This course will become read-only in the near future. Tell us at if that is a problem.

WK 3 - Group cognition/cognitive artifacts - Gerry Stahl

Week 3 (May 9th-15h)

  • One Meeting - On shared piratepad, Sat 5pm EST (negotiable)
  • Readings:
    (Thanks Josh!)

Task Discussion

  • Jennifer Claro   May 15, 2011, 9:07 p.m.

    Hi Nate,

    This is very cool and useful. Thank you so much for doing this for us. How did you do it? What software did you use? Did you have to tag/code every line of text? How long did the whole process take? Is it difficult to do, does it take time to learn how to use the software?

    This would be a great tool for chat analysis or discourse analysis! It's the first time I've seen anything like this. Are there other tools/software like this available?

    Thank you very much Nate, it's so great to see cool new tools like this. Maybe we should make a new page for new useful tools and software! We could add Zotero, PiratePad, and whatever this chat analysis tool is called. Plus more!

    Thanks again,


  • Nate Otto   May 16, 2011, 1:02 p.m.
    In Reply To:   Jennifer Claro   May 15, 2011, 9:07 p.m.

    Thanks: I'm glad you appreciate it.

    This chat analysis tool I made isn't a piece of software or anything. I kind of whipped it up as an intermediate step to a static infographic I'll put together later this week. It's just some manually coded HTML divs attached to a script that changes an attribute based on a class (or "tag"). After manipulating the text from this format:

    	14:20 Nate: it works for some learners better

    to this:

    	<div class="chatdiv nate oer infographics">Nate: it works for some learners better</div>

    I just adapted a script to change the width of the post's box when you click a button.

    This method, unfortunately, doesn't involve anything automatic. The chatlog needs to be tagged manually and needs to be done right in the code. Someday there may be some software to plug chatlogs into an analysis tool, but it would still involve manually tagging posts for the most part.

  • Jennifer Claro   May 16, 2011, 4:23 p.m.
    In Reply To:   Nate Otto   May 16, 2011, 1:02 p.m.

    Thank Nate, 

    I have no experience with programming - what you think is easy looks very difficult to me! Thanks again for doing it, it gave me some ideas about coding and using discourse analysis to try to track down how groups learn together. 

    Thanks again,


  • Nate Otto   May 15, 2011, 4:18 a.m.

    The Piratepad meeting was fun. I think we eventually made good use of the linear chat PLUS nonlinear collaborative workspace nature of the tool.

    Following in Stahl's technique (sort of), I mapped out the chat log by subject and user and put it into an interactive infographic. Check it out! It's far from an ideal tool for analyzing patterns in chat logs, as Stahl attempts to do to find evidence for group cognition, but some interesting patterns do indeed crop up.

    You can see where certain members jump in and out of the "flow" as the topic swings to the things they're most interested in; just look at the palette variation as you scroll (and I recommend zooming all the way out to get the big picture view, though you can drill down and see the context for the changes.) Anyway, I'll get to the static graphical version of this data in the next few days.

  • Joe Corneli   May 15, 2011, 4:47 a.m.
    In Reply To:   Nate Otto   May 15, 2011, 4:18 a.m.

    neato!  It would be great to find ways to continue this work.  I'm reminded of a short paper and preliminary demo I made called "Gravpad".  

  • Joe Corneli   May 15, 2011, 4:51 a.m.
    In Reply To:   Nate Otto   May 15, 2011, 4:18 a.m.

    Another point: I wonder if we could sync this with the replay of the Etherpad's contents itself: worth taking a look at it anyway,

  • Nate Otto   May 15, 2011, 1:25 p.m.
    In Reply To:   Joe Corneli   May 15, 2011, 4:51 a.m.

    I highly doubt you could integrate anything like this with Etherpad. I had to manipulate the chatlog extensively to get the HTML in there, and all the important stuff is the meta-content (the subject tags). And because it was the first time I used regex, I had to just strip out the timecodes so I could work with it. I don't think there's any way to feed tag data back into Etherpad.

  • Joe Corneli   May 15, 2011, 1:31 p.m.
    In Reply To:   Nate Otto   May 15, 2011, 1:25 p.m.

    I mean as a long-term project not a one day hack !

  • Marcy Murninghan   May 15, 2011, 1:47 p.m.
    In Reply To:   Nate Otto   May 15, 2011, 4:18 a.m.

    This is fun, a good start toward mapping the emergence of cohesion. One thing to keep in mind, though, is the frequent booting off that occurred; just because someone wasn't commenting isn't necessarily due to lack of interest or distraction. As Stahl points out in the "How to Study Group Cognition" article, we don't know all that's involved in our particular (synchronous) online experience--going to the bathroom, being interrupted by a phone call, captured by aliens.

    Still, it helps locate where the energy is.

    We might brainstorm other keywords that are relevant, too. I suggest: learning; collaboration; professional; education; development; commercial; monetize; free

    Thanks for taking the lead on this, Nate!



  • Monica Resendes   May 14, 2011, 10:40 a.m.

    At 5pm EST meeting on Piratepad is still a go (this link I directs you to the CSCL-meeting2 page, which we created last week. It's the one with the ideas for a group artifact.)

    The discussion can revolve around how we envision the rest of the course proceeding - where we wnat to focus our energy, perhaps we can select an option for an artefact and begin to work on it.

    (There is a slight chance I won't be there on time, but I am going to try my best.)



  • Jennifer Claro   May 14, 2011, 7:41 a.m.

    Hi Everyone,

    Are we meeting Saturday via BBB or PiratePad? Please let me know. 

    Thank you,


  • Monica Resendes   May 11, 2011, 3:40 p.m.

    Hi all,

    I suggest you read the latter two papers listed in this week's course readings:

    The first pdf (Group Cognition: is a whopper (494 pages)-- it was included in our piratepad by a participant and is definitely a good one to keep on the virtual shelf!!