Learning Piazza / Learning Space

Information system functions must evolve in concert with the evolution of faculty and student needs. Faculty innovators in different disciplines come up with different pedagogical models and research practices. Some will match the project-based model and the flex-lab. But since many modes of learning and creative scholarship are not served by such a model or architecture, let me suggest some additional functions which could be incorporated into a Scholars' Piazza.

By calling this cyberspace a Piazza I'm enlarging the Learning Space concept to include a fuller range of scholarly activity in an augmented experiential environment. http://topologicalmedialab.net/xinwei/papers/texts/scholarspaces/1.Introduction.html)

If we build such augmented spaces only for undergraduate service courses then we needlessly limit the range and depth of work that can be done in such a space. Why not build a library space that can augment work at a quasi- professional level? The models need not be limited to skill-training or survey courses involving extensive work -- spanning and integrating a broad range of information. They also could support what can be characterized as intensive work -- focussed deeply on one subject. (http://topologicalmedialab.net/xinwei/papers/texts/scholarspaces/3.Design.html )

I'll give examples below.

We can build spaces for the construction, manipulation and rendering of mathematical structures -- a dry lab for physics and other symbolic models. The aim here is not only to depict but to teach how to analyze, conjecture, confirm. For example, Blas Cabrera's pedagogical ambition was to teach apprentice physicists how to make and interpret their own models, rather than play someone else's simulation or worse, watch a string of videos. The ideal would be to couple a mathematical programming language with numerics packages and multimedia editors. For more precise proposal, see Overview: Liveboards for Mathematical Research; and Proposal: Intelligent 3D Blackboards for Geometric Research (7 pp, PDF).

We can build labs for empirical science. This, too, will require access to the most powerful computation engines on the net. For example, weather models and blood kinematics will require access to supercomputer numerics and delivered on Reality-Engine-class rendering systems. Multi-player labs will require even more co-processing. For this reason, it's not reasonable to expect a Piazza to be a self-contained net of computers. (see Nuts and Bolts.) Other examples include optics benches, organic chemistry labs. What we could use, then, are not so much hermetically complete labs, but workbenches of coherently designed virtual tools -- numerical integrators, data collectors, statistics packages, a database of compounds and objects with embedded physics. etc., that can be called together in a scientific programming language. (see section on Writing Systems.)

I put music in a class by itself, because, despite CCRMA's proximity, we still haven't got a concert-quality space which is wired for cyber-performance. This is because we've been too ocularcentric in our design. The Piazza should be able to import tools and music from CCRMA or the Archive of Recorded Sound, for example. At the very least, the Piazza should be usable as a performance space for live instruments or voice without amplification.

Applications range from multi-player performance to theater and psychoacoustics. With computer-support, we should be able to configure a space with different acoustics and even assign acoustic functions to different locations or gestures in space. T.V.Raman (DEC) (AsTeR -Towards Modality-Independent Electronic Documents, DAGS95.) and others have developed extremely sophisticated and rich interface systms that map fairly general structures into aural space. I suggest that we study such work before wiring in the sonics and accompanying software.

We should look to CADRE at SJ State and to SF State to see what can be done. For example, we could run multiprocessor SGI's to render synthetic + sampled backgrounds against which live actors could interact with "believable agents." (See, for example, MIT Media Lab's ALIVE: Artifical Life Interactive Video Environment, Maes, P., Pentland, A., Blumberg, B., Darrell, T., Brown, J., and Yoon, J., Intercommunication, Vol. 7, pp. 48-49, Winter 1994) The sampled backgrounds could come from SUL collections. Such experiments may yield insight into how we should embed computational artifacts into physical spaces yielding environments for richer scholarly work.

Physical arrangements are just one part of the cyber-architecture. An equally crucial aspect is the symbolic architecture. We seem to agree that physical flexibility is good. What does flexibility mean in the context of symbolic spaces?

- rich structures: one reason why digital video is so painful to work with is because almost all video is opaque. There's no machine-parsable structure underlying most digital video (except a timestamp). Imagine if your vision could not pick out words or sentences or if you could not tell in what language a text was written. This lack of structure hobbles casual editing and makes rich scholarly work impossible. The problem is quite general with digital media. We need to be able to easily associate rich symbolic structures with every physical or symbolic entity: scorefiles with musical recordings, TeX with Mathematica, and Mathematica with video, databases with people, places or realia.

- multiple modalities and models: We must also be able to work in multiple modalities. A Piazza built of only visual elements would be a deaf world. A Piazza should host musical performance or spoken word. We should be able to localize the Piazza in multiple languages (cultures), especially Pacific Rim and Latin American. This will require appropriate human support as well as software.

- variable focus: We should be able to support work that ranges in focus from the extensive to intensive. Extensive work is characterized by broad, shallow activities that integrate many fields -- like a puppet theater for a language course or a public show about the history of cosmology. Intensive work concentrates on a single subject -- like an analysis of a single movement from a Ligeti string quartet or the construction of a proof. It's important to realize that this is independent from the degree of collaboration, and applies to teams as well as individuals.

Appropriate Technology and New Writing Systems

It's crucial to keep in mind the ratio between staff support and scholarly pay-off. Right now, much effort is needed to juggle sets of applications and system patches to run a lab space supporting fairly extensive work. But this is, hopefully, a transient problem, due to legacy operating systems and application software and to a singular shockwave of transforming writing technology. Looking 2-5 years downstream, I see that much of the functionality of a flex-lab as currently designed should migrate out to the residences. This includes small-group planning, informal skits, website construction, residence-based seminars. SUL will pipe information in lowest-common denominator formats to the dorms.

So, what next? An augmented space -- a Scholars' Piazza -- in order to gain and to warrant economies of centralization -- should probably be a lab where we can construct a variety of public augmented spaces. What distinguishes public spaces from more intimate spaces like a tutorial office or a planning room? They're places where members of the Stanford community can gather to see a performance, to engage in scholarly conversations in front of an audience. But these are fairly static spaces. A Piazza can be a place where people can find or establish some scholarly enterprises that require deep or wide access to SUL collections and SUL-mediated information sources. (For example, the Humanities Center and the Stanford Humanities Review could open a portal on the Piazza.) This too, is a fairly straightforward extrapolation of network technology.

Designers of scholarly spaces need to augment commercial software to support scholarly "data" analyses which are not anticipated or supported in the wider market. See examples in Scholarly Spaces: Personal research)

Most innovatively, a Piazza could housesymbolic laboratories, provided we develop sufficiently rich structures and function spaces to operate on those structures. ( Scholarly Spaces: Structured data vs. opaque data) In this sense, a Piazza can support a virtual chemistry lab, a space for jointly constructing and manipulating models from quantum chemistry, differential geoometry, astrophysics or economics and certain social sciences. But in order to do this, we need more than the means to trnasmit data in lowest common denominator formats (eg. ascii or SGML), we need interconnecting functions (eg. spell-checkers, indexing, data-plotters, map-servers etc.) , and most importantly a set of languages and writing systems that will let scholars compose new entitities in the Piazza.

Switching perspective, we can view the Piazza not as an architectural space but as a system for writing symbolic structures and for transforming these structures .

These symbolic structures include English, Chinese, Sanskrit, but also QuickTime, ArcInfo (GIS), Mathematica, C, VRML (virtual reality modeling language). A transformation in natural language might be most trivially a citation (URL), but could also be one philosopher's restatement of another philosopher's argument. Transformations in physics go from an English description of a topic like conservative force fields to a mathematical model to a set of numerical simulations.

Just displaying the results of someone else's composition -- whether it be a QuickTime video or a pre-compiled software simulation -- trivializes the scholarly task of teaching students how to think or create in a (inter)disciplinary context. It is not enough just to install a set of tools or applications.

Perhaps most importantly, we should bear in mind that building such experimental augmented spaces cannot be justified merely as "communication" technology. Why? (1) there are much cheaper + effective ways to put people in touch with each other, for example working through ResEd to increase the number and quality of residence-based academic work: music societies, science and language tables, theme house seminars, etc. (2) Communication is not equal to bandwidth or mimetic depiction (See discussion in in Scholarly Practices: Mimetic Fallacy .) To teach people how to identify sophistry, or to understand alternative world-views or conflicting motivations, mimetic depiction is a seriously problematic technical goal.

In fact, "communication" may not be the appropriate category for justifying and motivating augmented spaces at all. A unique, centrally located and centrally administered space does not constitute an experiment in communication. Communication technology lets people "talk" with each other wherever they are using whatever media/modalities are appropriate -- telephone, fax, WWW, video etc. If a goal is to support students' entry into existing scholarly communities or to foster the emergence of new scholarly communities from existing social networks, then we should think about the embedding of unobstrusive information appliances into ordinary places like dorm rooms, dining tables, hallways, outdoor kiosks, and backpacks. In any case, more of such appliances will appear on campus in the near future.

Nuts and Bolts

Some of the nuts and bolts are discussed in Scholar Spaces: Technical Requirements B,C,D. Beyond the obvious necessities such as speedy network protocols (ATM, 1+ GBit backbone...), hardware-neutral and OS-neutral architecture, the Piazza should have several other features:

The Piazza should be a physically comfortable space, with scalable formality, so that people can hold a "non-internet" meeting in that space.

People in other locations, perhaps using a mobile field station, should be able to visit the Piazza. They should be able to project a neighborhood of themselves into the Piazza, perhaps accompanied by some pre-prepared artifact (a privacy screen, a document, etc.)

We should be able to broadcast the Piazza in multiple modes and channels: live video, objects, datastreams, to remote locations.

Distributed object software architecture: subsumes old client-server paradigm, accomodates mixed innovative hardware and design which is not constrained by a single operating system.

Persistent shared data storage integrated with Distributed Computing Group/Networking support. Persistent context-dependent metadata (database) storage.