I first met Shai Reshef, the indefatigable force behind University of the People, at DLD in 2009. Since then I have come to know Shai as someone who cares deeply about using education to empower people all around the world. He is taking quite a different approach from the MOOCs by scaling UoPeople much more slowly and deliberately and working diligently towards accreditation.
The approach has two premises. First, in many countries around the world a degree plays a dramatic role in changing the income trajectory for individuals. This has also been the insight behind Vittana which provides student loans. Second, Shai is looking to build a self sustaining model. To that end while courses at UoPeople are free, they charge a $10-50 application fee and a $100 exam fee per course. Because of UoPeople’s very lean model they can become self funding with only a few thousand students. From there Shai believes they can grow to tens and eventually hundreds of thousand students globally.
In the meantime of course $100 per course is still a huge amount of money in many parts of the world. Based on my advice, UoPeople has rolled out a micro scholarship program. Here you can give as little as $10 to help a student with his or her exams (there are mostly men because women currently receive a scholarship from HP). So if you are feeling fortunate because of where you are in life, go and help someone get the education to make their life better.
My older son asked me yesterday about how colors on the screen work. We started talking about the history of the RGB color model including early color screen technology. That eventually led to a discussion of how the RGB color space is three dimensional with 256 values on each axis for a total of 2^24 or roughly 16 million colors. Since a display is only two dimensional how would you go about showing the entire color space? From that question we talked about projections (from 3D to 2D) but also wrote a bit of code that let’s you slice through the color cube and display that slice on screen.
As we were playing around with that using Khan Academy’s visual programming environment, my son discovered some wonderful color patterns more or less by accident. That led to a discussion of generative art. And as it turns out the Khan Academy environment is based on the Processing library which has been used for years for exciting generative art projects. So I helped my son to take what he had stumbled upon out of there and put it on a web page. With a little more work we figured out how to enable fullscreen mode. You can see the results here.
Talking through all of this provided a great example of how knowledge hangs together and how context provides motivation. Each progressive bit of exploration here was motivated by the territory we were finding ourselves in. It is a stark contrast to our fragmented experience of education in school and even to this day online. As we are creating more and more of atomized learning objects, reassembling them in new ways around these kind of journeys is a tremendous opportunity.
One of our goals for 2012 was to get scuba certified as a family. We took a first run at it in a rushed fashion early in the year but abandoned that effort after a less than pleasant pool dive in Manhattan. In the fall we decided to start the process over and go with a proper PADI certification program. We found a dive center in Playa Hermosa in Costa Rica called Diving Safaris and structured our vacation around that.
Instead of wasting time on location with theory, we took the theory classes ahead of time online. The PADI e-learning course was quite well done. It consisted of short segments with lots of illustrations with both a recorded voice and a text you could read (Susan preferred the voice, I preferred reading the text). The short segments were grouped into subsections and then sections. Each subsection came with immediate review questions and then each section had a full on assessment. You had to pass each section assessment to qualify for the overall theory assessment (in fact, you can’t even access the overall assessment until you have passed the sections).
When we arrived at Diving Safaris, they conducted a brief review of the most important theory — this may be a test to make sure that people didn’t get others to take the online test for them. But other than that we went straight to the practical hands on equipment handling. Having done all the theory up front allowed the instructors to spend more time with us individually on equipment handling and underwater skills.
I had been thinking for some time that schools should switch much more aggressively to the inverted classroom model in which students learn the material at home (from videos, texts, etc) and apply it in the classroom. This frees the teacher up to see where the students have individual difficulties applying the concepts that they are supposed to have learned. Given our PADI experience I am now a complete convert.
I am convinced this is an amazing model and should be used much more broadly. It’s especially powerful if the application of the theory is as lively as our open water dives, although I realize that’s a high bar, especially because on our third dive we got to see seven whitetip reef sharks!
One of the reasons that we decided to move to Chelsea was the proximity to the USV office. The other was that our kids can walk to Avenues, a new private school with an ambitious program of building campuses in leading cities around the world. Another distinguishing factor of Avenues has been their embrace of technology which got a good writeup in today’s Wall Street Journal. Our kids each have both an iPad and a MacBook Air from school and use both of them heavily across a variety of classes.
There is one important missing component though so far and that is learning to program. That of course has been the subject of my Tech Tuesdays and I have in the past promoted Scratch as a way for kids to learn programming. In that post I wrote that “the use of Scratch can and should be pervasive throughout instruction rather than being something taught separately.” Here is just a short set of ideas for how to do that in different classes:
- History - changing maps over time; graphical relationships between concepts and people; animated historical timelines
- English - create word games; animated six word biography; create your own scene from a drama (“enter stage left”)
- Music - create electronic compositions; visualize sound and music
- Science - simulate experiments; graph the results from experiments; safely explode things
- Math - illustrate the number line; create math games; show the relation between algebra and geometry
Having spent more time writing about programming and also talking to my kids about it, I have become even more convinced of the importance of integrating it into other classes. The reason is that programming provides an exceptional way of learning a concept. It is reminiscent of the saying that you haven’t really learned anything unless you have taught it several times. Programming is “teaching” the computer how to do something. If you can’t teach it to the computer you have probably not completely understood it. Hence the “programming to learn” in the subject line of this post.
For instance, our kids are currently learning about identities and inverses in Math. If they were simultaneously learning how to write a program that kind find the inverse of a number for either addition or multiplication, I am convinced it would give them a much deeper understanding.
Thankfully the team running Avenues is receptive to this idea. I will be spending time at the school in early December to meet with several of the instructors to talk about Scratch and other ways to integrate programming into the learning experience. It is something I very much look forward to.
What this will change is the opportunities for the 50-year-old laid-off engineer. He could now show that he still has the analytical skills and brain plasticity to work alongside 22-year-old college grads in a 21st-century job. It would allow the smart, young, single-mother to re-engage with a real career. It would allow recent college graduates to prove they have skills beyond what can be gleaned from their majors and grade-point averages.
This unbundling of credentialing is also one of the hypotheses we have for investing in education at USV (see Christina’s blog post). We have seen the Internet bring dramatic unbundling to newspapers which historically pulled together news, commentary, sports scores, classified ads and more and expect to see a similar trend in higher education albeit with slower adoption due to regulation.
But unbundling generally doesn’t mean simply splitting up the pieces. It often means coming up with something different. For instance, when newspapers were unbundled breaking news went in part to Twitter which is unlike anything that existed before. Similarly, classified ads went in part to Craigslist, but job classified went in part to Indeed which presented a completely different model of classified by replacing the “publish” model with a search model.
So unbundled credentialing will likely come in forms that are quite different from simply having separate tests. For instance, github, StackOverflow and Behance are all examples of a broader view of unbundled credentialing. They all provide ways for individuals to demonstrate knowledge and have a community evaluate the contributions. What is appealing about these types of peer-produced credentialing systems is that they make it much harder to cheat than most traditional test taking as reputation is built over time.
There are also several startups targeting credentialing, such as Degreed and LearningJar. Mozilla also has something going there with their OpenBadges and it will be interesting to see what EdX, Udacity and Coursera will provide here.
Today’s Feature Friday is not about a single feature but rather about initially prioritizing user engagement and growth over other features and in particular over discovery and analytics. The idea here is simple but still often ignored. If you have engaged users then their activity will generate the data you need for discovery and analytics. In some ways this is a corollary to the idea that lots of data combined with simple algorithms tends to beat more sophisticated algorithms run over less data.
A company in our portfolio that has done this particularly well is Edmodo. Their early focus was entirely on solving a seemingly teacher-student communication problem by giving teachers an easy way to share content with their classroom. They added other features to that to ensure that teachers and students had a reason to come back to Edmodo frequently. The result of this utilitarian focus has has been rapid growth and increased engagement.
With its latest release, Edmodo is now unleashing the discovery and analytics that flow from their massive overall usage. They are calling these features Connections and Insights. Connections uses the data on what teachers are sharing with their students to make it easier to discover high quality content and other teachers. The beauty here is that any act of sharing by a teacher provides an implicit recommendation for that piece of content. Having access to many millions of shares means that Edmodo doesn’t need to ask teachers to explicitly rate pieces of content. Insights on the other hand does rely on explicit feedback from students (and teachers) to provide a kind of “pulse” for the classroom. Here too engagement is a critically important enabler: it is easy to ask a student to provide an emotional reaction if all it takes is one click or touch after an action they have already taken. Asking for feedback in the absence of engagement would lead to a much smaller and more likely biased response.
Both content discovery and classroom analytics have been the goals of a lot of prior technology efforts in education. Most of these have fallen flat because they were premised on static content repositories and on feedback disconnected from action. I am excited to see Edmodo approach these problems in a net native way. If they succeed here then I believe it will form the foundation for tackling the holy grails of education: measuring effectiveness and powering adaptive learning.
For a couple of years now I have been complaining to folks in the Harvard administration that the school is woefully behind when it comes to embracing the Internet. So I was thrilled to see the announcement of the edX initiative with MIT yesterday. The two combined with a $60 million commitment will make a formidable force in higher education on the Internet. For people everywhere who want to learn there is an exciting competition starting between some formidable institutions with Coursera (Princeton, Stanford, Michigan, Penn) and Udacity (unaffliated) also in the running.
And now for the real challenge for both MIT and Harvard (other than actually launching): how to integrate edX back into the schools themselves. That will make all the difference not just for the success of edX but also for the experience of attending one of the schools. If students on campus are connected to the world through edX and vice versa, then the two can serve to enhance each other. Part of what this means in my view is (eventually) abandoning the idea of four years on campus. By that I don’t mean junior year abroad, but something more radical. Maybe as far as spending one year at a time or even less on campus.
The opportunities here are enormous. I am excited to see where it goes!
I love it when a plan comes together. The plan in question is the Edmodo Platform. When we invested in Edmodo in the fall of 2010, the company was just beginning to really grow its network of students and teachers. One frequently asked question was how Edmodo would be able to keep offering its features for free. I remember going to the ISTE conference in 2011 and having several teachers come up to me concerned that they would have to stop using Edmodo once it would start to charge. I assured them that Edmodo would always stay free and that the company would find other ways to make money.
Well, the team has done an amazing job both building out the network, which now has over 6 million members, and also laying the groundwork for the Edmodo Platform. That involved a ton of development work on the APIs which allow third-party applications to connect to Edmodo’s features and education social graph. It also required building out partnerships with third party developers. The current soft launch of the platform includes over 35 partners that have built applications. Some of these apps will be free and others paid and that will be how Edmodo can maintain the network.
Nic Borg, Edmodo’s co-founder and CEO, announced the platform at SXSWedu where Edmodo hosted the closing BBQ. You can find coverage of the Edmodo Platform at Forbes, Fastcompany and Mashable. I am super excited about this as it will allow developers to reach the classroom directly without the need to sell to schools or districts or states.
Andy has a great post over at the USV blog announcing our investment in Codecademy. Knowing how computers work and how to program them I believe will be quintessential knowledge for everyone. It is shocking to me that I had to enroll my kids separately to have them learn typing. While they have learned a bit of Scratch in school, it has been an afterthought as opposed to being integral to their curriculum.
That’s why I am so excited to see the emergence of so many resources and services that completely bypass schools and let everyone learn these essentials directly. There are many wonderful in-person classes on Skillshare and with Codecademy there is now a great start on self-paced and highly interactive learning. For people who are more advanced there is Stanford’s AI class.
So everyone go and learn. Together, we will program the future!
Couldn’t resist an attention grabbing headline here after spending yesterday morning with my daughter on the kind of homework that just completely takes the fun out of learning. In geography her class is learning about latitude and longitude. As homework they got a work sheet that made them look up about 15 cities in the US based on Lat/Long and write down the Lat/Long for another 15 based on a map in their social studies book (a ridiculously heavy tome by Pearson). It was an incredibly dull and completely unnatural exercise with the net result that my daughter now associates lat/long with tedium. As a sailor and a fan of Dava Sobel’s “Longitude” I find that incredibly upsetting.
What could have been done differently? In an ideal world the teacher would have gotten the kids excited about Lat/Long with stories of sailors and pirates and being lost at sea. Seeing how it was “International Talk Like a Pirate Day” that would have been the angle I might have gone with. But even short of that there are dozens of ways of making the particular homework way more fun. For instance, the lat/longs to look up could have come from a famous journey of exploration or from the concert tour of the Rolling Stones. Or the first letters of the list of cities could have formed a secret message. Or they could have used Google maps or earth. Anything other than a random list of cities would have been a huge improvement.
Just like in business where it is much harder to win back a customer that had a bad experience, it is now that much harder to get my daughter excited about Lat/Long. It’s annoying to think for how many other topics this is likely to be the case.