About Harrison Baker

Harrison Baker works as an aquarist at Oregon State University's Hatfield Marine Science Center. His academic background is in animal husbandry, journalism and editing. He is currently pursuing an MS in Free-Choice Learning Science Education. His board game, Deme, is currently under development as a component of his MS project on games and adult learning.


Ladies and gentlemen, I present for your consideration an example of our signature rapid prototyping process. The handyman’s secret weapon gets a lot of use around here, and I even had a roll of Gorilla Tape on my wrist in case of emergencies.  Fortunately, it didn’t come to that.

The angles necessary for good face detection and recognition (up to about 15 degrees from straight-on) require careful consideration of camera placement.  The necessary process of checking angles and lighting isn’t always pretty, but I, for one, find the above image beautiful.

With Mark’s guidance over the phone, I spent a few hours today testing camera placement with a small Axis camera and its built-in microphone. One of my favorite security features of this camera is its built-in speaker, which can be used to make the camera shout “intruder,” whisper “pssst,” or bark like a dog.  None of these have any conceivable utility whatsoever for what we’re doing, but it’s always nice to know we have options.

So, I put it in the entryway.  I put it over and next to the octopus tank.  I put it over the front desk. I put it by the touch pool, which triggered a barrage of eyeball-seeking dust particles that had been guarding the overhead ethernet ports for untold eons.

Each vantage point tested presented a decent view and adequate lighting.  The model I used will not be installed in all positions, but it provides a great baseline.  We also received a new Axis dome camera with a microphone, which we can use up-close at individual exhibits.

To record a few audio tests, I directed the system output of one of our Macbooks into Audacity using Soundflower. Having recently spent several late nights playing with open-source audio software, I improvised this solution a bit more easily than I had anticipated. I never expected that my private dubstep habit would prove to be a reservoir of generalizable workplace skills, but it goes to show that free-choice learning happens all the time.

Alan Alda and the Center for Communicating Science have a challenge for scientists: explain a flame to an 11-year-old.  Brilliant.  You can read more about this (and submit your entry) here.

“As a curious 11-year-old, Alan Alda asked his teacher, “What is a flame?” She replied: “It’s oxidation.” Alda went on to win fame as an actor and writer, became an advocate for clear communication of science, and helped found the Center for Communicating Science at Stony Brook University. He never stopped being curious, and he never forgot how disappointing that non-answer answer was.”

Alda’s guest editorial for Science, wherein he issued his challenge, is also well worth reading.  This can also be found at the Flame Challenge site.

Do it for yourself.  Do it for the kids.  Do it for Hawkeye.


Mark and I did some scale-model wave tank testing this afternoon.  An initial test presented some hurdles (waves splashing over the far end of the tank, waves rebounding and creating mid-tank chaos, etc.).  Mark introduced a novel scale-model component (a scouring pad at the end of the tank) to disperse the wave energy and prevent the waves from bouncing back.

With this humble addition, the model tank performed admirably, providing practical reassurance that the proposed measurements for the final design will demonstrate the relevant concepts without soaking the floors.  Any handle, button, lever, knob or switch in an exhibit space must be built to accommodate a range of perceivable affordances.  If pulling the lever triggers an interesting result, pulling it ever harder and faster might produce even more interesting results.

This can sometimes put wear and tear on exhibit components, but it’s part of what makes hands-on exhibits fun for learners (and learning researchers, too).


Michelle will be posting this week from the Exploratorium.  She’s currently working with NOAA scientists and some of our iPad apps.   Stay tuned.

In the meantime, here’s something to keep you occupied.  An AI called “Angelina,” developed as part of Michael Cook‘s Ph.D. project at Imperial College, generates (almost) entire games procedurally.  From the New Scientist piece:

“Angelina can’t yet build an entire game by itself as Cook must add in the graphics and sound effects, but even so the games can easily match the quality of some Facebook or smartphone games, with little human input. ‘In theory there is nothing to stop an artist sitting down with Angelina, creating a game every 12 hours and feeding that into the Apple App Store,’ says Cook.”

The capacity of games to teach is a research interest of mine, and I think the most interesting thing about Angelina is its ability to run through its own creations to determine (presumably using human-defined parameters) how engaging they are.  It shows in the New Scientist-commissioned “Space Station Invaders” demo game, which is a retro platformer with some nice simple jumping challenges.  The player character’s immortality is a welcome inclusion, as the aggressive procedurally-generated enemy behaviors give new meaning to that classic gamer complaint: “The computer cheats.”