I had several very helpful meetings today. =)

The first was with Greg Wilson about a really
interesting experiment in social bookmarking for software engineering.
I wish I could have the brainspace to do it justice, but my research
supervisor feels I should concentrate and get my primary research out
the door first. Greg Wilson is way cool,
though, and I should definitely share ideas with him. =)

I discussed my research plan with my supervisor,
Mark Chignell. I described the waterfall-ish
division of time that I blogged about yesterday, and was relieved when
he suggested that instead of holing myself up in a library and reading
everything that's ever been written about the topic, I should instead
capitalize on my strength at prototyping things. I can build all these
little systems and watch what people do with them. When I see
something unusual, then that's the time for me to go and figure out a
theoretical framework to use in order to explore and evaluate the
situation. How nice it is to have a research supervisor who knows how
much I like to hack—and how hard I sometimes find it to focus on
something purely theoretical!

He also told me that he'd be happy if I could go for maybe one
conference and one journal paper. That'll probably be CSCW, then. This
means I don't have to worry too much about breaking my project into
publishable things. Think simple master's thesis, not PhD
dissertation.

My original plan was to be almost done by April of next year. He
thinks that's doable but ambitious, and that we'd both be better off
(read: less stressed) if I target June instead. From his experience,
masters students generally take two months to finish writing – April
and May. I need to have something reasonably promising by October in
order to convince IBM to keep funding me. My nearest milestone is
August 17, the intern day at the Cambridge lab: must have something
cool to talk about then.

Mark also suggested that I take his statistics course (“Every educated
person should know statistics”). Knowing my interest in business, he
also suggested the course on the business of software. I wouldn't mind
crossing over into Rotman for a business course, although it's also
offered under CS.

The last meeting was with Steve Easterbrook, Greg Wilson and Mark Chignell. We talked about CASCON. I like Greg's suggestion of a Ruby on Rails + mashups hands-on session that leads naturally into a student-oriented Hack Night. That would be great! I'm excited again. =)

If all my meetings could be like the ones I had today… wow!

Another thing I want to build for IBM CASCON 2006 is an easy way to
create an OPML file for conference registrants and session attendees.
Imagine if you could associate your registration with a blog URL and
then be able to:

• import an OPML file of all the conference registrants so far
• read an aggregation of all the conference registrants
• do the same for all the people who have registered for a particular session

Certainly, speakers with blogs should have them all listed. Tomorrow I'll ask for permission to get in touch with all of the speakers and ask them for blog URLs. We'll put together a page, export some OPML, throw up an aggregator (maybe even just a public Bloglines), and boom! Happy happy happy.

Even more advanced stuff: imagine a small-scale tech.memeorandum running against the speakers, the conference registrants, or session attendees… Imagine doing that with bookmarks, too! Maybe I can convince Pranam Kolari to do something like that.

In the future, people might even want to associate multiple blog URLs
with their profile. For example, if they write topic-focused blogs,
they might want their business blog to be aggregated with all the
other blogs for a marketing session, while their technical blog might
be better for the programming sessions. I don't think we're quite at
this point yet, but it should be easy enough to build.

Sounds like a terrific tool. I have one month to build this and all
the other nifty things I want to make for IBM CASCON 2006! I wonder if
my developer sponsor and my research supervisor mind if I do this…

E-Mail from Aaron Kim

I got so carried away making lunch that I nearly missed the planning conference call for IBM CASCON 2006's social computing workshops. I dropped in just in time to hear Stephen Perelgut and Steve Easterbrook talk about real-time collaborative note-taking, and I chimed in with my two cents about how wonderful it is to have backchannels during the conference.

A backchannel is an informal way for participants to talk to each
other in the background while the speakers are talking. Backchannel
chat is a great way to find out about other interesting sessions and
meet other people who are into similar things. We've also used the
backchannel to coordinate our attendance at sessions. (“I'm heading
over to session A.” “If you're blogging that, then I can go to session
B…”)

If the backchannels are logged, they can be the start of collaborative
notetaking. We tried backchannel transcription at one session during
Mesh. People were distracted because the backchannel was projected
onto the main screen behind the panelists. Most people have a hard
time keeping track of two or more streams of information, particularly
as they were both verbal. In addition, the IRC channel used for the
backchannel chat also included people in other sessions, which made it
hard for many people to separate the messages that were related to the
current session. Still, it was a good experiment, and that resulted in
a number of side-conversations during the session.

I think one of the things that would be great to have for IBM CASCON
2006 is a backchannel that people can get to through IRC and the Web.
I'd love to set up one of those, but it needs to be promoted somewhere
so that everyone with wireless can hear about it.

An alternative would be to encourage everyone to liveblog it and to use Technorati or a similar web service to aggregate all the posts tagged, say, cascon2006 and the session's tag.

HEY! There's an idea! If we suggest tags for each session and a tag
for the entire conference, then we make it easy for external bloggers
to make their posts discoverable. And I can so totally modify the CASCON blog to make it easier for people to “BLOG THIS SESSION” – they can post their content on the session blog and then retrieve it for crossposting onto their blog… That _would_ be totally sweet.

Think!Friday's tomorrow. Let's make it happen!

E-Mail from Aaron Kim