If you want to see my photos on Flickr, here they are.


I was struck by this instance of the Twitter “similar to” algorithm working so well for something as subjective and qualitative as comedy. I have to say, I do consider these 5 guys to be similar to Carl Reiner.

The Elliott Masie lead conference Learning Directions was a great day of thought provoking conversation. Raw notes below.

  • Learning Personalization
    • Biggest change happing in learning
    • Like TV, choice leads to on demand consumption
    • Good question “If you had to learn something for work, how would you want to learn?”
    • The predominant filter applied to learning is “teach me what I want to know that I will use or be held accountable for”
      • People don’t want to learn what they think they know or what they could look up
  • It has been difficult for educators to personalize curriculum but now the learner is able to personalize for himself
    • Matt thought: The popularity of the myth of the charismatic, transformative teacher who magically connects with students seems to be a response to the deadening experience impersonal curriculum
  • Compliance driven learning is extremely depersonalized and doesn’t increase anyone’s performance while giving a bad rap to elearning – and easy improvement would be to let people test out
  • What is being published on the supply side is pretty much the same, but consumption has changed – dramatic increase in the learners watching short duration video
  • Learning systems aren’t changing to take advantage of the online environment. Learning systems give you only 3 data points: Did they take the course? Did they finish? Did they pass? And those 3 data points tell you nothing about how optimal the learning experience is. However online learning can give you thousands of data points
    • For MVA “did they come back for more” seems to me to be the biggest reflection on the quality of the learning experience for the visitor
      • If we could get a really well tuned point system we could get some very good data about how optimal the user experience is
  • Question from audience – does increased personalization lead to a loss of community?
    • Enable a shift from individual learners to team units of learners
      • As a species we seem to need cognitive practice with other people
      • Think of the experience of an audience at a theater – each individual has their own experience but that experience feeds off and is shaped by the crowd around them
  • The learner can’t be left in an ambiguous environment – need to have input about how effectively he or she is mastering material “am I getting to competence?”
    • Don’t take away the freedom to fail on the way
  • Industry knows how to charge for cost recovery, but can’t get to pricing based on relationships
  • Learners who are affectively involved (who feel the rush when curiosity meets knowledge) will collaborate to completion
    • We need to synthesize until we can articulate – cognitive rehearsal, accomplishment and transfer
    • Trust as a competency – how do you design for authentic trust?
    • Post break looking at survey
      • Total spend on learning against compensation is a very interesting metric for a company
      • Even though it is interesting and widely practiced, educators are not trained to provide the support structure around “on the job training”
      • Mobile
        • Question to ask is “what does a device afford us to do that we couldn’t do before?”
          • The slide with the r verbs is a great list
  • Matt thought: Social and personal are not on his mind map for mobile, but in terms of learning and freedom to be less than perfect (freedom to fail) it’s interesting to consider how the Real Self community http://www.realself.com/ uses mobile to research “whisper topics”
    • Also, what would instagram for learning or screengrab galleries on MVA look like? Images very easy to consume and share on devices
    • Looking forward to meeting with the Fuse / Socl folks
  • BYOD is better name than “mobile” because it’s not about doing something unique with the device but making sure you can do what you want on as many devices as possible
    • Matt’s version: Write great data and service layers that flow to flexible presentation
    • Dynamic curating
      • Can be a low cost / high yield way to experiment
        • Rather than authoring, the learning is ranking, annotating, archiving etc
  • Curate.masie.com they are using scoop.it
    • How to get SME’s to curate?
  • Mark thought: Student playlists would be good (vis 90 days to certification genesis)
  • Post lunch warm up
    • Globalized – trend is to create dynamics that will be the same everywhere with permission to change for local needs
    • Curriculum – content people need to articulate it like a map that includes the level of knowledge expected at each point: memorize it? Get the framework only? Reference level knowledge? Etc
      • And build badges you get along the way rather than as a big bubble at the end
      • eBooks
        • Putting forward the idea of an L book: a container of learning content that supports a sharepoint like set of scenarios (including learner tracking)
          • Digital magazines pushing the form
          • Elementary science books too
  • For MOOCs monetization might be in the “textbook” and other assets
  • Look for prototypes at his fall conference
  • Big Learning Data
    • Look for the whitepaper from ASTD
    • It will change the measures we look at
    • Check out the design analysis in the metrics that matter these cats are up to http://knowledgeadvisors.com/
    • How to get ready for the big data asks
      • Be specific about the constituents and what they want
      • Use humor to get at what strategically measures the need
      • Get a data scientist on the team
      • Social and Collaborative learning
        • David and Roger Johnson of U of Minnesota doing good work
        • Don’t confuse social media with social learning (collaborative learning)
        • Collaborative learning = process of leveraging more people to enhance and accelerate learning for the learner
        • If someone is required to teach what they learned to someone else then mastery and transference goes up significantly
          • Matt: recruit volunteer mentors from the students for the good of the mentor
  • Cognitive rehearsal
    • Remediate what you don’t know
    • Putting something in your own words indexes it (ability to retain and transfer)
  • Benefit for the learner is hearing a 2nd voice
    • Powell “Content, context and field truth”
  • You also get better questions from the learners
  • How do we design collaboratively?
    • Need to map to people’s available time and energy
      • Do NOT design a course and attach a collaboration space to it after the fact that you then throw people into
    • Design so social team ups have value to the learner
    • Acknowledge not everyone in a group will pull their own weight
    • At a human level video is a more powerful story telling medium than text
      • Builds trust
      • Gets the subtleties and texture into learning that’s not there in text
    • It has to work for the introverts
    • Matt: How do you put the trolls to work for you?
    • Video
      • Delivers access to experts – learners benefit from multiple points of expertise being available
      • Presence – brings people into environments in non transactional ways?
      • We are missing a way to let learners know that the optimum video for them is there
      • Flipped classroom (Daphne of Coursera referenced her experience of this as well at the CA summit)
      • Video can bring the field / workplace into the classroom

The individuals in the community as point of comparison to the organizations in previous posts. This is still looking at “Who is interested in these people?” and it’s where you see more engagement among the audience. (The other direction that I haven’t looked at yet is identifying representative “students” and analyzing whom they follow.)

Most obvious difference between people who follow people and people who follow organizations is that there are a lot more @ contacts among people followers. Intuitively makes sense, but as always, nice to put some numbers against the idea.

Also as a note, the follower count reports are a great example of data sets where there is a huge difference between average and median.

Jeremey Foster @codefoster

15.7% > 500 followers (301)


3.3% URL > 25% (61)

0.8% RT > 25% (13)


2.4% @ > 25% (42)



Rachel Appel (@RachelAppel)

23.5% > 500 followers (1,260)


3% URL > 25% (159)

0.7% RT > 25% (41)


2.9% > 25% @ (152)



Rick Claus (@RicksterCDN)

48.8% > 500 followers (2,506)


18.6% URLs > 25% (951)

3.2% RT > 25% (163)


6.3% @ > 25% (323)



Veronica Wei Sopher (@Shih_Wei)

51% > 500 followers (2,305)


14.7% URL > 25% (668)

2.3% RT > 25% (101)


10% @ > 25% (448)



This video What it means to be on a MOOC is an awesome artifact of the Coursera class E-Learning and Digital Cultures

Apologies that I can’t get the embed code working here. It’s one smart phone complaining to another about how hard he gets worked now that his owner in taking a MOOC, and it lays out key mobile scenarios for e-learning.




Udacity 8.3% > 500 followers (3636)


Coursera 10.1% > 500 followers (6498)



Udacity 1.1% URL > 25% (477)   0.5% RT > 25% (195)


Coursera 1.1% URL > 25% (741)  0.5 % RT > 25% (314)



Udacity 286 (0.6%) > 25%


Coursera 510 (0.8%) > 25%

Bio clouds



Key things you learn about your audience by analyzing their social networks are context (how they group together) and communication flow. Based on whom they follow and engage, Twitter users self-segment based on interests much more efficiently than Facebook users, whose connections and engagements are rooted in knowing each other personally as opposed to sharing common interests. The Followerwonk Twitter reports provide a great deal of information about context and can be used to map out optimized communications for targeted messages. This post is intended to provide a reference point for a conversation about what conclusions could be drawn from these reports and what actions could be recommended.

Click on the images to see them at their full size.


By comparing users of various accounts in Followerwonk you can create segments based on interest and overlap of interest to effectively target messages to those segments. The image below shows the overlap in reach between MSLearning and MVA on Twitter, with Reality_Nation followers included as a baseline because it’s a community I know to be prolific in their reach and gregarious in their engagement.


If you dig deeper into the follower counts of each, you see that MSLearning not only has more total followers, but their followers generally have more followers. About 14% of their followers have more than 500 followers themselves, compared to about 10% of both MVA and Reality Nation.





Reality Nation


Engagement (Sharing URLs and Retweeting)

If your goal is to drive visitors to a particular url via Twitter, you want to identify the followers who are likely to include URLs in their tweets. If your goal is to also disseminate a message broadly, you want to identify the followers who are likely to retweet a message.

1.5% of Reality Nation’s followers are likely to include a url in at least 25% of their tweets. 0.8% are retweeting someone else at least 25% of the time they are tweeting.


1.9% of MVA’s followers and 3.7% of MSLearning’s followers are likely to include a URL in their tweets. 0.5% of each audience retweets others 25% of the time.



The download in CSV feature on Followerwonk doesn’t seem to be working today, so I’ve included a screen grab of the MVA followers most likely to include a URL in their tweets. As you can see, the tool enables you to create specific sets with a great deal of granular information about the individuals in each set.



@contacts are a way to measure how much of the audience’s conversation is directed at or about others as opposed to simply being broadcast. For Reality Nation 1.6% of the audience mentions someone in particular in at least 25% of their tweets. Although MSLearning followers may be more engaged at what could be assumed to be a task based level (more interested in specific urls and messages), only about 0.8% of them have timelines consisting of tweeting about others more than 25% of the time. For MVA it’s about 0.7%.




Coursera and Udacity comparison in depth in separate post


Over on Coding Horror, the wonderfully witty Jeff Atwood’s post For a Bit of Colored Ribbon explicates a great example of how a feedback system that shows how you compare to your peers delivers huge value for you (the user) and the community of your peers.

Short version is seeing a graphic example on his energy bill comparing his use to his neighbors drives him to do more and more to reduce his consumption.

Aside from the clear to understand analogy to what they built at Stack Overflow and Stack Exchange, I totally love his description of designing the types of communities he does “what I do, what I’m best at, what I love to do more than anything else in the world, is design massively multiplayer games for people who like to type paragraphs to each other.”

Thanks, Jeff!

Was glad to be able to watch some very worthwhile talks from the live stream of re:boot conference about the future of online learning in CA higher education. A few interesting notes:

  • According to Sebastian Thrun Udacity sees 8 – 15% completion rates for their courses
  • Using professional mentors greatly improves student engagement (much more effective than peer mentors)
  • Daphne Koller spoke quite a bit about how to enable deeper, more meaningful active learning (and had great learning from traditional campus experience that extend well to online)
  • One special and powerful attribute of online learning she called out is the ability to leverage interaction between students via social networks and other online communication (the chat next to the video player provided great examples, but some spam too) This interaction is far more meaningful than attending a lecture. (When she made class sessions optional but replaced lectures with interactive sessions among the students in person attendance doubled)
  • Response time for student questions on message boards is a great metric – and it’s much faster if you have very large classes
  • Completion rates improve is students come in prepared by knowing what they are interested in and whether the class they have chosen will actually meet their needs
  • Steve Klingler of Western Governors described a professional mentor system that was very effective at delivering individualized learning and very comparable to the amount of one on one attention undergrads get at traditional campuses
  • Their tuition is based on time not number of credits, so the students have incentive to do more learning more quickly to get full value and their average time to graduate is an impressive 3 years
  • Online learning can do a lot more with mobile devices in regard to leveraging students being always connected
  • From an employer’s point of view, he presented the interesting metric of whether or not their graduates need less time to on board. (Because of the success of their competence based approach they have a 96% employment rate among their nursing grads)
  • They believe in embracing good content wherever they find as opposed to feeling they need to create the best of everything themselves in house.


It’s pretty common to have a Facebook graph that consists of a few unconnected or only loosely connected clusters. In the visualization above there are distinct “constellations” of people I play cards with, people I worked with at Microsoft, people I am related to, people I went to high school with, and people I went to college with. Most of the people in those groups are fairly interconnected with each other but not at all connected to people in the other groups, or perhaps only loosely connected. For example the two nodes between 3 and 4 are my brothers who went to the same high school as me and therefore bring the high school and family clusters into proximity, but they aren’t connected to anyone I went to college with or have worked with in the past 10 years. Those clusters are the type of graphs I’d expect to see from the organic use patterns of a typical Facebook users, more or less.

So what’s with the really big blob? That’s the social network of someone who behaves like a very, very passionate fan of reality TV, in particular Big Brother and Survivor. I’ve been managing social media around the Big Brother Live Feeds for a while, and like my customers, I’ve connected to lots of stars and tons of fans. And all of those folks are connected to each other through multiple connections themselves. Over in constellation 6 are people I’ve worked with at RealNetworks, and the nodes connecting that group to super fan cluster are folks who work directly with customers, like our chat room moderators, my social media team, editors and video hosts, as well as some key folks in program management and operations.

Here’s the key if you’re curious

  1. Bridge players
  2. Microsoft friends (past gig)
  3. High school friends
  4. Family
  5. Knox College friends
  6. RealNetworks (current gig)
  7. Reality TV fans

Aside from the density of connections, you can see by the size of the circle which nodes are highly connected. The average number of connections in the graph is 55 but that average is thrown off by the folks represented by the very large circles, who have between 150 and 200 connections each. Some of those are stars who have been on the shows–and reality stars tend to be very accessible and connected on both Facebook and Twitter. They don’t just accept friend requests from fans, they will actively and frequently engage. However, it’s worth pointing out that some of the most connected members of my graph are just fans. They are the type of fans who will travel to Las Vegas after the season is over to party with the stars. They know each other very well and are as connected and influential in the fan community (and among my customer base) as any of the stars or casting directors.

Here’s a close up of the Reality TV fans


And here are the high school / family clusters so you can see just how much sparser the connections are


Note: This post was an extension of the first week’s lesson of the Coursera course Social Network Analysis taught by Lada Adamic. Thanks, Lada!  The visualization tool I used is Gephi.