Common Knowledge

Social Applications with a purpose

Fred Stutzman blogged earlier this week in reaction to Peter Brantley's comments regarding what the success of Facebook tells us about where education is going in the future.  I think both of them have it right when they observe that Facebook has/will alter the way the current college/high school generation communicates well beyond their time as students.  It's like they say, "The only thing harder than learning how to use an application is learning how to unuse it."  (I have been painfully reminded of this everyday since under duress I was forced to adopt Office 2007 and unlearn an interface I had been using for 15+ years.)   I also agree with Stutzman and Brantley that the deep and rich communication/social interaction process that goes on inside Facebook indicates that there is a tremendous opportunity to rewrite the rules for how education works.  These tools show how it may be possible for peer to peer education to become as important if not more so than the traditional top-down models.  (Such a solution is coming just in time -- with the glut of information that exists and that is growing exponentially, we need, as I have discussed in the past the kind of "massively parallel processing" that can only come from a peer to peer model.)  However, and I think this is part of what Stutzman is getting at, there is still a lot of work to be done.  By and large the activity on Facebook is fun because it is about the ultimate kind of fun -- interacting with friends.  On the other hand, learning is a "purposeful" activity that, while deeply satisfying, is not "fun" in quite the same way.  What needs to happen now is for people to begin to experiment with creating "well-designed" tools that make the purposeful activity of education, easier, faster, and more rewarding.  At Carmun, we are working on a subset of this problem -- specifically related to the typical research and paper writing activities that high school, college, grad students, professors, etc. go through.  In the couple months that we have been live, we have already learned a lot and know that there is a lot more to learn.  Facebook and others have shown the promise -- now it's time to do the hardwork of finding the answers.

Back to the future

These days it seems like the chorus for making education a social process keeps getting larger and larger -- David Weinberger says things I think here but more eloquently.  Even better, his post introduced me to Ellen Hoffman, another like minded thinker.  Both raise the question of what the role of the teacher will be in the upcoming educational revolution, which I have started to noodle on myself. I think the simple way to think about all of this is to go back to one of the hoariest of all educational saws -- "Give a man a fish and he will eat for a day; teach a man to fish and he will eat for a lifetime"  In the new world, I think teaching becomes less an less about imparting "facts" and more about turning us into auto-didacts.  Teach me a fact and I will have learned something -- show me how to learn and I will keep doing it for the rest of my life.

Take it to 11

Larry Sanger published an article earlier this year proposing a new model for the "university" in which a students "shoppped" for teachers in an open marketplace (think eBay or Craig's List).  A radical notion no doubt -- but often when I read visions of the future of education like this, I wonder if they are radical enough.  Is it possible that the Internet will actually render the concept of a "degree" irrelevant?  In the past, we needed degrees as a way of overcoming information disparity.  Without a piece of paper from Harvard saying you were an economist, how could I ever know?  Now, we leave digital tracks all over the place and if we participate in communities (eBay again, Slashdot), we develop reputations that are easily discoverable.  Google, LinkedIn, Facebook, blogs tell me more about someone's qualifications than a piece of paper ever could.  In a world like this, who needs an institution to certify someone's credential?  That is taking the notion to 11.

Helpful Irrelevance

It’s hard not to be amazed at the power of search to deliver relevant results.    It seems that, no matter how obscure one aims, search engines return multiple Web pages related to the query.  Five years ago, this phenomena was so striking that word nerds like myself used to play a game called Googlewhacking – where you attempted to pose a search query that would return one and only one result.

However, relevance is one thing – and helpfulness another.  When you are doing research, one hundred relevant sources that have overlapping information are generally less helpful than two sources that complement one another in a meaningful way.  Helpfulness is a human concept that is hard for machines to judge with effectiveness with which they determine relevance. 

The difference can be seen in another form in alternative approaches that computers and humans use when playing chess.   Computers employ a brute force approach in which they analyze the consequences of all possible moves several generations ahead.  Humans, on the other hand, look fewer moves ahead, and instead rely on pattern recognition for a given arrangement of pieces.  As chess is, relatively speaking bounded (albeit with a high degree of complexity), computers have been consistently improving their ability to beat human opponents.  However, the complexity of all the possible ideas that might be connected is well beyond the ability of today’s, and perhaps even tomorrow’s, machines to compute.

All of this means that for computers to be truly helpful they need to return relevance and complementary irrelevance, which requires human pattern recognition, not computational ability. 

Put another way, the intersection between technology and education needs to be the creation of community as much as the indexing of information.

Massively Parallel Processing: Academic Collaboration in an Age of Information Overload

For a long time, the iconic image of the “hard work” of research has been the solitary scholar tucked into a library cubicle, cut off from the rest of the world.  With this as the paradigm, the idea of students sharing research ideas verges on heresy.  If someone tells you what sources to use, what lessons are learned?  Where has all the hard work gone

The problem with this model is that it depends on a number of assumptions that just don’t hold true anymore.  First, libraries were once bounded institutions. Combing through a library’s “entire” collection on a subject may have been time-consuming, hard work, but it was do-able.  Second, there were “experts,” professors and librarians, that “knew” the collection and could guide students’ research, while still leaving plenty of room for the hard, solitary work of research.

With the advent of the Internet and the digitization of academic content, these two fundamental assumptions no longer hold – Prometheus has been unbound.  Not only is there more information availablethan any single person, even an “expert”, could legitimately navigate, but it is growing at an exponential rate.  As a result, the research paradigm has to shift. Sharing and collaboration will be critical to solving the problems of information overload.    Now instead of teaching students how to work on their own – we need to teach them how to work together.  Students need to learn the social skills and judgment necessary to building trusted networks of “collaborators”.  This will also make it necessary for them to learn the importance of giving back. 

Parallel processing only works when the nodes on the network produce as well as consume information.  The new paradigm will not eliminate the importance of students engaging with sources individually – on the contrary, it calls for a balancing of the social and a-social to collectively harness the power of the massive amount of information that now lies only a few keystrokes away.

Pedagogy and curriculum in a 2.0 World

As recently as 10 years ago, textbooks really were the first and last words in the curriculum. What a textbook said, and in what order, drove almost entirely the way people learned about a particular subject – from reading to math to health to science, etc. Under this framework, curriculum and pedagogy made learning a very controlled journey. A geology textbook would decide whether you learned about sedimentary rocks or igneous rocks first, and when to show you a picture of the Grand Canyon and when to show you a picture of an erupting volcano.

Things have loosened since then. CD-ROMs and Web sites produced by educational publishers provided learners with the opportunity to bounce around a subject – but only within the boundaries defined ahead of time by the publishers. Now, we live in a hybrid world where textbooks and interactive media are packaged together. But, the value they attempt to provide is much the same as it was ten years ago – a defined and architected approach that directs students through material.

It seems to me the Web is poised to blow this wide open. More and more high quality content is moving to the Web and more people are combining and presenting it in unique and varying ways. All of this new content is just a search a way. In this world, people won’t stay inside any single bounded presentation of a subject. They will read something. It will spark a thought. They will type that thought into the search box and be off on their own journey – not one that was programmed for them.

This radically changes the way we need to think about curriculum and pedagogy. Now, it is not so much that we need to teach people a subject. We need to teach them to search. We need to make it easier for them to evaluate what they find. And finally, we need to give them the tools to collect and knit together this unbounded journey into their own unique and individual curriculum.

“The tide of web 2.0 use has flooded over the campus walls an it’s too late to stop it, so embrace it. There comes a point where bottom up becomes so compelling that it becomes the new top-down.” Donald Clark

From Bacon to Bakhtin

Will Smith introduced the concept, and Kevin Bacon made it fun.  As the theory has it, all people on the planet can be connected through a relationship chain consisting of no more than 6 people.  In the case of Kevin Bacon, it turns out that no living actor in the English speaking world has a “Bacon Number” of greater than 4 – i.e., the chain from any actor back to Kevin consists of at most 3 intervening actors.

As the CEO of a social media site, I have been reading more formally on the subject of the interconnectedness of people – especially the book, Small Worlds by Prof. Duncan Watts.  It is good and hard, thought-provoking stuff.  Networks of people, even actors, are fascinating – but I want to know about the networks of ideas.

Right now major classifications schemes like the Library of Congress are organized hierarchically and not as an interconnected mesh of ideas.   As Peter Suber has written, some people are looking at using a network approach to measure the “quality” of journals based on how often they are cited by other journals.  Google of course already does something similar in its search for relevance with the Web and is trying it with scholarly publications as well.  But, where I think this all gets really interesting is when we can start developing a network mapping of ideas and writings that at first glance may be unconnected.  Where people can easily discover and take advantage of how people all over the world connect ideas together.  In a paper I once wrote, for example, I connected the computer scientist Douglas Hofstadter and medieval theologian St. Augustine.  Not obvious – but people are doing interesting things like this all the time.  The Internet will be the place where we can finally create a practical environment to explore these kinds of connections.  Then we can finally discover how many steps it takes to get from Bacon to Bakhtin.

Credit where credit is due

Last week I blogged about some of the barriers that I think are currently limiting the amount of academic material originating on the Web. The biggest challenge that I saw was people’s risk aversion when it comes to consuming academic content. One of my co-workers, Jessie Goldhirsh, pointed out that this may not be the only challenge. In the academic world, receiving proper credit for your ideas is one of the most important forms of currency. In the Internet world, on the other hand, ideas often get cut off from their originators very quickly. It’s not that Internet culture is inherently incompatible with the idea of “credit” – it’s just not a current norm. Nonetheless, before the Web can become an ideal environment for the “production” and “distribution” of academic content, meaningful approaches that handle the “credit” issue will have to emerge.

I think the issue of credit has some interesting implications for some of the discussions going on in the information community about bibliographic control, as discussed here in Nancy Fallgren’s summary of presentations by Karen Markey  and Timothy Burke . Each make interesting points about the importance of the bibliographic record from the end user perspective as a “search tool” and from the librarian’s perspective of managing a collection. 

But, I think there is an additional dimension here related to the issue of credit because the bibliographic record contains all the information necessary for proper source attribution. Right now, it is often static data, and as such it is very un-Web like. But somehow, I think the kernel to solving the credit problem will be intimately tied into the use of bibliographic records.

 

The Means of Production

Even after years of studying medieval Celtic literature, sometimes I still think like an economist (my undergraduate major). And, in some circumstances, I think the dismal science actually provides a useful way of looking at things. In particular, the economist’s perspective helps me to understand the impact that the Internet is going to have on the production function for ideas.

People have long since noticed that the Internet has lowered the “distribution” cost for ideas – bits are cheaper than paper and ink. As a result, we are in the midst of an explosion in the production and consumption of ideas in lots of idea categories (e.g., entertainment gossip, investing, dating, etc.): via email, blogs, message boards, and others means.

However, with respect to scholarly ideas, it’s not clear to me that we have seen an equivalent shift in production. I think this has a lot to do with people’s “risk aversion” when it comes to “consuming” ideas. If you use poor quality information in a paper, you risk a bad grade or missing tenure – let alone losing an argument. Thus, most of us continue to rely on published works (journals or books), where we use brand and the cost of production as evidence that there are good goods inside.

The challenge for us working at the intersection of the Internet and education/learning is to find a way to meet these legitimate issues. I think key pieces to the solution are starting to emerge, but still need to be honed further:

• First, the wisdom of the crowds, e.g., Digg and Wikipedia. Lots and lots of people commenting on an idea creates a Darwinian environment in which only the good ideas survive.
• Second, implicit metadata, e.g., del.icio.us, provides context for the well from which ideas spring.
• Third, idealets, e.g., message boards, blogs, etc., where we can “consume” lots of ideas without investing a lot of time.
• Fourth, social networks, where we can connect with and engage the people behind a given ideas Somewhere in here is the secret sauce for radically altering the production of scholarly ideas.

I can’t wait to see what happens.

Finding Jerry

As the founder and CEO of a start-up, I read with a fair amount of sympathy a recent post by Steve Fisher on VentureFiles  reacting to Stu Philips’s blog about VCs saying “No”. Hearing “No” from a VC is one of the hardest things about being an entrepreneur. Carmun can only be the amazing business that it has the potential to be, if I can find people willing to fund our growth. That having been said, I have always felt good about the process that has gotten me to “Yes” with the folks that have invested thus far and “No” with those that haven’t.

It seems to me that VCs have a real challenge in wading through all the potential deals they see. Although I have never been a VC myself, I was a senior deal maker at AOL for seven years. During that time, I worked with a lot of companies both big and small that eagerly wanted a deal with AOL. Figuring out whether to say “Yes” or “No” was more challenging than you might think. The issue for me was never the idea itself but whether or not I thought the partner could execute on it. Determining who could do the best job was certainly more of an art than a science.

I suspect the same is true for VCs. Unfortunately, my sense is that radical innovation is rarely going to be present in such a way to make decisions “easy”. Even looking at recent consumer Web successes like YouTube, Facebook or MySpace. Not one of them succeeded with an idea that in some way had never been hitherto fore heard of. Just ask, iFilm, CitySearch, or Geocities. Without considering the ideas that truly make zero sense, it seems that VCs are in exactly the same place we were at AOL – trying to make “rational” decisions among “similar” investment opportunities when there are few “rational” criteria to apply.

I think it all comes down to finding a great relationship. Entrepreneurs need to find the right VCs and vice versa -- connections that leave each side feeling like “You had me at hello.” You want people that “Get” the idea intuitively and “Get” you as an entrepreneur. I know that Carmun is going to be a huge business and make a lot money for a lot of people. However, I know the ride will not always be smooth. During those sideways moments, I want, and I think every entrepreneur should want, an investment partner whose belief in the business and in me is so strong that they help me solve my problems rather than adding to them. When I was at AOL and I found myself in deals that either I or the partner never truly believed in, the moment the first issue cropped up everybody got a serious case of “Buyer’s Remorse” and started trying to figure out how to cut their losses. On the other hand, when problems arose in situations where the personal and business commitment was strong, like with Mark Rosenthal at MTV, or Megan Smith, at PlanetOut, or Andy Sturner at Sportsline, we all pulled together and the business emerged intact, and usually better.

As much as I need money for my business, I really need a partner. I don’t underestimate the challenge most people will have in trying to figure out whether Carmun is right for them or not. As a result, I don’t get discouraged or frustrated with the process. I know the right relationship is out there.