Presenter: Nathan Bos
Participants: Umesh Thakkar, Daniel Weber
Discussants: Jonathan Cummings, Atul Prakash
Buscope is a project to make the electron microscope at the Beckman Institute available to K-12 classes via the Web. Student groups send in insect samples, then access the instrument remotely at scheduled session times.
Bugscope followed on the heels of Chickscope, which was Beckman's first foray into this area. Chickscope was very popular with K-12 classes, but demanded too much personnel and resources at Beckman to be an ongoing concern. (One estimate puts the cost at $5,000 per class).
Bugscope uses a different model, and has proved to be popular and sustainable. It can be scheduled into 2-hour blocks of unused instrument time. The sessions themselves require minimal staff time. Student helpers can run the online sessions and do most of the sample preparation. The project director can handle awarding and scheduling of sessions through online forms.
Bugscope uses the IRMA software that was developed at Beckman. Both a Java and a Browser version are available.
Bugscope has conducted 82 online sessions since March 1999. Survey of users show it has good ratings for software usability, and teachers rate it as useful.
Interesting questions to ask about Bugscope:
- Which stakeholders have the most to gain (or lose) from the Bugscope collaboratory?
- Stakeholders: K-12 students, K-12 teachers, General public, Beckman Institute, Scientific researchers, Scientific fields, Software developers...
- Issues of Scope: Is this model scalable? How would scaling put new demands on:
- Instrument operators?
- Internet software?
- K-12 classrooms?
- Curriculum support?
- Student helpers?
This is clearly a successful educational outreach project. But at this scale, these aren't Computer Science CSCW questions because the technologies being used (chatroom, etc.) are mature. There are no compelling design questions involved in enabling shared instruments.
There would be a set of interesting CSCW question if this project were to be scaled up including:
- How to accommodate large numbers of users?
- How to coordinate access across multiple instruments?
- How to enable 'open' access?
- How to enable user-to-user collaboration with back-and-forth communication?
- How to allow contribution of "Software Instruments", e.g. model?
- How to handle IP or security/privacy issues?
(see slides for more details)
Q: For the people on the project, what are the prospects for scalability?
A: This is a very interesting question. We recently got an award from NASA to make this a research-grade tool for NASA scientists' remote use. The limiting resources on scalability are:
Time during sessions to reposition scope
Time for setup between sessions (~1 hour)
Working on purchasing more scopes
Q: Would it be possible to pre-scan and then simulate the session, with different resolutions, etc.
A: That is on our to-do list (creating simulations to supplement or replace live sessions). We would like to sample the entire stage and stitch the graphics together. This would also allow users to 'zoom out' further than is possible with the microscope, which helps users stay oriented. Would also allow us to simulate this ahead of time so teacher or researcher can prepare for the session, and focus on important parts.
Q: You mentioned an entomologist who was an ad-hoc advisor. Is this an important limiting factor (on scalability)?
A: Yes, that's an important part of the project (classes can ask questions during the session), and staff can do some research to answer question.
Q: Would it be possible to foster more collaboration between classes?
A: From SOC staff: The web archives of Bugscope sessions go a long way toward supporting this. The logistical problem of cross-classroom collaboration is usually because of incompatible schedules. So (with the archives) if a class comes up with a nice set of images they can be shared asynchronously.
Q: Are the archives used?
A: Not known. The log files would need to be analyzed
A2: Classes can also log in as observers on other class' sessions. But this hasn't happened very much. Teachers are very busy and curriculums are crowded.
Q: Would like to know more about collaboration within the classroom, and collaboration between students and operators. Would like to ask, how much do students deviate from presets?
A: Classes vary-some just bounce from preset view to preset view. We have looked for a predictive variable for this without success. Usually we set the presets so that some interesting things are hidden-they need to go to a preset, but then zoom in to see it, and some classes miss these, even when we make suggestions. Others schools do a lot of exploration. And have also had cases where they totally ignore us, and go off and do something cool on their own. Bugscope staff are very interested in this question of when they might 'go outside the box' in their inquiry.
Q: In other examples (e.g. Keck) it's very clear where the advantage is. But I don't see things like Keck scaling up. Instead there are things like the Sloan Digital Sky Survey, where you capture a larger dataset. So this might be the kind of thing that is scalable.
Comment: Simulations might prove useful, but there's also some value in having (customized) specimens.
Comment: One Sky Many Voices is an example of a project that is having cross-classroom collaboration, but that's because they're getting a lot of NSF money to support it. There are no pressures within the educational system for collaboration across classes.
One aspect of Bugscope that we may be under-emphasizing is the access to remote expertise. Chatting with experts may be very useful. Could also imagine system for classes to send in questions (asynchronously) for consultation.
back to the Final Summary