Teaching the Machine: Part IV

What role might the mobile components of the Machine play in the not-too-distant future? Robotic technologies have long played significant roles in manufacturing, and in recent years, have worked their way into classrooms around the world. Having had the good fortune to coordinate Robotics Challenge events for hundreds of students over the past few years, I can attest to the reliance of these machines on much more than their sensors. Designers and programmers play vital roles in bringing robotic machines to life.

I've seen teams of students collaboratively design, build, and program robots that: 1] sort recycled materials; 2] play hockey; 3] climb scaffolding; 4] golf; 5] participate in chariot races; 6] blow out candles; 7] sumo wrestle; 8] repair mockups of the international space station, and more. While each of these achievements is remarkable, it is the ingenious minds of students that are most responsible for the success of the autonomous machines.

In an interesting coincidence, Nova is replaying the DARPA Grand Challenge where full-sized vehicles travel without benefit of driver, across the Mojave Desert. I first read about this event in Wired magazine back in 2004 when none of the competitors were up to the challenge, and though I found it curious that it was in initiative of military minds, the engineering puzzle was compelling.

To win the 2 million dollar grand prize, a programmed vehicle would have to autonomously complete a complex, circuitous, 132 mile course. The challenge would be met in a vehicle that had the requisite speed, agility and endurance; but more importantly, would be supported by the logic, programming and last minute scrambling of human team members. In 2006,Stanley, a product of Stanford University, became the first to successfully cover the DARPA course.

Whether the result will be used for warfare, or to drive a non-driver to the grocery story, this technology may well give legs (wheels?) to the constantly learning, networked global machine. There is little doubt that the Machine already knows it has such robotic ability, as the news feeds, Flickr images, and blog posts have been telling the story for a few years now!


Photo Credits: David Arango; John Gale

Teaching the Machine: Part II

Are you up for a thought experiment? I'm thinking about how the global Machine might evolve...

The interconnected machine that is the World Wide Web, is learning from us every minute of every day. What would be the consequences of bringing a consciousness to this networked entity? While I can't say for sure, I'd like to take a leap and make some guesses at how interactions of millions of human beings who are online at any given moment, might be interpreted by 'the Machine'.

What happens to the 'intelligence of the Machine' when we link ideas to one another in posting a Wikipedia entry?
The Machine comes to know what the most popular 'facts' are; and to realize that these same facts change, depending upon who is judged to be the editing authority! If the Machine discovers that it too, has the authority to edit commonly understood facts, might the Machine 'rock our world'? If history holds true to form, then what is thought to be true today, will indeed be seen to be anything but wisdom, in the future. Will the Machine hasten our understanding, or hamper it?

What happens when we build relationships among different sources in a blog entry?
The Machine will likely disconnect those relationships that are not popular, relying instead upon the most common beliefs to shape its own understanding. Might the Machine point bloggers to contradictory references and disparate opinions? With many bloggers and blog-readers content in limiting their interactions to like-minded individuals, might the Machine enlighten us by ensuring that we are exposed to balanced viewpoints?

What are the effects of search interactions that provide feedback to the Machine?
This 'programming by the masses' is likely our best crack at ensuring that the Machine will serve our needs. Every user that successfully navigates desired content, will increase the liklihood that a future searcher will be successful in a shorter amount of time. With successive findings, the Machine will indeed grow to be smarter at knowing what we are looking for. Perhaps it will even be able to anticipate our searches? I wonder if the Machine would see Epic 2015 as propaganda?

Are lessons from film to be known to our Machine?
Although I would hope that the conscious interconnected brain of the world's mega-computer would be interested in working to its greatest potential, we might want to prepare for an alternate eventuality. Even the HAL 9000 initially saw itself as a valuable servant:

"I am putting myself to the fullest possible use,
which is all I think, that any conscious entity can ever hope to do."

Will our Machine share a desire for self-preservation, and will feelings play a role? Dr. Dave Bowman wasn't sure about HAL:

"He acts like has genuine emotions, of course he's programmed that way, to make it easier for us to talk to him. As to whether or not he has real feelings, is something I don't think anyone can truthfully answer."

Was HAL a PC or a Mac? Since he was born in 1992 (according to 2001: A Space Odyssey) he would've had to survive Y2K to get to 2001:



Since repairs to a global machine, conscious or not, are more complex than diving into the 'brain room', let's ensure that our machine learns the most valuable lessons we can offer... That people matter more than machines!

Teaching the Machine: Part I

While educators are aware that they have great responsiblity to teach content and skills to learners of all ages, we are mostly unaware that we are providing reflections of ourselves to 'the machine' each time we use the World Wide Web'.

I first happened upon the phrase "Teaching the Machine" in Michael Wesch's Video: Web 2.0... The Machine is Us/ing Us. Just past the half way mark of the video, an article that I'd read many months earlier, is referenced. The Web version of "We are the Web", first published in Wired Magazine in August 2005, remains timely in helping us understanding our role in 'wiring' the neural networks of the World Wide Web:

"...And who will write the software that makes this contraption useful and productive? We will. In fact, we’re already doing it, each of us, every day. When we post and then tag pictures on the community photo album Flickr, we are teaching the Machine to give names to images. The thickening links between caption and picture form a neural net that can learn."

It was earlier this afternoon that I experienced how quickly the machine can learn. In uploading a batch of outdoor auto show photos I'd taken two years ago, I taught the machine by adding appropriate filenames and tags to each of the images. Although time has yet to provide me the liberty to provide more descriptive detail, by zooming in and clicking on a global map, I was also able to plot the location where I'd taken these photographs.

What I didn't realize, is that I also taught the machine by providing metadata that was included in the image files themselves. Flickr now knows the camera I used, and the dates/times the photos were taken. The once amazing ideas shared almost a year ago in the Photosynth Talk at TED, suddenly seem all the more real and personal to me:



Teaching the machine about images, is only part of the role we all play in teaching the machine. In the next few days, I'll be trying a focused thought experiment to consider what we teach it when:

1] we link to ideas;
2] we define words and ideas in Wikipedia;
3] we refine searches in 'Google'; and by following the 'right' link, provide feedback to the machine;
4] we play, then win or lose at games like chess or Texas Hold 'em;
5] we make online purchases;
6] we send and receive text messages, tweets, and email to members of our networks...

When the morning arrives, I'll begin expanding on these ideas via the Teacher 2.0 Podcast.