Crosstalk meeting on February 27, 1998

New Models of Student Computing

The discussion focussed around the questions:

• WHAT do you do now with regard to use of computing in your classes? What parts of the current environment do you use? What could be different?
• WHAT would you like to do in the ideal world? What is already there? What needs to be added to the current environment?

Where the word "WHAT" in the previous sentence = both CONTENT (e.g. hardware, software applications) and PROCESS (e.g. communication, connectivity, ubiquity, etc)

Vijay started by pointing out that the Environment is defined by:

• Functionality provided (software, communications, support, ...)
• Location of the machines (e.g. dorms, clusters, ...)
• Type of machines (e.g. Intel, Mac, UNIX, ...) and also by Ownership:
  • Centrally owned
  • Departmentally owned
  • Student owned

Opinions about the current environment and possible changes to that environmentthat were expressed during the discussion included the following. Opinions varied, but the general consensus was that we need to increase the heterogeniety of the environment while maintaining its stability, robustness, and support.

• Faculty need help in producing Web-based materials for teaching. Faculty should be focussing on content and pedagogy, not production. There needs to be a support structure on the department level, but also going beyond that.

• A centralized system leads to a more uniform interface, which is better for students. This can also lead to cross-fertilization between departments.

• What's good about the current environment is projection in the classroom, network connectivity in the classrooms, and electronic classrooms. What's lacking is the ability to get at stuff from home. Need to support MediaOne as well a Tether, DHCP, etc. We need to work on getting student licneses via MediaOne to software they use at school.

• We should shift expenditures away from a large public computing environment toward student ownership and specialized clusters for things such as visualization. We should be looking at PC software as our base, and Project Pismere is a good step in that direction.

• The current cost of maintaining PC's in a lab setting is very labor intensive. The PC is not robust enough. Athena boxes take care of themselves and we're very happy with them.

• The value side of NT is the options it gives for software applications.

• The PC is more maintainable in a private ownership/single user model

• We can't expect every student to license $1000 worth of software for their PC. The big question is "how do we provide affordable, reliable software in a distributed model with self-ownership."

• We'll never have a "one size fits all" model. We need the mix of back end heavy duty machines, a mix of public machines (both high end and read-your-mail use), and student owned machines.

• What's needed is something that preserves the benefits of Athena:
  • Stability
  • Infrastructure
  • Software licensing and support

• The big question is "How do we provide an increased level of heterogeneity without sacrificing the consistent, reliable service that we currently provide?".

• The price that was paid by Athena is that MIT went one way and the rest of the world went another (e.g. no Microsoft Word on Athena). We need a front-end suite of tools that are world standards.

• Laptop ownership will increase as the cost goes down. There is now a laptop that sells for less than $2000.

• We need network drops in all sorts of informal places, as well as in the Libraries. The socialization aspect is important to having computing more integrated into the student environment. People in the Rotch Library already ask for group study spaces with network drops.

• The fact that MIT is in an urban environment is an impediment to laptops because of physical security aspects. (What happens if it gets known that MIT students carry laptops across the Mass Ave bridge at night?)

• We need to continue to have our infrastructure (e.g. central servers, central filespace), and broaden the multimedia representational aids.

• The Foreign Language faculty have been developing materials that are currently delivered in the Language Lab. They need to be accessible across campus and beyond the physical campus. To do this, we need high speed access so that we can deliver digital video over the network. Students need to be able to use these materials and collaborate wherever they are, not just in the Language Lab. Developers also need to collaborate for production and development of these materials.

• Faculty need to know what are IS's plans as far as timing re high speed network access for things such as digital video.

• Servers need to play more of a role than just storing the data, doing some amount of computation and letting the user bring over just what he needs to display.

• Do we have or need a program to teach students computationsal skills? Answer - our students do OK with computational skills. What they need is more within the information skills arena.