Monday, February 23, 2009

Siftables: Spatially Aware Building Blocks

In his recent TED talk, David Merrill shows how, once again, a simple idea can be quite powerful when executed in just the right way. In the video of the talk, embedded below, you will see Merrill demonstrate Siftables, little tiles with LCD screens that are spatially aware of each other.

As described on Merrill's Siftables research page:

Siftables aims to enable people to interact with information and media in physical, natural ways that approach interactions with physical objects in our everyday lives. As an interaction platform, Siftables applies technology and methodology from wireless sensor networks to tangible user interfaces. Siftables are independent, compact devices with sensing, graphical display, and wireless communication capabilities. They can be physically manipulated as a group to interact with digital information and media. Siftables can be used to implement any number of gestural interaction languages and HCI applications.

The components of the Siftables blocks are nothing new on their own. In fact, I wouldn't be surprised if even the combination of these ideas has been tested before. But when you see the interesting applications of such a system, such as the mathematical equation setup or word play game (seen in the video), you can't help but be excited.

Funnily enough, when showing the video to my husband, the first thing he thought of was how fast young children would probably destroy the little blocks, which he assumed were delicate. This worry was soon squashed as Merrill showed several videos of children testing out the word game as well as a really creative interactive cartoon application. No blocks were harmed - not bad for what was probably just a research prototype.

I'm convinced that Siftables could be used for a wide variety of interesting educational games. They might help students visualize complicated, spatially oriented word problems in math class, for example. Or they might be used by chemistry students as they build various molecules as they study organic chemistry. Maybe even youngsters introduced to computer science could benefit as they see the effects of rearranging various parts of an algorithm. All this from nothing but a few simple blocks!

Friday, February 13, 2009

Turn Around and Try Again

Ah, research. The land of never knowing when you'll have to take a few steps back and choose another path. Though this is how I had imagined the world of research, I secretly figured it wouldn't happen to me. But, it has, and so I am not "done my research by Christmas" as I had originally hoped. (To be fair, I also secretly figured I wouldn't be done by then, anyway.)

My last thesis update was way back in November '08. I gave the most recent description of my research, and talked about a small bit of theory that I needed to confirm before moving on. Back then (and up until now), my thesis was supposed to be a simple experimentation of a particular class of interest point detectors to show how well they worked for matching photos with cube and cylinder panoramic images.

Unfortunately, I was never quite able to get the photos and panoramas to match automatically in any sort of reasonable way. I tried a few different experiments to show this, but it was difficult to convince ourselves that the strategy used just wasn't working until I really pared it down to the bare essentials. Basically, what I ended up showing was the inadequacy of the interest point descriptors in the context of comparing them in the panoramic and photographic images. You can see some of the results here.

Now I must seek another reasonable method for matching these particular images, given the following:
  • The panoramic images are not terribly sharp in their quality. This is related to the fact that some areas of the image are sampled from several sources. Fine features are thus unlikely to be reliably detectable.
  • In most comparisons, buildings will be the only structure that will appear in a scene consistently. The ambiguity from repeated structures in these buildings (such as windows) will pose a challenge.
  • It may be reasonable to assume that we are always matching images of the same scene thanks to a previously run recognition algorithm and/or GPS location attempt.
  • The matching process must result in being able to find the geometry between the cameras of the two scenes.
I have a couple of different possibilities to look into at the moment. This time, I'm trying to start with the basics. Let's hope I don't have to go too far down any other wrong paths, because this time, I really do have to finish by May (I'll tell you why later).

Wednesday, February 11, 2009

Mommas don't let their babies grow up to be engineers

I wrote up this quick post for the CU-WISE blog after seeing the article in IT World. Figured it was worth sharing here, too.

Check out this article from IT World: Mommas don't let their babies grow up to be engineers.
More than 85% of students today aren't considering careers in engineering, a new survey found, as more parents encourage girls specifically to become actresses than IT professionals.
Really, eh? Have you ever heard of your girlfriend's telling her to hit Hollywood back when she was a wee tot? I know I haven't. Either way, that's a scary thought.

The more important statistics for us to consider are the following:
  • 44% of students polled said "a lack of knowledge around engineering as the top reason they would not pursue such jobs"
  • 30% thought engineering was too geeky or boring
  • 22% didn't feel their math and science abilities were adequate (I once heard that this is despite the fact that female students usually scored higher than their male counterparts - the guys were just more confident)
  • 20% of parents encouraged a career in engineering
As they say in the article:
"It's clear that there is a low level of interest and knowledge about engineering careers for both parents and children," said Maurice Ghysels, chair of ASQ's K-12 Education Advisory Committee. "Educators and engineers need to work more closely together to get students excited about the profession and spotlight interesting role models."
That's why one of the main goals of CU-WISE is to help young women see that they can do science and engineering. We know there are girls out there that have the potential to not only succeed in these fields, but to really and truly enjoy themselves in the process. We want to help them discover this potential. By participating in our many outreach activities, you, too, can help with this cause!

Tuesday, February 10, 2009


Mentoring has been a concept I've been thinking about a lot lately, mostly in the context of my work with Carleton University's Women in Science and Engineering group. We are hoping to implement some kind of mentoring program either this semester or in time for the beginning of the next school year. While funding and support from our Deans is not a problem, we haven't quite nailed down how exactly to run such a program.

The main issue to me is that we want to get it right the first time. If we start a program that pairs faculty or professionals with undergrads and high school students, but don't design it in such a way that mentoring relationships don't fizzle out, the original participants aren't likely to try again in future attempts.

With this in mind, the biggest mistake we could make would be to "just create an online form or forum where mentors and mentees could connect." Unfortunately, left to their own devices, most people would find other priorities to fill their time, and that online mentoring site they signed up at last month can slowly slip from their minds until it's completely out of the picture. And that's just the people who actually took the iniative to register in the first place - surely a minority.

That's not to say that an online mentoring site can't be successful. Alberta's CyberMentor seems to be doing well. But as a relatively small student-run group (we only have five executive members and a huge program to run, after all), CU-WISE cannot set up such an elaborate online presence.

I also don't want to suggest that an online component of a mentoring program wouldn't be valuable. Of course, the more ways to communicate, the better.

The main point here is that we need something more. Incentive to participate. Physical meeting opportunities. Support on how to get the most out of the mentor/mentee relationship. A reason to care. This is where I'm still somewhat at a loss. What is the best way to accomplish all this?

If you've had success with mentoring programs in the past, or heard of some great ideas, I'd really love to hear about it in the comments.

Friday, February 6, 2009

I Am "Modern Graph Theory"

I couldn't resist posting my results for The Springer GTM Test. They were uncannily accurate in my case.

All you do is answer a few simple questions, and then find out what Springer graduate text you are. Here's my result:

If I were a Springer-Verlag Graduate Text in Mathematics, I would be Bela Bollobas's Modern Graph Theory.

I am an in-depth account of graph theory, written with the student in mind; I reflect the current state of the subject and emphasize connections with other branches of pure mathematics. Recognizing that graph theory is one of several courses competing for the attention of a student, I contain extensive descriptive passages designed to convey the flavor of the subject and to arouse interest.

Which Springer GTM would you be? The Springer GTM Test

Now, the reason why I think this is so accurate is that my biggest passion so far has to be my desire to improve the academic lives of students. That's why I was President of the undergrad computer science society, and why I'm working so hard with the Women in Science and Engineering group now. I am quite interested in how to best teach a topic to students so they remain interested - hence my use of video games to teach grade 8 girls computer science topics in my mini-course. I also want to stay current in terms of "the real world," so I try to choose projects that allow me to continue practicing my software development skills.

How'd they know all that??

Jade Redmond

I finally got the chance to learn more about Jade Redmond, who I previously only knew as "the woman who made Assassin's Creed." The funny thing is that I did so with an article found on that begins with a sexiness rating.

The cool part is that the sexiness description includes more than just physical attributes, and her biography contains none. I learned that she is Canadian born (from Montreal!), and graduated from McGill University with a degree in computer science. Her first job was creating Sony Online's first R&D group, after which she moved to produce The Sims Online at Electronic Arts. Her huge successes at these companies landed her the position of lead producer for Assassin's Creed, a game that's been called by Game Spot, among other wonderful things, “one of the finest gaming experiences ever created.” Nowadays, you can find Jade on The Electric Playground, "bringing her face-to-face with video game industry colleagues and the fans who’ve supported her over the years." All this, and she's only 33!

I can't believe I've missed out on knowing this awesome role model until now! I mentioned her in passing during my mini-course on video games for girls, but I think I'm going to have to change that for this year's edition. I find it amusing that it took a male friend reading a men's site (which is practically obligated to talk about how sexy women are at some point) to finally dig deeper on Jade's background. I am relieved, however, to see that was respectful in their descriptions, and gave Jade all (or at least most) of the credit she deserves.

Oh yes, one more thing: I think I'm going to have to buy a copy of Assassin's Creed now...

Tuesday, February 3, 2009

Blog Template Redesign

Please excuse the appearance of this blog for the next day or so as I try to fit it into a new template. I'm hoping the new layout will be a little easier to read, particularly for its fluid width for the content column. Thanks!

Battle Royale Round 2

On January 24th, 2009, gamers in the Ottawa area flocked to Carleton University's Porter Hall for 24 hours of hard core gaming action. The spectacle is lovingly named Battle Royale and was celebrating its second year.

Battle Royale

I stopped by the big event wearing my Google Campus Ambassador shirt, goodies in hand. I was very happy to be able to connect the organizers of Battle Royale with my contacts at Google, and even happier that Google was willing and able to sponsor the event with giveaways and prizes. I visited each gamer personally, giving them their choice of Google pen and sticker set. I was pleased to find that most didn't mind risking whatever game they were in the middle of playing to select their swag and chat for a minute or two.

Spread the Google Love

Google sent me two raffle prizes as well: a backpack and a jenga set with the Google logo. They always find such creative ways of branding everyday items with their famous logo. I hope the winners enjoy being the envy of their friends. :)

Google Backpack Prize Google Jenga Prize

Overall, I was pretty impressed with the setup of this event. It was obviously not trivial to satisfy the power requirements for over 100 computers, and I wouldn't even know where to begin when trying to network them all together. There were two big projector screens for playing Rock Band and Super Smash Brothers, drinks and snacks for sale (the first two of which were included with the ticket price), and probably enough prizes for everyone to walk away with something. Kudos to the organizers!

A few more photos of the event can be found here.