|
|
PROPOSAL FOR PHILIP GREENSPUN'S CLASS AT MIT (Fall 2003)Organization Name: Discussion Systems (tentative) Contact: Bruce McHenry, bmchenry@alum.mit.edu, 202 460 1492 (mobile) Spirit of Project: Test the viability of collaboratively constructing domains with explicit and directly awarded compensation (such as cash and stock) for contributions. Approach: Extend an annotation system for the web with generalized accounting (the currency could also be academic credit or political capital) to support pay-per-link (with ability to dispute or renegotiate the credit after viewing -- and be assessed for consistent underpayment). Automatically adapt link prominence according to proven value. Enhancements will include processes for raising and managing disputes (to include alerting contributors of high repute) and consensually approving the integration of the resulting corrections and other improvements. The merger of such annotations into their referencing context would also be accompanied by a payout or miniature stock swap to reflect the expected revenue from the merged annotation’s unrealized future clickstream. How important: Central to the mission of the company (a start-up) Why not built yet?: The idea of bi-directional and priced hyperlinks dates to the 1960's with Douglas Engelbart and Ted Nelson. Prototypes were built before the Internet became ubiquitous and authors could seriously consider on-line publication. Success depends on gaining acceptance for a new paradigm for both acquiring and creating knowledge. Readers would have to start actively engaging in shaping content to suit their needs by valuing the components and sometimes comparing alternatives. Contributors would need to focus on integrating unique contributions into evolving domains over which they have no absolute control. What kind of content: Although this approach should be equally valid for almost any kind of intellectual product, the company will be concurrently developing the key section of the business plan -- the identification of appropriate markets with highly valued knowledge. If client content is not ready in time, materials from an OpenCourseWare course could be used. Attracting users: The company will focus on high value markets where there already exists a community of users. This does not imply high volume and the company expects that some of the best clients will be found in areas like investment banking where knowledge has high monetary value to a relatively small community. Long run: The process will tend to create natural monopolies in the domains where applied because the most authoritative resource becomes the one to challenge and, because of being continuously challenged, becomes even more authoritative. However, unauthorized copying into alternative domains will develop when they are better adapted to other languages, cultures or users’ ability to pay. So while the monopoly is natural, it is not assured because copying can be difficult to detect and is impossible to stop. So the defense of proprietary works will depend on maintaining an attractive pricing structure as well as enforcing intellectual property laws and keeping secure the most valuable components. Budget: We aim to raise $.25 - .5MM for initial marketing, recruiting a seasoned executive team and developing a working model. Full development costs are expected in the range of $3 - 6MM and are to be raised by management from private equity sources or contracts to deliver systems based on the core technology. > -----Original Message----- > From: Philip Greenspun [ mailto:philg@mit.edu]> Sent: Tuesday, September 02, 2003 5:05 PM > To: bmchenry@alum.mit.edu; hal@mit.edu; aegrumet@alum.mit.edu; > ben@mit.edu > Subject: RE: proposed client project for 6.171 > > Thanks, Bruce. Great idea obviously but this doesn't really sound > like an online LEARNING community though. I.e., it is not a site > where the > discussion forum would be the most valuable part of the service. > > Philip
Philip, Au contraire, learning communities need to be motivated. Expression is one-half of the learning cycle, and will be more highly motivated if there is meaningful feedback with direct consequences for students' grades, the subsequent view of the "discussion", and even the possibility of discovering an especially effective explanation or learning exercise that propagates into subsequent versions of the course. Chat, threaded discussion and the other content should be integrated by one seamless application. Chat should arise in the context of content -- something to chat about -- and there should be a potentially escalatory process that invites others to join based on the value they put on their time and the importance attached to the issue. Accordingly, conversational moves would be elevated into a more persistent discussion when valuable contributions appear to have been recognized. Furthermore, it should be possible to propose improvements to the corpus, collaboratively refine them with expert advice, and incorporate them by consensus. I think this all has to start with marking up content in a structured way with a system of meaningful credits. If successful, it will lead to a new layer of Internet protocols and human behaviors. The discussions, motivated by the likelihood that valuable contributions will be recognized and rewarded, will be their most valued property. The approach I suggest below will of course need to be expanded and refined. I would be pleased to have the class participate in that process. Bruce |
