NMIS Project Final Report 1993 - 1997
Executive Summary
The explosive growth of the Internet in the 1990s created enormous interest in
how many services, particularly those involving diverse media, will be
delivered in the next century. In a sense, the Internet represents an entirely
new medium - one that is global in scale, interactive and with tremendous
potential for both economic and social change. However, much of this promise
depends on how interdependent economic, technological and policy issues are
ultimately resolved. While there have been innumerable studies that examine
some distinct technical, economic or policy issue, few have looked at how
different issues in each of these different domains impact on the others.
The Networked Multimedia Information Services (NMIS) Project was as a
collection of experiments and research studies that explored the new frontier
of Internet-delivered multimedia. NMIS focused on what we called the "end to
end" problem, i.e. how multimedia information will be created,
regulated, distributed, paid for and used by consumers. For the most part, NMIS
approached these interdependencies through a series of major experiments that
implemented concrete examples of networked multimedia services. In order to
understand how new, interactive multimedia services might be created and
delivered, we created services which were much more than laboratory-based
"proofs of concept". Instead, the NMIS experiments were much closer to fully
operational, production systems that could be used and tested by numerous
consumers.
NMIS was a joint project undertaken by three universities: MIT, Dartmouth
College and Medical School Carnegie Mellon University, and two corporate
sponsors: the IBM Corporation and Turner Broadcasting operating under grants
from the Defense Advanced Projects Research Agency and the National Science
Foundation. As with any research initiative exploring new Internet services,
the program underwent numerous revisions as the Internet (and the technologies
on which it is based) changed. This report summarizes the results of this four
year long research program.
For the most part, the research undertaken in the NMIS Project centered on
five, interrelated questions:
- How can existing media services be transformed in a way that can exploit the
special properties of the Internet?
- What new information services that utilize the Internet might be created in
the future, and what will the value of those services be to society?
- What new technologies are needed to make Internet delivery of multimedia
information viable?
- What are the economic issues that need to be resolved for delivery of
multimedia to homes, schools and offices to be viable?
- What are the remaining standards and policy issues that need resolution?
The prototype multimedia information services created by the NMIS Project were largely focused on education and training of distinct groups. In particular, NMIS created and delivered three distinct Internet-based services:
- Internet CNN Newsroom - Each school day, Turner Broadcasting produces a thirty minute news show for secondary school children. The content of this program was indexed, digitized, captioned, and delivered every school day via the Internet. Teachers and students could search the accumulated archive of news stories and view relevant stories on demand. An ongoing evaluation study in two test schools was done to assess the educational value of this service was conducted.
- Professional Seminars and Workshops - Seminars and workshops created by the MIT Center for Advanced Educational Services were made accessible through the Internet. The target audience for this service was professionals seeking to enhance their understanding of new, emerging technologies.
- Interactive, multimedia for health care professionals - NMIS developed a multimedia application, "Preventive Medicine in the Combat Theater", is a simulation that teaches physicians and other health care the problems of preventing disease in combat units. This program was tested and evaluated in the Army Training Support Center and the National Guard.
In developing and delivering these services, NMIS researchers often had to
create new technological solutions. Many of these technologies filled gaps in
the current state-of-the-art that existed at the time the service had to be
delivered. Often, the solutions developed by NMIS were then followed by
commercial solutions, some of which were heavily influenced by NMIS work. Some examples of these technical innovations include:
- The creation of a series of software modules that process incoming, analog video, an teacher's guide to the video, and information from the producer of the analog news show to produce a Web page with links to indexed, digitized news stories;
- Server software which can extract segments and tracks from MPEG-1 encoded digital video and export them to clients as MPEG-compliant streams;
- Authoring environments for creating interactive multimedia software;
- One of the earliest streaming digital video players;
- A SGML-compliant markup language called Smart Video Markup Language (SVML) that describes video files, the origins of the segments that make up that video file and copyright information.
In the areas of economics, standards and policies, the NMIS Project:
- Worked with the Society of Motion Picture and Television Engineers (SMPTE) on the design and acceptance of standards for headers and descriptors of video content;
- Participated in the development of Carnegie Mellon University's NetBill
system for secure, verifiable, low cost electronic commerce over the
Internet.
- Collaborated with the Carnegie Mellon University's Informedia digital library
project to examine problems of interoperability between their implementations
and the INTERNET CNN Newsroom service.
- Developed an integrated economic model for Internet Streaming Video.
As this summary report documents, NMIS developed several innovative services
that provide working examples of the potential and power of the Internet. We
are clearly at the beginning of a major change in global communications, and
further experiments such as those fostered by NMIS will be needed before we
come to understand the possibilities that a global, low cost, high bandwidth
network has to offer.