Triple Helix

So these past few weeks I have been going to Hackett Middle School twice a week to host an after school session to get the kids acquainted with the skateboarder CSDT. It’s been going well. I’ve mainly just tried to let the kids play around on it so that they come across things they don’t understand and then I can explain things to them.

This week I wanted to try something different, however. I borrowed a group of students’ past project from Ron Eglash called the Fast Track. It’s a hot wheels track wired with an arduino and four sensors and measures the speed a car takes to finish the track (http://pdi-studio5.wp.rpi.edu/fast-track/). I put the opensource code onto my laptop and took the track to the school today. Through this project, the kids learned about speed (being equal to distance over time) and about how gravity and mass help to increase speed. It was a success as there was 100% participation and the competition really excited them. Through our talks about speed, they learned ways to increase speed without changing the track (i.e. taping weights to their cars). But this could also be a negative if the weight is too great, so they had to find that perfect balance.

A cool thing I noticed also was that we encountered other interesting aspects to explain to the students through this exercise as we ended up talking about wind resistance and extraneous data. My main goal in this project, however, was to give them a physical way to think of the skateboarder CSDT. We talked about how they could increase the mass in the CSDT as well as changing the angles of the arcs and line segments to increase gravity. So my plan was to get kids excited about the project using toys and to help them understand real world applications for the CSDT. So I can reference that next week when we go back to working with skateboarder.

by Kirk Jalbert

During the development of the RPI sensor in collaboration with the MDL team Chris, Louis and I ran into a rather serious problem over differing opinions on how copyright should be applied to the completed software and hardware. While the Statement of Work stated that the project would be fully open source as seen in the excerpt (below), the MDL administration decided all creations exiting the MDL lab were property of the lab…

3.0 Objectives
We will create a prototype environmental sensor system, using mixtures of proprietary and non-proprietary components.  This sensor system will operate as a generic platform from which a larger infrastructure can be built.  Designers of custom sensor technology, such as Sawyer/Shing and others will then be able to utilize this infrastructure to deploy their technologies into the SOOS sensor community.

The MDL team will be primarily responsible for developing the sensor system/hardware and device interface with the end goal of launching a stable prototype by end of Spring semester 2011.  During this time Gutierrez will be developing the software infrastructure for the SOOS online community. While much of the basic online structure will be built on available Open Source technology, some special-purpose utilities will be developed in partnership with the MDL team to support the sensor device.  The MDL team will also work in collaboration with the oversight team to create supporting documents and educational tools for the community of users (e.g. circuit diagrams, instruction modules, physical layouts and other documents not normally part of open source but critical to the SOOS community).

The crux of the problem was a core disagreement over what “open source” implied as an educational directive in research projects. As far as we were concerned, if the project was developed with open source principles and the parties agreed to the language in the contractual SOW all was well. Various offices across campus, however, each had their own definition of what constituted our intellectual property vs. the inherent right of the school to claim ownership over work done in their facilities. Ultimately, the ruling decision was made by the “Office of Technology Commercialization” that the SOW was indeed a binding contract and the Open Source agreement had to stand.

Interestingly, the MDL administration agreed to this mediation by justifying it as a financial argument:

In this regard, our general policy and stated objective in the Design Lab has been to identify sustainable funding sources for service oriented projects that will facilitate our working for charitable causes in to the future.  As an exception to this general policy and in the interest of promoting entrepreneurial initiatives on the part of the Navajo and Ghanaian people, I’m proposing that we (i.e., RPI and the Design Lab) do not claim IP protection on either of these two projects.

Nevertheless, this is a far more complicated story than I can tell in this blog entry. If you’d like to see the full version go see the 2-part presentation I gave to the Rensselaer Center for Open Source in July of 2011 here:

Back to Culturally Situated Community Sensing Project Main Page »

by Kirk Jalbert

After much effort negotiating open copyright with the MDL administration in Spring of 2011, the RPI sensor team was ultimately faced with a more difficult challenge…their technology didn’t work. This may be only a partial truth – certain parts of the MDL sensor design worked according to specifications, but many aspects of the hardware and software were riddled with bugs and design curves we simply didn’t have the resources to sort out in time for our field tests in Summer of 2011. Since the MDL built their platform on an Arduino architecture, we were able to switch the core processing components over to off the shelf hardware and build around this accordingly. In the end we came up with a hybrid solution to bring into the field.

The end device was capable of sensing volatile organic compounds (VOCs), carbon monoxide (CO), temperature, relative humidity, and soil moisture. Louis and Chris also tried to get a dust particle sensor working to no avail, but it was a valiant effort. To be truthful, the VOC and CO sensors didn’t work to our satisfaction in this first model, but it did give us relative values for some educational comparison purposes. You can read the full user manual for the sensor device here.

Here are some photos of the RPI Sensor model 1.0…

Photos credit: Kirk Jalbert

Back to Culturally Situated Community Sensing Project Main Page »