• Ricky Jay’s Magical Secrets

    Ricky Jay is a legend, and is without a doubt one of my favorite magicians to watch. His shows leave you with a greater appreciation for the magician’s craft, rather than just the immediate illusion they present.

  • Homer's Last Theorem

    A look into the deep, dark, strangely complicated world of Simpsons mathematics.

    I knew that Futurama has a lot of great mathematics involved, but I didn’t realize that it started in the Simpsons.

  • Pipino: Gentleman Thief — EPIC MAGAZINE — Medium

    Magicians, Mafiosos, a Missing Painting, and the Heist of a Lifetime

  • The Time Everyone “Corrected” the World’s Smartest Woman

    In 1990, Marilyn vos Savant correctly answered a probability puzzle in her column for Parade Magazine. And then, the world called her an idiot.

    I still remember the chaos caused by the Monty Hall problem in our high school math class. Here’s my explanation for how it works:

    • Your first choice could either be a goat or a car.
    • If you picked the car, you will win the car by not changing your guess.
    • If it’s a goat, you will win the car by changing your guess.
    • Since you’re twice as likely to have picked a goat than a car, you’ll be twice as likely to win by switching for your second guess.

    Note that picking a goat and changing your guess will guarantee the car, since the host has revealed the second goat.

    The best strategy is to try to pick a goat in round 1, and plan to switch once the host has revealed the second goat.

  • A Teenager’s View on Social Media — Backchannel — Medium

    Very informative description of how teens and college students use the major social networks

  • The Rise of the Professional Cyber Athlete

    Very cool profile of Scarlett, a professional StarCraftII player

  • 真如堂

  • AWS Lambda

    You can trigger an AWS Lambda function to automatically create a thumbnail when an image is uploaded to Amazon S3, verify address updates in an Amazon DynamoDB table, or process click-stream data in an Amazon Kinesis stream, without having to manage any compute infrastructure.

    Billing is metered in increments of 100 milliseconds, making it cost-effective and easy to scale automatically from a few requests per day to thousands per second.

  • ゲームの空間認識と今後の進化


    マリオはなぜ世界でヒットしたのか? チームラボ・猪子氏が語る、日本的空間認識とクリエイティビティの関係性


    西洋の空間認識だと、パースペクティブを意識する。 よって、絵の登場人物になりきると視点が変わって絵の全体を意識しなくなる。絵の全体の意識が変わる、と言う方が適切なのかもしれない。

    日本の空間認識だと、レイヤーで物を見て、重ねて表現する。 絵の登場人物になりきっても、絵の認識が変わらず、絵の全体を意識し続ける。




    Oculus Rift 用に開発中のLucky’s Taleの話を思い出した。

    Lucky’s TaleはVR機能を活かした3Dのプラットフォーマーなのだが、自分が主人公になりきるわけではない。キャラクターの視点と自分の視点は完全に別で、自分が3Dの世界の中で、主人公の隣に居るという見せ方。



  • Relationship Test - Dataclysm

    My team and I wrote an app that will apply findings from a recent research paper to your Facebook graph. The app won’t post to your wall but it will show you both the shape of your friend network and which of your friends are most mathematically important to your life.

    Instead of looking at a metric like the number of friends you have in common, this algorithm takes it a step further. It looks at how well connected your mutual friends are with each other, and suggests that having more diverse mutual friends indicates a stronger, more important relationship.

    If everyone at work is connected via Facebook, then you will have many friends in common with your coworkers, but this doesn’t mean that these are your most important relationships. Since all of your common friends are within the same social circle, you know that these relationships are limited in scope.

    But if you and a friend’s mutual connections include coworkers, college roommates, and cycling buddies, then you know that your relationship with this person spans many social circles, and they are more likely to be important to you. This also is effective at picking out romantic relationships.

    There seemed to be issues with outliers when I tried it because there were a number of people with little or no mutual friends with very high scores, but in general the results were surprisingly accurate.

    It’s also fun to look at the graph of your network and see the clusters of different social circles.