You are viewing devjoe

devjoe [entries|archive|friends|userinfo]
devjoe

[ userinfo | livejournal userinfo ]
[ archive | journal archive ]

Solving an incomplete multiplication [Jun. 4th, 2014|10:12 pm]
[Tags|, ]

The What Would Martin Gardner Tweet account just tweeted this puzzle of his:


I don't have the original Gardner source handy, but I thought I'd write my own solution.

Read more...Collapse )
LinkLeave a comment

Google Code Jam 2014 Round 1C [May. 11th, 2014|11:09 am]
[Tags|, , ]

I qualified for round 2 already, but I woke up less than an hour into the time of round 1C, and I decided to check out what one of these rounds looks like live when you aren't competing. The answer is that you can see the scoreboard and the problems, but you can't download inputs (must less submit results) until the round ends. This let me mock-compete, starting my own 2.5 hour clock when I was ready, and when the contest ended, I checked results for the first two problems, which I had then completed. If I had been competing, I would have gotten the points for those problems and nothing else, and finished around rank 210 counting the time I needed for debugging and one penalty I would have gotten for a wrong small result.

Detailed analysis (no code) for round 1CCollapse )
LinkLeave a comment

Code Jam 1B problem C: The Bored Traveling Salesman [May. 5th, 2014|08:53 pm]
[Tags|, , , ]

In my previous post, I noted the idea I had with only a few minutes left in the round to solve my one unsolved problem. Tonight I coded up that idea and confirmed it works.

I apologize in advance for a couple aspects of this Python 2 code. For one thing, zipcodes is a list and is indexed from 0 to n-1, while flights is a dictionary and is indexed from 1 to n. I don't know why I did this stupid thing but this is real code, how I programmed it during the event, only modified to add on the isconnected() check. I have added copious comments to help explain the code, though.


This time the code is behind the cutCollapse )

Link1 comment|Leave a comment

Google Code Jam 2014 Round 1B [May. 3rd, 2014|05:42 pm]
[Tags|, , ]

I didn't qualify in round 1A last weekend. I blame tiredness; work has been taxing lately and it was at 9 PM on a Friday. I got the first problem easily enough, and I think I figured out the right algorithm to do problem B small and large inputs, which would have qualified me if I had completed it soon enough, but I was just too tired in the last hour of the event to finish writing the program.

So I tried round 1B today, and finished 253rd to advance to round 2. The rest of this post is about today's problems.

Detailed round 1B analysis, no codeCollapse )
Link9 comments|Leave a comment

Google Code Jam 2014 Qualification Round [Apr. 12th, 2014|11:21 pm]
[Tags|, , ]

The Qualification Round this year just ended. As I realized when competing for the first time last year, this round is really meant to distinguish those who know how to program (at all) from those who don't, and to introduce people to the system. You just needed to score 25 points to advance, out of a possible 90 points, and you have slightly over a full day to do it. I worked on the problems last night shortly after they came out (but during dinner and such, so not rushing through them in full time-trial mode), and ultimately finished 606th among those who got all 90 points.

Detailed analysis (no code)Collapse )
LinkLeave a comment

BAPHL 9 [Apr. 7th, 2014|01:26 am]
[Tags|, , , , ]

BAPHL 9 happened, down in Providence where I had not been since the NPL convention was there not too long ago. I registered, under the name plugh which has gained quite a reputation in some of these small-team hunt events, a team which consisted of four people. Besides me, there was dalryaug, Mike Booth, and Nick Poulos. We finished first, 25 minutes ahead of Death from Above.

Somebody asked us afterward how we can solve puzzles so fast, and "practice" was the number 1 answer. I've been doing Mystery Hunt since 2000 and lots of other puzzle events over the years. I know quite well the typical ways that puzzles can lead to final answers, as well as knowing the standard puzzle types well. My teammates have a little less experience than that, but in recent years we have been doing quite a few different puzzle hunts, so the experience builds rapidly. Also, the ability to play Wheel of Fortune helps sometimes. If you know where those letters are going, write out your partial extraction with blanks for the missing letters. You never know what you'll pull out of it.

Some more possibly spoily details on things that helped us with this particular hunt follow the cut.

the spoiler alert cut thingCollapse )
Link3 comments|Leave a comment

MIT Mystery Hunt 2014, part 5 (Saturday evening through end of Hunt) [Jan. 28th, 2014|11:59 pm]
[Tags|, , ]

This is part 5 (last part) of a post which begins here.

Read more (spoilers)...Collapse )
Link1 comment|Leave a comment

MIT Mystery Hunt 2014, part 4 (Saturday afternoon with all the rounds open) [Jan. 26th, 2014|11:08 pm]
[Tags|, , ]

This is part 4 of a post which begins here.

Read more (spoilers, natch)...Collapse )
Link2 comments|Leave a comment

MIT Mystery Hunt 2014, part 3 (bits of several rounds up to my sleeping break) [Jan. 25th, 2014|10:02 pm]
[Tags|, , ]

Part 3 of a post which begins here.

Read more (spoilers inside)...Collapse )

Continued here.
Link1 comment|Leave a comment

MIT Mystery Hunt 2014, part 2 (parts of the Tea Party, Mock Turtle, and White Queen rounds) [Jan. 22nd, 2014|08:41 pm]
[Tags|, , ]

This is part 2 of a post which begins here.

Read more (with spoilers)...Collapse )
LinkLeave a comment

navigation
[ viewing | most recent entries ]
[ go | earlier ]