Last week we had two interesting visitors who each gave talks in the area of tangible computing. (Briefly, tangible computing explores ways of interacting with computers using real-world physical objects; much more info can be found online including at the Tangible Media Group at the MIT Media Lab). FXPAL has done a number of tangible interface projects over the years, including the PostBits project, the Convertible Podium, and others.
Looks like I missed a good paper at JCDL 2009: A Polyrepresentational Approach to Interactive Query Expansion by Diriye, Blandford and Tombros. As with many good ideas, this paper describes an approach that is obviously useful once described, but one I had not come across before.
Manual query expansion can be useful when relevance feedback fails because it doesn’t know why a person found a document relevant, but people are often reluctant to use the suggestions offered by information seeking systems. This paper offers a new twist on these recommended terms: When suggesting query terms for expanding a user’s queries, they show terms with some representation of the context in which they occur. Evaluation showed that this contextual information allowed users to understand query terms better, and that it improved their ability to make relevance judgments with respect to documents that contained the suggested terms.
In Cerchiamo, we offered users term suggestions based on relevance judgments made by search partners. While the suggested terms were useful for identifying other relevant documents, they weren’t always used. It’s likely that term recommendation in collaborative search situations would benefit from these techniques even more than in the standalone search because in the collaborative search case term recommendations may be based on documents that a searcher has never seen.
Bill Kules and Robert Capra have an interesting poster at JCDL 2009. “Designing Exploratory Search Tasks for User Studies of Information Seeking Support Systems” describes a study that evaluated an algorithm for determining whether a search task makes a good exploratory or known-item search candidate. They evaluated their approach by having 18 study participants run candidate queries (four exploratory and two known-item). After performing the specified searches, participants answered the questions about
The study reported the following significant differences between the two conditions: People said that they found exploratory tasks more difficult, they had lower confidence that the task was fulfilled, they had to find multiple documents to fulfill the request, and their goals changed during the search session.
This is a nice summary of the distinctions between exploratory search and known-item or navigational searches, and underscores the need to build interfaces that support the differences among these tasks.
Update: The poster one page description is now online, as is the poster handout.
No, really, it’s on the TV schedule this time (a couple of weeks ago the show got pre-empted for a pledge drive): You can get a look at our Virtual Factory and some of our molecular dynamics animations on “The Science of Chocolate” which is showing tonight on Channel 9 (in the Bay Area) as part of the KQED Quest series. The story is focused on the hows and whys of chocolate making, not on our Virtual Factory project, but it’s still fun to see some of our work on the air.
Model of a theobromine molecule
All these 3D models and animations were created by FXPAL’s resident Art Guy, Tony Dunnigan, with Sagar Gattepally handling the virtual world construction; the video embedded in-world was shot by John Doherty.
The show is on tonight, June 16 at 7:30PM on KQED, Channel 9; will repeat at 1:30 AM Wednesday June 17; and should also commence streaming on the KQED web site as of tomorrow. It looks like the “Science of Chocolate ” story is one of two stories in this show.
Brynn Evans has been doing some interesting research on looking at how social networks support collaboration over information seeking. She uses two dimensions—search goals and search locations—and looks at how social interactions affect search activities, and discusses the implications that this work has for the design of tools to support information seeking. Slides are available through SlideShare, and a video of the talk is also online.
Or, when worlds collage…
In yesterday’s post I promised to discuss my favorite feature in the beta version of the 3D web browser Exit Reality. This was the discovery: as a way to create rich 3D worlds quickly, you can stack worlds and models — and their accompanying scripts and animations — inside of each other, all inside one browser window. The Exit Reality 3D search provides a rich source of 3D objects and worlds; you simply drag-and-drop them from the search results into your open world-window.
The collage effect is less of a mess than you might expect, despite differing scales and environment settings. Or OK, it’s a mess, but an interesting mess.
Arguably the two most common topics on this blog are search, especially collaborative exploratory search, and virtual worlds. Now, the new browser-based 3D platform ExitReality, has piqued my curiosity by bringing these topics together. As part of their 3D platform, they offer a search engine optimized for finding and displaying 3D objects and worlds. You can either enter the found 3D sources as a world entire unto itself, or (my favorite) drag-n-drop it into your current 3D space in the browser window. (If you’d like to see the 3D search via a normal 2D web page, that’s available here.)
Note that this 3D search engine is one that searches for 3D objects, models etc., not something like the SpaceTime browser that displays standard search results in a 3D(ish) format.
So it was this morning I found myself standing with a wizard, a Doberman and a rat on the outskirts of Stonehenge, contemplating several quite nice Moon lander models and a gigantic purple flower (Cattleya – one of the search results for “cat”). This all in the space of ten minutes’ carefree clicking around. Dali would’ve had a ball.
Browser-based 3D with 3D search engine
ExitReality’s tag line is “the entire Web in 3D.” The idea is you can convert your own website to 3D via a fairly simple process – and it’ll still look the same in 2D; you’ve just added a 3D button. In general the interface is very well thought out – where it falters is most likely due to its beta status (e.g. avatars can’t yet fly or change clothes, though you can change avatars).
My second favorite feature so far: when other people visit the web site you’re viewing, you see them as avatars (if they have ExitReality installed). It’s possible to use in a standalone kiosk mode, or in secure mode behind a firewall. My first favorite feature? Check this space tomorrow for details.
Damon Hernandez’ blog has a very nice synopsis of last week’s Metaverse U at Stanford. The conference addressed new applications for virtual worlds, and open source platforms for supporting them.
Damon caught me (and several other people) for a short interview that’s now up on his website (and YouTube).
First, I agree with Daniel Tunkelang and many others that this is an important step for Google, an important departure from the ranked list. Hurray! My sense about this way of performing exploratory search is that Google Squared addresses (at least) two different kinds of activity: filtering through structured data, and collecting different instances of some topic. The former, of interest to the set-oriented retrieval crowd, should make it easier to do feature comparison for a variety of domains, typically with some sort of a shopping angle. In this post, I will focus on the second kind, the less structured exploratory search. Continue Reading
I wasn’t able to attend or contribute to the Collaborative Information Behavior (CIB) workshop held recently in conjunction with the GROUP 2009 conference, and was looking forward to reading about the discussion, which was posted today by Sharoda Paul. The discussion touched on several interesting points, and perhaps we can get some of the participants to comment more on them (here or elsewhere). The notes highlight discussion around models, behavior vs. technical design, appropriate research methods, explicit vs. implicit collaboration (yeah!), and evaluation.