Canada, Moving Lead to Slacking Blogger

Sorry I have been a lazy blogger.

Here is a photo with several of the graduate students in my lab in Montréal, Canada.

Toys from Google Labs: Finding Sets

I stumbled across a toy under development at Google, Google Sets. If I am reading things correctly, this particular tool must not be under very active development because I believe its inception date was 2002.

At any rate, the basic idea of the tool is to enter up to five search terms, and the algorithm will attempt to make either a Small Set (15 items or fewer) or a Large Set from those items.

For example, if you enter titles of reality television shows, it returns more reality shows.

However, it appears to be a bit useful for stalking.

I entered my name and the names of two former colleagues, "samuel bradley", "yongkuk chung", and "mija shin". If you do that and click Large Set, you get a pretty nice list of my former colleagues in the Institute for Communications Research at Indiana University.

Cool ... and spooky.

Put Corn Down and Slowly Back Away, Americans

Gasoline is, like $800 a gallon, and I eat approximately 1 lb. of corn every 30 seconds.

The population continues to grow. I admit it. I am worried.

Read more in the New York Times.


Thanks Wendy and/or Wes.

Just Wait until They Try Skype

From CNN.com:

Landlines go dead as more users rely on cell phones

WASHINGTON (AP) -- For nearly three in 10 households, don't even bother trying to call them on a landline phone. They either only have a cell phone or seldom if ever take calls on their traditional phone.

The federal figures, released Wednesday, showed that reliance on cells is continuing to rise at the expense of wired telephones. In the second half of last year, 16 percent of households only had cell phones, while 13 percent also had landlines but got all or nearly all their calls on their cells.

Personally, we're using Skype at home.

Invisible Social Forces in Name Popularity

How do people name their kids?

Often people name a new baby after some relative or friend of the family.

However, in the mass media era, naming children has an often unnoticed social component.

I have four daughters, and we went through a different process in naming each of them.

The source of the names, however, interests me each year when the Social Security Administration releases their top names.

They have a very nice Web site where you can track the popularity of names over time. For me, this is particularly fascinating process. For each of my first three daughters, we selected a name that was trending upward.

In the above figure, I have tried to denote the year we picked the name with a vertical black line. However, keep in mind that we would have been looking at SSA data from the year before. We did not know what names would be selected in 2000, for example.

The fourth and final kid's name does not fit the pattern -- it has been stable for a few years -- however, we really picked that name in 2000. It was an alternate for the second kid. And back then it was very much trending upward.

Now let me try to let you in on our basic algorithm. We tried to pick names where they would not have three other kids in their class with the same name, but not so uncommon they would hate us ("Hi, this is my daughter, Rihanna").

Somehow other people must have been having similar thoughts, whether or not those parents' intentions were conscious.

These curves really do fascinate me. With very little deviation, these names became increasingly popular each year. That's an interesting social phenomenon in a country of 300 million people.

Somehow this reminds me of Adam Smith's invisible hand working its way through society.

I'd be interested in your thoughts.

Perish the Published: Making Texts Affordable

From the New York Times:

EDITORIAL
That Book Costs How Much?
Published: April 25, 2008
Colleges and universities will need to embrace new methods of textbook development and distribution if they want to rein in runaway costs.

Artificial Intelligence: My Current Passion

Reading some social network site or the other, I saw that colleague James Angelini said it has been a year since his graduation ceremony at Indiana University.

That means that it's been three years since my own graduation. It may not seem like that long, but it seems like an eternity to me.

It's been a great three years at Ohio State and Texas Tech, but one part of my brain has not received much attention: computational modeling.

My love for these models flourished at Indiana's program in cognitive science. That program is among the handful of top programs in the world, and it marked the three greatest intellectual years of my life.

I'm slowly working on a project with Tim Laubacher, former OSU master's student and current Columbus, Ohio, advertising executive. This is the first new computational idea in a while, and it is quite exciting.

Nonetheless, I am reminded of he dormancy of this part of my brain on days such as today when I open the New York Times to read:

TECHNOLOGY

Pursuing the Next Level of Artificial Intelligence

By JOHN MARKOFF

Published: May 3, 2008

Daphne Koller’s work has led to advances in artificial intelligence that can be used to predict traffic jams, improve machine vision and understand the way cancer spreads.

Ah, I wish that I were doing work such as this!

Cool Colleagues: My Friends in Columbus

A creative approach to work

Dublin ad agency's outside-the-box policies aimed at inspiring workers

Friday, May 2, 2008 3:23 AM

By AMY SAUNDERS

THE COLUMBUS DISPATCH

At Dublin advertising agency b&a, employees are allowed to drink on the job, Hula-Hoop in the conference room and return from vacation whenever they feel like it.

As a result, they come to the office, behave responsibly and do their work. Really.

"Why wouldn't they?" asks principal and founder Jack Buchanan. "They have a job to do."