was doing living in an apartment.

I didn't recognize the names on the documents, which had been printed in dark purple ink on aging white office paper, about 8"x10.5", using a ditto machine. One of them might have been Donald something. Both were male.

The hunt was themed around special relativity. Some of the puzzles had a grid which was too small, a timeline that was too short, or (in some fashion the authors did not describe in their brief) not enough mass. Others had extra time, space or mass. You could make one puzzle bigger or smaller at the expense of another by calculating your own velocity relative to each puzzle. Since each puzzle came with a set of coordinates, this process would eventually lead to a spacetime diagram of lots of solutions flying around the cosmos at high relative velocities, which in turn would lead to the final set of puzzles.

The authors also provided two sample puzzles, in which one had to transfer length from one set of answers (which was clearly departing at relativistic speed and thus redshifted, so that its answers appeared longer than they actually were) to another (which was clearly arriving at relativistic speed and thus blueshifted, with the reverse effect). I don't recall anything about the puzzles themselves.

It sounded like a really neat idea for a ridiculously complex scheme of puzzles, but the major problem is that only university-level math & physics geeks would enjoy it. That's all well and good for a demesne like MIT, but I haven't even been a UBC student for many years. I can't think where to market it.
osmie: (Default)
The short version: with 143 points, it would take a miracle for me to place among the top few Canadians. I made mistakes on half a dozen puzzles and had to start over. I left four or five puzzles unfinished -- when I could have spent the same amount of time finishing one or two for points. I failed to grab pages #1 and #2 out of my printer, losing me the chance to grab some quick points on the easiest puzzles. I forgot to submit my answer to one question, losing ten more points which I actually earned.

I have already booked vacation for this fall, and I will be travelling to Montréal in September, not Vilnius in October. So even if I do qualify, there's no tension over going to the Worlds or not: no is a real decision this time around.

More details under the cut )

When you add it all up, I spent more than an hour pursuing false trails: puzzles I later abandoned, or on which I turned out to have made an early mistake, or both. This left under 90 minutes to achieve the score I did, which is extremely respectable. But I'd be a lot happier if I'd achieved it in the first 90 minutes of the test, instead of the last.

Humbug. It's sure a good thing I do these puzzles for fun, or I might be permanently disgruntled.
osmie: (Default)
The short version: with 143 points, it would take a miracle for me to place among the top few Canadians. I made mistakes on half a dozen puzzles and had to start over. I left four or five puzzles unfinished -- when I could have spent the same amount of time finishing one or two for points. I failed to grab pages #1 and #2 out of my printer, losing me the chance to grab some quick points on the easiest puzzles. I forgot to submit my answer to one question, losing ten more points which I actually earned.

I have already booked vacation for this fall, and I will be travelling to Montréal in September, not Vilnius in October. So even if I do qualify, there's no tension over going to the Worlds or not: no is a real decision this time around.

More details under the cut )

When you add it all up, I spent more than an hour pursuing false trails: puzzles I later abandoned, or on which I turned out to have made an early mistake, or both. This left under 90 minutes to achieve the score I did, which is extremely respectable. But I'd be a lot happier if I'd achieved it in the first 90 minutes of the test, instead of the last.

Humbug. It's sure a good thing I do these puzzles for fun, or I might be permanently disgruntled.
osmie: (Default)
Well, I've spent an awful lot of time wishing in the last couple of weeks, but the bare truth of it is that I will not be attending this year's World Puzzle Championships in Rio de Janeiro.

The response to my appeal was far higher than I expected, and I'd like to thank everyone for your offers of support.

But...

When I sat down to work out the details of buying my plane ticket, I had to face the facts:
  • My savings account has been drained by this spring's ongoing financial crisis. I can't afford to buy a plane ticket outright until around September.
  • I haven't quite managed to finish paying off my credit card debt, and I'm not about to add another $1200 to it now. I'll be able to afford a plane ticket around the beginning of September.
  • Until I'm clear of revolving debt, I do not want to open another line of credit. Many thanks to the two of you (yes, two) who offered interest-free cash flow loans, but I know how my brain stresses out in direct proportion to the number of loans I have on the go, and I cannot afford to say yes before, well, the end of August.
  • So far, I've been offered more than $500 in outright donations. This is incredibly generous, and I'm very grateful to everyone who responded. However, this number includes contributions from almost all of my closest friends. It's reasonable to assume that I'll get a lower average donation from acquaintances, and still lower from complete strangers, for a roughly equal total among each of the three groups: meaning that if I spend all my time fundraising this summer, I might just barely be able to cover my costs.
  • If I'm lucky.
  • And if I have no other financial crises.
  • I just don't like the way it feels to make these last two assumptions.

[livejournal.com profile] aceofbirds remarked the other week that it's too bad I don't have a demo CD out yet ... which is true on so many levels, but especially the one which challenges me to define my priorities. And you know what? Attending the World Puzzle Championships, in this or any year, just isn't as important to me as recording my first CD. If I'm going to set a goal for this summer's spare time, I'd rather it be creative than competitive, and (though it pains me to choose between them) I'd rather it be musical than puzzlical.

So I'm not going.

But you're all invited to my CD release party. How's Thanksgiving weekend for you?
osmie: (Default)
Well, I've spent an awful lot of time wishing in the last couple of weeks, but the bare truth of it is that I will not be attending this year's World Puzzle Championships in Rio de Janeiro.

The response to my appeal was far higher than I expected, and I'd like to thank everyone for your offers of support.

But...

When I sat down to work out the details of buying my plane ticket, I had to face the facts:
  • My savings account has been drained by this spring's ongoing financial crisis. I can't afford to buy a plane ticket outright until around September.
  • I haven't quite managed to finish paying off my credit card debt, and I'm not about to add another $1200 to it now. I'll be able to afford a plane ticket around the beginning of September.
  • Until I'm clear of revolving debt, I do not want to open another line of credit. Many thanks to the two of you (yes, two) who offered interest-free cash flow loans, but I know how my brain stresses out in direct proportion to the number of loans I have on the go, and I cannot afford to say yes before, well, the end of August.
  • So far, I've been offered more than $500 in outright donations. This is incredibly generous, and I'm very grateful to everyone who responded. However, this number includes contributions from almost all of my closest friends. It's reasonable to assume that I'll get a lower average donation from acquaintances, and still lower from complete strangers, for a roughly equal total among each of the three groups: meaning that if I spend all my time fundraising this summer, I might just barely be able to cover my costs.
  • If I'm lucky.
  • And if I have no other financial crises.
  • I just don't like the way it feels to make these last two assumptions.

[livejournal.com profile] aceofbirds remarked the other week that it's too bad I don't have a demo CD out yet ... which is true on so many levels, but especially the one which challenges me to define my priorities. And you know what? Attending the World Puzzle Championships, in this or any year, just isn't as important to me as recording my first CD. If I'm going to set a goal for this summer's spare time, I'd rather it be creative than competitive, and (though it pains me to choose between them) I'd rather it be musical than puzzlical.

So I'm not going.

But you're all invited to my CD release party. How's Thanksgiving weekend for you?
osmie: (Default)

So this year I placed second among Canadians on the Google US Puzzle Championship, earning me the right to fly to Rio de Janeiro for the 16th World Puzzle Championship.

This is expensive. Many teams around the world have sponsorship—for instance, Google gives the Americans an all-expenses paid vacation—but not in Canada.

My budget would look something like:
  • Entry fees, accommodation, meals ⇒ €400 ≈ $587.60
  • Brazilian Visa ⇒ $90
  • Spending money ⇒ $400
  • Plane fareeither $1179.94 or (7300 - N) air miles + ($245 + $0.30N)
  • EDIT: Carbon offset for plane emissions ⇒ €154 ≈ $226.23
  • EDIT: Travel insurance ⇒ $105
The total is either $2588.77 or ($1653.83 + 4184 air miles + any proportional combination of 3116 air miles and $934.94).

I can afford about $1000.1

I have approximately one week in which to let the team organizers know whether or not I'm coming.



If you would like to help me get to Rio, please let me know as soon as possible if you're able to contribute any amount of money, or any number of air miles, to my cause. Please do not actually send any money or air miles yet! Based on people's responses to this preliminary plea, I'll decide within a week whether the fundraising drive is actually on. You'll hear.

Note that if 16 people all contribute $100, I'm there…but if 160 people all volunteer $10, that's just as good!

I am screening comments to this post by default, but I will unscreen any comment which includes the word "penguin."

Thanks.

EDIT: How to donate air miles )
EDIT #2 )
EDIT #3: Carbon offsets )

1Although I've been saving up for this event for most of a year, we've suffered some financial setbacks in the past month (details available if I know you personally), which mean that I can afford only about $1000.  Nothing terribly dramatic, just expensive and unexpected.
osmie: (Default)

So this year I placed second among Canadians on the Google US Puzzle Championship, earning me the right to fly to Rio de Janeiro for the 16th World Puzzle Championship.

This is expensive. Many teams around the world have sponsorship—for instance, Google gives the Americans an all-expenses paid vacation—but not in Canada.

My budget would look something like:
  • Entry fees, accommodation, meals ⇒ €400 ≈ $587.60
  • Brazilian Visa ⇒ $90
  • Spending money ⇒ $400
  • Plane fareeither $1179.94 or (7300 - N) air miles + ($245 + $0.30N)
  • EDIT: Carbon offset for plane emissions ⇒ €154 ≈ $226.23
  • EDIT: Travel insurance ⇒ $105
The total is either $2588.77 or ($1653.83 + 4184 air miles + any proportional combination of 3116 air miles and $934.94).

I can afford about $1000.1

I have approximately one week in which to let the team organizers know whether or not I'm coming.



If you would like to help me get to Rio, please let me know as soon as possible if you're able to contribute any amount of money, or any number of air miles, to my cause. Please do not actually send any money or air miles yet! Based on people's responses to this preliminary plea, I'll decide within a week whether the fundraising drive is actually on. You'll hear.

Note that if 16 people all contribute $100, I'm there…but if 160 people all volunteer $10, that's just as good!

I am screening comments to this post by default, but I will unscreen any comment which includes the word "penguin."

Thanks.

EDIT: How to donate air miles )
EDIT #2 )
EDIT #3: Carbon offsets )

1Although I've been saving up for this event for most of a year, we've suffered some financial setbacks in the past month (details available if I know you personally), which mean that I can afford only about $1000.  Nothing terribly dramatic, just expensive and unexpected.

Rio.

Jun. 21st, 2007 07:48 pm
osmie: (Default)
Well, I was wrong. Even 221 points was enough to place second in Canada, meaning that I qualify to go to Rio in October.

That is, I qualify to spend $3000 of my own money to go to Rio in October, because the Canadian team still has no sponsors. Which is to say, the way my finances are shaping up, I get to stay home again and wish I was independently wealthy.

I'm almost disappointed to have done so well, because it means I have to decline another invitation.

Rio.

Jun. 21st, 2007 07:48 pm
osmie: (Default)
Well, I was wrong. Even 221 points was enough to place second in Canada, meaning that I qualify to go to Rio in October.

That is, I qualify to spend $3000 of my own money to go to Rio in October, because the Canadian team still has no sponsors. Which is to say, the way my finances are shaping up, I get to stay home again and wish I was independently wealthy.

I'm almost disappointed to have done so well, because it means I have to decline another invitation.
osmie: (Default)
—this time in the August Games, with two more fairly ordinary Penguin Paths. I thought they were going to publish one of my variations this time around, but apparently not.

That's three publications & seven puzzles since my last submission. I really must get around to designing some more. I have this neat idea for a Christmas-themed logic puzzle, which needs to get to them before July if I'm to have a hope of placing it in the December issue.
osmie: (Default)
—this time in the August Games, with two more fairly ordinary Penguin Paths. I thought they were going to publish one of my variations this time around, but apparently not.

That's three publications & seven puzzles since my last submission. I really must get around to designing some more. I have this neat idea for a Christmas-themed logic puzzle, which needs to get to them before July if I'm to have a hope of placing it in the December issue.
osmie: (Default)
Just out -- the July 2007 issue of Games magazine, featuring four of my "Laser Tag" puzzles! Everybody go buy a copy!
osmie: (Default)
Just out -- the July 2007 issue of Games magazine, featuring four of my "Laser Tag" puzzles! Everybody go buy a copy!
osmie: (Default)
VCon has moved from Thanksgiving to a later weekend this year -- which means that I can attend from October 19th to 21st and travel to Brazil October 6th to 11th for the World Puzzle Championships!* I get to do both! Yay! And I won't even have to be jet-lagged!

The biggest difficulty I foresee is that it appears to be impossible to travel from Vancouver to Rio de Janeiro without changing planes somewhere in the US. This is going to test my resolve not to enter that country until they reinstate human rights for foreign citizens, and I fear my resolve may not be up to the challenge. Maybe if I'm lucky, the US Congress will decide to reinstate the Bill of Rights before October, and it won't be an issue. Go US Congress!

It's been a very busy couple of weeks at work, but life continues on. Last Sunday I climbed my first 5.8 route, and tonight I'm going to see about tackling my second. Edit: I did it, too! Oh, and I've also spent quite a bit of time on Google trying to figure out what these numbers actually mean, qualitatively speaking. It's harder than it sounds. All I've been able to find out is that a 5.8 is "harder than a 5.7 but easier than a 5.9" (thanks; I figured that out already) or that it's "about as hard as the Such-and-such Climb in Such-and-such State Park" (great trivia; not helpful on a limited travel budget). Does anybody know of a good resource which actually defines these numbers in terms of anything concrete, like the techniques that you might need to use to scramble up the route?

Last night I tried out a recipe I saw on the food network while standing in line at the bank t'other day: fried shallots & garlic in broth, layered beneath puréed peas & mint, layered in turn beneath seared salmon filets. Numm. I swapped veggie broth for chicken broth, since I don't eat chicken and all, and skipped the lemon zest in the broth on account of not having any lemons on hand, but the whole thing worked remarkably well. Puréeing peas turns out to be much easier with a rolling pin than with a blender, though it must be noted I still haven't figured out what blenders are actually good for -- apart, that is, from searing & liquefying two centimetres of food at a time, at which point the blender must be taken apart and the blades scraped off before reassembling it with another two centimetres of wannabe purée.

Then I unwittingly invented hamburger cookies: cookie dough which turned out to be rolled several times the thickness I originally intended, baked until it rises to more than half a centimetre thick, allowed to cool, then set in pairs with likewise-denser-than-I-planned chocolate icing between. What was supposed to resemble homemade reverse-colour Oreos ended up looking a lot more like miniature hamburgers. They still tasted yummy though.

==> And I still need to find some more open mike nights. If anybody should notice such a thing at your local cafe, please copy down the address & contact numbers & let me know!



*Assuming I qualify, that is. Which isn't altogether unlikely, given my record these last few years, but is by no means certain.
osmie: (Default)
VCon has moved from Thanksgiving to a later weekend this year -- which means that I can attend from October 19th to 21st and travel to Brazil October 6th to 11th for the World Puzzle Championships!* I get to do both! Yay! And I won't even have to be jet-lagged!

The biggest difficulty I foresee is that it appears to be impossible to travel from Vancouver to Rio de Janeiro without changing planes somewhere in the US. This is going to test my resolve not to enter that country until they reinstate human rights for foreign citizens, and I fear my resolve may not be up to the challenge. Maybe if I'm lucky, the US Congress will decide to reinstate the Bill of Rights before October, and it won't be an issue. Go US Congress!

It's been a very busy couple of weeks at work, but life continues on. Last Sunday I climbed my first 5.8 route, and tonight I'm going to see about tackling my second. Edit: I did it, too! Oh, and I've also spent quite a bit of time on Google trying to figure out what these numbers actually mean, qualitatively speaking. It's harder than it sounds. All I've been able to find out is that a 5.8 is "harder than a 5.7 but easier than a 5.9" (thanks; I figured that out already) or that it's "about as hard as the Such-and-such Climb in Such-and-such State Park" (great trivia; not helpful on a limited travel budget). Does anybody know of a good resource which actually defines these numbers in terms of anything concrete, like the techniques that you might need to use to scramble up the route?

Last night I tried out a recipe I saw on the food network while standing in line at the bank t'other day: fried shallots & garlic in broth, layered beneath puréed peas & mint, layered in turn beneath seared salmon filets. Numm. I swapped veggie broth for chicken broth, since I don't eat chicken and all, and skipped the lemon zest in the broth on account of not having any lemons on hand, but the whole thing worked remarkably well. Puréeing peas turns out to be much easier with a rolling pin than with a blender, though it must be noted I still haven't figured out what blenders are actually good for -- apart, that is, from searing & liquefying two centimetres of food at a time, at which point the blender must be taken apart and the blades scraped off before reassembling it with another two centimetres of wannabe purée.

Then I unwittingly invented hamburger cookies: cookie dough which turned out to be rolled several times the thickness I originally intended, baked until it rises to more than half a centimetre thick, allowed to cool, then set in pairs with likewise-denser-than-I-planned chocolate icing between. What was supposed to resemble homemade reverse-colour Oreos ended up looking a lot more like miniature hamburgers. They still tasted yummy though.

==> And I still need to find some more open mike nights. If anybody should notice such a thing at your local cafe, please copy down the address & contact numbers & let me know!



*Assuming I qualify, that is. Which isn't altogether unlikely, given my record these last few years, but is by no means certain.
osmie: (Default)
I've continued to tinker with my Perl crossword generator, adding optimizations and minor features and watching it fail to construct a crossword on a fairly open grid with lots of 7+ letter words.  (That my test grid be so intransigent is important at this stage; how can I optimize a program that keeps finishing its job?)

I've also written another, much shorter Perl program which slurps an arbitrary number of text files, identifies all of the individual words they use, and outputs a set of crossword-building index files, organized by length.  (It also outputs a single master file, so that I can go through the list and remove all the useless, unclueable words—e.g. proper names of minor characters, phonetic spellings of people with their mouths full, typos, etc.)  Currently my puzzle generator is working with a list of every word used in any of Bleak House, Middlemarch, Anne of Green Gables, Someone Comes to Town, Someone Leaves Town, and The Special and General Theories of Relativity, all courtesy of the Gutenberg Project.  It's not a bad word list, as these things go, but it clearly needs a lot of editing—and probably a lot more input texts!—before I can create any useful crosswords.

Even at just a couple of hundred lines, this is the most ambitious Perl program I've ever written, and I'm feeling very pleased with myself.  So far my most exciting discovery has been an honest-to-goodness bug in the language.  After poring over my code for hours looking for the circular reference which was causing Perl's garbage collection to fail and my program to gobble 3 megs of memory every second, I finally learned how to use the Perl debugger.  Within ten minutes I'd discovered the memory leak…and it wasn't in my code at all.  The default MacOS implementation of the first() subroutine in the List::Util module is buggy—at least when comparing the list items to a regular expression.  Beats me whether this bug has already been found & fixed in the current Perl revision, but for my purposes it was easy enough to replace the command with an explicit loop.  Presto!  The program now runs for an arbitrary amount of time in less than 4 megs.

That said, it can still take days and days to decide that a given grid has no solutions.  Which isn't really surprising, when you think about it.  It starts with the longest word in the grid, but it still has to try every single (say) 10-letter word in the language—or at least every class of 10-letter word, to the best of its optimizing ability—to be dead certain that none of them will fit.

Yay me!

Soon I'll get back to those Laser Tag puzzles so that I have enough to submit to a magazine.  Has anybody out there had a chance to try them yet?  I haven't heard any feedback at all.
osmie: (Default)
I've continued to tinker with my Perl crossword generator, adding optimizations and minor features and watching it fail to construct a crossword on a fairly open grid with lots of 7+ letter words.  (That my test grid be so intransigent is important at this stage; how can I optimize a program that keeps finishing its job?)

I've also written another, much shorter Perl program which slurps an arbitrary number of text files, identifies all of the individual words they use, and outputs a set of crossword-building index files, organized by length.  (It also outputs a single master file, so that I can go through the list and remove all the useless, unclueable words—e.g. proper names of minor characters, phonetic spellings of people with their mouths full, typos, etc.)  Currently my puzzle generator is working with a list of every word used in any of Bleak House, Middlemarch, Anne of Green Gables, Someone Comes to Town, Someone Leaves Town, and The Special and General Theories of Relativity, all courtesy of the Gutenberg Project.  It's not a bad word list, as these things go, but it clearly needs a lot of editing—and probably a lot more input texts!—before I can create any useful crosswords.

Even at just a couple of hundred lines, this is the most ambitious Perl program I've ever written, and I'm feeling very pleased with myself.  So far my most exciting discovery has been an honest-to-goodness bug in the language.  After poring over my code for hours looking for the circular reference which was causing Perl's garbage collection to fail and my program to gobble 3 megs of memory every second, I finally learned how to use the Perl debugger.  Within ten minutes I'd discovered the memory leak…and it wasn't in my code at all.  The default MacOS implementation of the first() subroutine in the List::Util module is buggy—at least when comparing the list items to a regular expression.  Beats me whether this bug has already been found & fixed in the current Perl revision, but for my purposes it was easy enough to replace the command with an explicit loop.  Presto!  The program now runs for an arbitrary amount of time in less than 4 megs.

That said, it can still take days and days to decide that a given grid has no solutions.  Which isn't really surprising, when you think about it.  It starts with the longest word in the grid, but it still has to try every single (say) 10-letter word in the language—or at least every class of 10-letter word, to the best of its optimizing ability—to be dead certain that none of them will fit.

Yay me!

Soon I'll get back to those Laser Tag puzzles so that I have enough to submit to a magazine.  Has anybody out there had a chance to try them yet?  I haven't heard any feedback at all.

Profile

osmie: (Default)
Osmium Penguin

April 2016

S M T W T F S
     12
345678 9
10 111213141516
171819202122 23
2425262728 2930

Syndicate

RSS Atom

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags