Remote Robotic Labs and Smart Science® Explorations

Recently, someone asked about the differences between remote robotic labs and the Smart Science® exploration online hands-on labs. This question may arise in the minds of many.

The various remote robotic labs (RRL), including MIT's iLab, are different in some important aspects. Because our approach is so different, educators often have trouble understanding the differences compared to approaches with which they may be familiar.  This explanation should help to clear up any questions.

A RRL provides data automatically. You set up your parameters, whatever those may be, and push a virtual button. Often, you see nothing transpire at the remote site. After a brief pause, you're handed a sheaf of data electronically. For advanced students, that may be just fine, but for ordinary students, all of the trouble of setting up the RRL has been wasted. You might as well have stored the data from yesterday (or last year) along with any imagery and provided that. In that event, you could have just provided this information locally. The students wouldn't know the difference and probably wouldn't even care.

RRLs have limited range. They cannot do Sordaria crossing over or seed germination experiments. You can imagine doing tides, but the real-time aspect is lost because students are not there in real time the entire time that data are being captured. And so it goes. You cannot base an entire biology or chemistry course on just RRLs.

RRLs have limited access. If you attempt to scale RRLs, you must have more pieces of expensive or unique equipment. Depending on the precise experiment being run, the time that the machine is available controls how many students can use it during a given hour-long period. It's not unlimited. You know that you cannot deliver to a million students per hour and probably not even to a thousand.

Our approach takes the online hands-on lab (OHOL) path. We toss out the pretense of real-time experiments. (I say pretense because there's always a delay between data capture and arrival at the student workstation.) In its place, we open up entire new vistas of learning science.

The OHOL way does not deliver data automatically. Students truly must interact to take their own data. As in the tides example, those data are different for different students doing the same experiment with the same parameters.

With OHOL, you have a visual experience. With tides, you watch the actual tides and then measure them yourself.

An OHOL can be created for any experiment you can record on video and take data from. The data may be quantitative, semi-quantitative, or qualitative. They are your data, not those of a machine. The experiment videos may be from a high-speed camera or from a time-lapse camera. They may even combine multiple cameras as with the shadows lab where one camera follows the Sun with a fish-eye lens and the other tracks the path of a shadow.

What do OHOLs and RRLs have in common? None of the data are invented. They all come from the real world. The various forms of real wet labs also have this feature. However, only manual wet labs and OHOLs are truly hands-on in the sense that you take your own data point by point.

Our technology allows for an unlimited number of scenarios. We're only limited by our imagination and our resources. We have done as many as 100 experiments to create one lab. The number of experiments available is also a function of the pedagogy. Students can be confused by having 30 experiments available. Some will think that they must do every one rather than exercise judgment (actually think) despite our telling them otherwise. It becomes the instructor's task to handle this issue because instructors control grades, and students who do every single experiment available are doing so because they think they'll improve their grades. The instructor must convince them that lack of thought will reduce their grades. Our best efforts cannot do so because we do not hand out grades.

There's much more to this picture. For example, we insist on students making predictions before beginning experiments. We provide introductory (pre-lab or formative) assessments and summary (post-lab or summative) assessments. We provide extensive background resources and an online lab report that can be customized for your classes.

The above is not to say that RRLs have no value. On the contrary they are the go-to labs of the future for college engineering courses. They open up the use of expensive equipment that many schools cannot afford to undergraduate engineering students. They have limited use for college science courses. The limitations are those of the medium that requires complete automation and relatively quick experiment completion. They're of little value in K-12 education. You can find better ways to learn any science concept at that level, with the possible exception of advanced or honors courses and then, as with college science, only with a very few investigations.

© 2013 by Smart Science Education Inc., U.S.A. www.smartscience.net
Follow this author on ETC Journal

MIT's iLabs are Great -- or Are They?

Sometimes a great concept just arrives too early or too late. I'm looking at the MIT iLabs project. NSF has kicked in $1 million to make it work for non-MIT students, i.e. regular students. The interface currently in place is too difficult for those below genius IQ to master.

However, I'm not writing to criticize the iLabs interface. I'd like to think really hard about exactly what they're doing and what Kemi Jona would like to do with the NSF money. He talks quite glibly about creating an eBay of online real-time programmable labs.

The iLab requires remotely programmable equipment with the ability to put results on an Internet link. That fact limits the range of experiments possible. Such equipment necessarily costs lots of money. Few schools will have that sort of equipment available to share.

Furthermore, the student will see the equipment as a black box and must have lots of additional instruction to appreciate fully the nature of the experiments being done. The information coming back from the equipment (as currently structured) is a string of numbers, not very exciting to the average student.

I see little chance that the iLab concept will expand to cover much of science education. If it remains viable, it may be a great experience for some students as a part of their science classes.

Consider that each time an iLab experiment is performed, all of the information becomes digitized before being transmitted. This information could be archived on a server database and provided to others on demand. Such a scheme would allow greater use of the equipment because if someone requests the same identical experiment, it will be immediately available from the database.

If some object to the repetitive nature of this scheme, you can readily record the same experiment several time to allow for normal experimental variation and chose the one for replay randomly. Take that concept one step further record all of the experiments that students might request. Then, the expensive equipment must be used only for a short period of time, rented if you will. The cost and feasibility of the entire operation goes way down and the likelihood of success goes way up.

You can also provide additional information in the digital feed such as images of the equipment while operating, images of the inside of equipment, and so on.

Moreover, you can embed the experience in a full learning scaffold so that students are forced to think about the experiment, must make predictions and analyze results. It can include post-lab assessments and online lab reports as well as substantial supporting materials.

Once you've created the system to store and deliver these experiments along with the learning support, there's no reason to limit the experiments performed to just those that can be run on programmable apparatus. After all, the programmable apparatus was only used so that experiments could be run on demand. With some clever video techniques and highly interactive software that allows students to collect their own personal data, you could cover all science areas that involve experiments and data collection.

Now, you have all of the benefits of the iLabs without the great expense or the problems associated with running an eBay-like facility for schools. You also have a much greater range of science that can be done. You have to wonder why the iLab people aren't proposing this marvelous extension of the iLabs idea when the technology to do it clearly exists.

Perhaps, it's because it's already been done - ten years ago!

Just take a look at www.smartscience.net.

© 2015 by Smart Science Education Inc., U.S.A. www.smartscience.net
Follow this author on ETC Journal.