Hide full article
A few weeks ago, I attended portions of the ACM's (Association for Computing Machinery) 1998 Conference on Computer Supported Cooperative Work. The conference centered on using computers to make it easier for people to work together, an extremely worthy goal. Think about it: for the last fifteen years, the computer industry has focussed on helping individuals use computers more productively, but has placed much less emphasis on creating applications and devices that help us work together. As an example, look at the ubiquity of email: we all use it, and for many people email is by far the most important aspect of the Internet because email lets us work with others. But how many email innovations have there been in the last ten years? Sure, it's faster, prettier, and easier to use, and modern email programs provide individuals with powerful features. But how many new features help us better work with other people?
<http://www.acm.org/sigchi/cscw98/>
The best way to immerse yourself in a sea of ideas at this conference was the demo session, where about 30 groups showed research projects. Some were far beyond my ken, such as the programming toolkits aimed at cooperative work, but others were more easily grasped and had stunning implications.
Meme Tags -- Unfortunately, I didn't see the demonstration of this project, though one of the principals at the MIT Media Lab explained it to me. First, a meme is an idea that spreads and evolves by the rules of Darwinian evolution (survival of the fittest, and all that). A meme tag is an electronic name tag that stores, displays, and trades names and memes. Meme tags explore how we interact within groups.
The idea is that when you meet someone and talk to him, face to face, you can see his meme, and by pressing a button on your tag, accept or reject it (via short-range infrared transmission). The tags track both who you've talked to and how the memes spread, reporting back to a centralized database, which analyzes the data to determine the best schmoozers, the most successful memes, and so on. The centralized reporting takes place on a "community mirror" - a large screen that reflects the trends in real time. Especially interesting was the way people wanted to cause trends that would change the display by evangelizing memes or increasing their schmoozing.
I want meme tags now, but in a different form. Consider an average trade show: you meet hundreds of people, stop at numerous booths, and frantically try to stuff tons of information into your head. Business cards trade hands, often with cryptic messages scribbled on their backs, and you gamely try to make sense of it all after the show. What if whenever you met someone, your name tag automatically recorded the information on that person's business card, plus anything else they wanted to encode, along with the time and date. Interesting information at a booth could be presented both in paper form and as a URL beamed to your badge when you pressed a button. You could exchange data with other people's badges, and when you got home everything could be dumped to database. In short, although the research implications of tracking this information are fascinating, I think meme tags have the making of a real-world product.
<http://lcs.www.media.mit.edu/people/fredm/ projects/memetag/>
The Hummingbird -- While the meme tags worked at very short range, another project from the Viktoria Institute in Sweden worked over longer distances, but provided less information. The Hummingbird, a specific instance of what the group calls an "inter-personal awareness device," is a small electronic device that constantly monitors the surrounding area for other Hummingbirds. When it finds one, it chirps and displays the name of the person wearing the other Hummingbird on an LCD display.
Although the information conveyed is relatively minimal, it can still be useful. The group tested the Hummingbirds in two situations, a rock music festival and a standard office. At the rock festival, the Hummingbird wearers reported liking the knowledge that another Hummingbird wearer was nearby, but they also found it frustrating that they couldn't necessarily find the person (the Hummingbirds' range is up to 100 meters). In the office situation, the Hummingbirds proved more practical, since the members of a workgroup could easily tell when another Hummingbird wearer entered or left the building. Obviously, if everyone had a Hummingbird, the devices would have to be configurable to look only for people in specific groups.
The important issue here is awareness of physical presence. Our forms of remote communication are necessary these days, but face to face interaction remains compelling and useful, and the Hummingbirds helped their wearers determine when face to face meetings were feasible. The Hummingbirds struck me as more theoretical than practical, although trying to catch up with people in large office buildings can prove sufficiently annoying that I see the utility of a system generating physical awareness information.
Interestingly, the Hummingbird group also had some matchmaking devices from Japan called Lovegetys (They're popular; reportedly over 1,000,000 have been sold this year). The devices come in male and female versions, and you enter what level of relationship you're interested in (Chat, Fun, or Friend). If you have a male version and run across (via radio waves) a female version set to the same level, both devices flash and optionally beep to let you know that you've found someone compatible. They're sort of a cross between the meme tags and the Hummingbirds, since they convey physical awareness beyond line-of-sight, but they also transmitt additional information and act on it if necessary. Future versions will reportedly increase the range and offer more precise settings.
<http://www.viktoria.informatics.gu.se/groups/ play/ipad/>
<http://www.love-gety.com/>
<http://db.tidbits.com/getbits.acgi?tlkthrd=490>
Virtual Work Rooms -- I said before that email hasn't seen much innovation for cooperative work, but the Virtual Work Rooms project from a team at Carnegie Mellon University (CMU) aims to change that. Most email programs look like spreadsheets, with rows and columns. You can sort the rows and add a level or two of hierarchy by applying labels or moving messages to other mailboxes.
But what if the information in those rows was not quite so static? That's the idea behind the Virtual Work Rooms demonstration, which essentially put a dynamic spreadsheet backend onto email. The demonstration used a situation where someone is spoofing email from different machines. Several system administrators must share information on the spoofed messages to track down the miscreant. If this were to happen today, it's entirely likely that one person would set up a spreadsheet (people love tabular displays of information) to track the information and responses.
The demonstration system, however, showed an email client (as a Web page initially, although a Eudora plug-in is under consideration) in which the users could create customized views using spreadsheet-like formulas. This level of flexibility enables users to ask questions of the data, such as "Who hasn't responded to the request for log data?" It's easy to make new queries, so new views could be created quickly.
Although the concept takes some getting used to, it's promising, especially in situations where the information coming in via email is well-defined. For instance, if you need to ask 100 people several questions in email, you could use a system like this to tabulate the results and to see who has and has not responded. The trick would be the interface - using a system like this would have to be easier than using a spreadsheet to manage the data.
The project brought to mind the flexibility of the Newton data soup, which eschewed typical hierarchical filing systems. Imagine an email program where all messages essentially float in a database, and constructs like mailboxes are merely queries. In such an email program, adding some intelligence to the queries would provide a system similar to what the CMU team has proposed.
<http://muffin.hss.cmu.edu/vwr/web/index.html>
inTouch -- The inTouch project prototype from the Tangible Media Group at MIT's Media Lab consisted of two boxes, each holding several wooden rollers. When you moved the rollers in one box, the corresponding rollers in the other box mimicked the action. Two people using the boxes can passively feel the manipulation of the rollers, cooperatively move the rollers, or fight over the state of the rollers. It's difficult to see real-world utility for this sort of "haptic communication" outside of the gaming world, but areas like telemedicine's remote surgery or remote control of robotic devices might benefit in the future, and it's interesting to see the concepts of cooperative computing extended beyond things people are meant to see or hear.
<http://tangible.media.mit.edu/projects/inTouch/ inTouch.html>
CLIVE -- IBM showed CLIVE (Collaborative Live Interactive Voice Environment), a customer support application with a high level of utility. Implemented as a Netscape plug-in, CLIVE enabled two people to view and interact with the same Web page, all while talking on the phone. Each person saw the other's cursor in a specific color, so you could see what the other person was doing, and both people had a crayon tool for circling specific parts of the page. All the interactions went through a proxy server that acted as traffic cop and ensured privacy. Overall, I was quite impressed, both by the system's apparent simplicity and its utility. As Web-based interfaces grow more complicated, allowing two people to work on the same page at the same time may become increasingly important.
The demonstration showed a situation where a user was calling for support on how to work with a Web-based online banking interface. The example hit home because recently Tonya and I were adjusting our retirement accounts via the Web and had to call for support because we didn't understand the terminology. With something like CLIVE in place, the support person could not only have explained what we needed to do, but also showed us. Apparently, a new IBM organization called Corepoint will market CLIVE.
<http://www.research.ibm.com/resources/magazine/ 1998/issue_2/cybercoll398.html>
<http://www.corepoint.com/products/products.html>
Ideas Meme Business -- One aspect of the conference that I enjoyed was the step back into academia, where ideas are the coin of the realm. Although the people attending and presenting at the conference weren't all academics, those from the corporate world tended to hail from research departments. I found my mind racing with possibilities after chatting with people or seeing a demo. Never mind that the proposals were rough, the presentation materials often needed editing, and the demos seldom showed a marketable product. The beauty of an academic conference for someone like me is that I could put aside practical considerations and focus on the fascinating ideas - a refreshing change compared to the all-out commercial assaults of industry trade shows.
