In the past, I have ranted about the fundamental flaws with threaded discussion interfaces as opposed to a simpler flat interface, where all posts in a conversation can be seen on one single, scrolling page. I am going to up the ante by making a more specific claim and challenging the research community to prove me right or wrong.
My hypothesis, based purely on personal experience, is that threaded discussion interfaces lead to far less synthesis in educational conversation than flat interfaces do.One of the most important activities in an educational conversation is pulling together the various ideas and viewpoints that have been expressed previously. You want to see a student pointing to two or three different comments and weaving them (or responses to them) together into one coherent position. My experience in flat discussion interfaces is that this happens a fair bit. The fact that a student can scroll up and down to see and pull together all of the pieces when formulating a response affords synthetic thinking. On the other hand, I see far less synthesis typically taking place in a threaded interface. This is because threads atomize the conversation. You have to click to see each post, and if you want to look over several posts as you formulate your response, you have to remember the subject line for each one and click around quite a bit to see all the content to which you want to respond. Before students can lay out all content for themselves, the interface forces them to play the hyperlink equivalent of a game of Concentration. For most people, it’s just too much work. They don’t do it.
It would be fairly straightforward to set up a study to test this hypothesis. You’d want to look at decently large cross-community (or cross-course) sample of conversations taking place in both threaded and flat boards. Analyze the number of posts that reference posts other than their immediate predecessor, both generally and through specific quoting or linking. I’d be willing to bet money that, on average, discussions taking place in flat interfaces have significantly denser networks of references among posts other than immediate parent-child posts.
If anybody is interested in taking on this research project, please feel free to contact me. I’d be delighted to help out in any way that I can.
I don’t think I have enough experience with “flat” discussion board interfaces (in a class setting) to really have an opinion on this, but I’m intrigued. I’d like to see the results of that research.
Michael, I’m going to jump in here with some unformulated thoughts, but thoughts nevertheless.
Bb, for instance, allows instructors to control threading in their Discussion Forums, and gives all users the option to open all posts to create a single scrolling page. But that’s as far as it goes. The real problem may be this: what happens in a discussion forum is NOT a conversation, and there is no way that we can make it into one. It’s a different beast. In a sense, you (we) are looking for a visual and structural representation that approximates the auditory, visual and proximate nature of a face-to-face conversation, and that’s probably not in the cards. You know that I prefer a “concept mapping” interface that, as you pointed out, could potentially capture the “semantics” of different “conversations” and give the participants inroads into all sorts of ancillary “dialogue.” (Gosh do I hate these qualifying quotation marks.) I used to describe asynchronous online discussions as purified experiences, exempt, somehow, from all that bodily stuff that gets in the way of face to face conversations. This, I thought, was an attribute. Now, I’m not so sure. One thing I am sure about though is that the CMS model for discussions is pretty poor. It’s 100% controlled by the instructor, it looks terrible, it is hierarchically arranged – not by the quality or frequency of the ideas being discussed, but by time – which, ironically, puts the most current posts at the bottom. The best pedagogic success I’ve had with asynchronous discussions is using them as material to pump up face-to-face conversations in the classroom AND vice versa. That’s why blended courses have such appeal to me. I’ll think about this some more and try to define what I think is going on in these asynchronous discussion forums.
I don’t agree. It’s certainly true that a thread on a discussion board doesn’t have to be a conversation, and it’s probably also true that there are times when thinking about a thread as something other than a conversation is beneficial. But I’ve had plenty of natural, organic conversations in flat interfaces, and even a few in threaded interfaces. It’s probably a somewhat subjective judgment, but I don’t see enough difference in either the affordances of a flat interface or my experience of the interchanges that have taken place through those affordances to say that that they lacked the essential features of a conversation. True, they are different sorts of conversations. But they are (or can be) conversations nevertheless.
actually, the old “freenets” often had flat discussions. The best derivative of these systems is Caucus which not only allowed one to see all discussions in a thread but allowed easy links across conversations at several levels.
Open Croquet, a 3D peer-to-peer, open source, system which is about to be released in a developers version creates exactly that for which you wish by essentially creating a large world where conversations take place much as in a large party atmosphere where you can wander through conversations and even exchange objects or create private conversations.
Interesting. Tom, are there web sites available for Caucus and/or Open Croquet?
Michael, I’m not sure that it’s helpful to say that two things are the same when they are actually different because then we lose sight of what makes each special in its own way. My SMS conversations are different than my phone calls which are different than my F2F meetings which are different than my online discussions. That’s why I’ll ask someone to email me, or to meet, or to do it online, though, more often than not, I like it best when it’s a combination of all three. Each seems to be appropriately nuanced for a slightly different portion of the task, but it’s the fact that we’re able to communicate with one another in all of these environments that is important, and what’s needed, I think, is to figure out how to maximize the effectiveness of each. I would much rather be having this conversation with you F2F right now, but we would both be missing out on Tom’s contribution. I’ll ask Tom a few questions or piggyback your own; maybe we can do some emails or a skype thing, but probably, if he’s on to something, I’d like to sit and brainstorm with him and you and whoever else jumps it to this – with a scotch. Again, that’s why this blended thing is so appealing to me because it is premised on the legitimacy and exploitation of all of these communication tools working together.
When an f2f conversation goes bad, we usually use words to fix it; it is, in a sense, auto-correcting. The problem with the discussion board is that the actual interface interferes with the type of communication we expect to generate from within it – which is an altogether different story
hi michael
yes, there is info on opencroquet- just google. Also, you could then contact Julian Lombardi at the University of Wiscosin-Madison to get specific information. A developer’s version is supposed to be released at the end of this summer
as the name implies, its open source
Caucus has been around for many years. One of the most effective users of the system is Group Jazz, lead by Lisa Kimball- again a google well get you there.
Eric’s comments are well taken. We should be careful not to try to map a past communication mode, e.g. f2f from brick space into click space but use the best of each. Opencroquet allows for multiple levels and types of exchange from phone via a selective voip package, to text exchange to group or community exchages- but no teleportation for f2f- video is the closest here.
tom
Hey, you get your wish!
I wanted to let you know that The Canadian Institute of Distance Education Research – CIDER (http://cider.athabascau.ca/) is hosting a session on April 8th where you can pitch your research idea…
Here is the session’s info:
Title: Content analysis of online discussions.
Facilitator: Dr. Elizabeth Murphy, Memorial University
Date/Time: 8 April 2005 at 11:00 AM (Mountain)
Mode: Audioconference with slide presentation (Elluminate Live!)
The session will start with a 20 minute presentation summarizing three years and eight studies of content analysis of online discussions. The presentation will be followed by a discussion period in which participants can explore their research questions related to online discussions. Participants are invited to come prepared with a research question (one sentence) as well as a method (one sentence) they believe would be suited to exploring answers to the questions. Participants can post their questions and method using the DM.
Find out more here: http://cider.athabascau.ca/events/cidersessions2005
I have found that the threading nature of the discussions in WebCt tends to confuse beginners but at the sametime felt that showing my students how to ‘flat line’ their discussions might totally overwhelm them (like the conversation we are having here – comments bounce at them every which way in the flat format)…I’m thinking for my next course, though I will share this information and at the end of the course, I can see which way they prefer to read the discussion board – I rarely have my WebCt discussion board threaded to read it because I like to get a sense of the overall discussion. Thank goodness for the “unthreaded” option.
I realize you’re talking about posting not reading, but I think they go hand in hand.
My feeling is that density is not a function of format (flat or threaded) but rather a function of the discussion (or content) and perhaps the players (those discussing).