strottrot.com » computing

Using PicoCrickets to Teach Debugging

strottrot — Thu, 07 May 2009 04:31:40 +0000

We ran into trouble yesterday trying to make a cat meow when left in the dark. The cat was a PicoCricket, a programmable kit for making creations that move, make sounds, and light up based on inputs like touch, sound, and light. We were trying to make a cat that purrs when it’s dark, but the thing purred no matter the light conditions. I did not plan it, but it turned into a good debugging exercise for my daughters (ages 10 and 12). The image on the left shows the program with which we began, which was running in concert with a similar program (stack of blocks) to make the cat’s collar light up. (This is one of the sample projects sent with the kit.) The program on the left basically instructs as follows: "Keep doing the following: if the sensor is picking up light at a brightness of less than 20, then play the sound of a kitten." ("20" what, I am not sure.)

We were not sure if the ever-present purring problem was physical or programmatic, so we checked out some of the physical things first: Cables snugly connected? Batteries strong enough? We tried swapping in a different light sensor. Next we moved on to programmatic debugging. We eliminated the additional set of blocks (program) to see if that was part of the problem. Still the cat purred as soon as we turned it on. We tried a different variable (the input of touch instead of brigthness). This time it recognized the sensor and did what we intended: It did nothing until we pressed the button on the touch sensor, at which time it purred. (I realize there’s no recognizable cat here: imagine a felt thing with ears, eyes, and whiskers with the PicoCricket parts hidden inside.)

Next we tried going back to the original program, but this time testing for "dark" instead of "brightness less than 20." Perhaps the way we were using brightness was faulty. Alas, dark also did not work: the cricket purred as soon as it was switched on.

The question of what the light sensor measures was now paramount. If we were working on a screen, I would output as much dynamically generated information as possible. This was the key. We added a "display brightness" command to the program. It turns out that the fluorescent light in our living room rates a brightness of only 3 or 4 on the display. The cat purred right away when the test was "brightness less than 20" because it was indeed less than 20. Covering the sensor with our hands, we found that "dark" is triggered by a brightness rating of 3 or lower. So the program and our sensors were working; we just needed to change the brightness test to "brightness less than 2" in order to have consistent control over when our cat purrred.

I loved how intrigued the kids were (both at home at at the local tween center where students I work with did an outreach activity with the PicoCrickets). The crickets were developed at the MIT Media Lab to engage young people in computing through art and storytelling. Scratch is a programming app (animation, games, art, music) from MIT that kids can download (free), create stuff with and then upload to share. Scratch uses a similar graphical interface with snapping blocks like the ones seen on this page. Storytelling Alice (from Carnegie Mellon student Caitlin Kelleher) is a "programming environment designed to motivate a broad spectrum of middle school students (particularly girls) to learn to program computers through creating short 3D animated movies." The National Center for Women in IT (NCWIT) offers Outreach in a Box kits to help people get started doing small outreach projects to get young girls interested in computing. Time to do some outreach!

Shuttle with Frances Allen

strottrot — Sat, 20 Oct 2007 12:08:49 +0000

I had the honor of sitting next to Fran Allen on the shuttle to the airport at 6 am this morning. (Wow–the Turing Award winner opts for the shared shuttle.) I asked if there is particular work she is excited about in the coming year. Allen said she will be speaking a lot about the current crisis in computing, which she described as probably the largest computing crisis we’ve ever faced, but one that she believes can be addressed by her field. (Shame on me–I didn’t ask what the crisis is, thinking she would tell me if she wanted to discuss it further at 6 am on the way to the airport, but looking back, I’m guessing she would have been happy to discuss it… Ironic, given how struck I was that she reminded me of my mother-in-law, an incisive reporter who no doubt would have asked.) After a bit of googling, I think Allen was referring to energy crises and power consumption of high-performance computers–and the potentials for parallel computing and optimization to address these problems. Regarding the topic of the conference we were departing, she seemed optimistic about engaging women in the field of computing, calling that problem “the easy one.” She joked about how she would prefer not to focus on being the first woman to win the award, and rather to focus on her work. She said her friends at ACM laughed at the possibility of leaving that out of the conversation. We sat with another woman from IBM, who does mentors many IBMers in her work. Both she and Allen spoke about the need to have discussions about gender and cultural differences–to put the issues on the table–in the workplace and in academia, and that not having such discussions remains a serious problem.

Grace Hopper Women in Computing

strottrot — Thu, 18 Oct 2007 03:21:39 +0000

I left HighEdWebDev in Rochester NY this morning and I’m thrilled to have landed in Orlando FL for the Grace Hopper Celebration of Women in Computing. There was great energy at the poster session when I arrived. It’s striking to go from the HighEdWebDev scene–professionals managing the bureaucratic change-fearing worlds of colleges & universities, striving to bring innovative technology to their learning communities–to the Hopper scene of women in doctoral programs sharing their fascinating projects exploring future possibilities. Makes me want to go back to school!

Last night’s banquet was quite moving as Fran Allen, the 2007 Turing Award Winner (first woman ever to win) was honored. Each speaker was asked to offer 20 words of advice for the women in the audience: Fran Allen told a story about climbing a 14,000′ Bhutanese mountain. As she was climbing, she noticed that one of the guides was putting stones across a brook for her to walk over. She reassured him that it wasn’t necessary–she could get wet going across. He explained that in Bhutan, they have a tradition: IN my country, we make the path easier for those who come behind. After Allen spoke, IBM announced (and awarded) a new Fran Allen Fellowship which will include an IBM mentor for each recipient–for the first year Fran Allen herself.

Dr. Monica Martinez-Canales, chair of the Richard Tapia Celebration of Diversity in Computing, spoke most convincingly about the importance of mentors. She encourged everyone to find a mentor and hold on tight: ask them how you can be like them, and take their advice.