Information on Demand

Digital Libraries


Overview


Consumer Needs

Industries primary motives for building broadband networks lie in the economic consideration that there is potential to attract consumers to spend to allocate more entertainment dollars on mass media resources. The market trend as it currently stands indicates that companies building broadband networks are planning to deliver video-on-demand as the primary application. "Industry sees video-on-demand, as the ability to select from among movies, television shows at will." DreamNet, MacWorld, October 1995.

However, according to several national surveys, consumers are much more interested in using emerging networks for information access, community involvement, self improvement and communications. The Benton Foundation's Communication Policy Bulletin 2 illustrates public sentiments also measured by the Mac World survey of October 1994. In fact in the Mac World survey consumers ranked video-on-demand as 10th among services most desired. However, what consumers want and what they are willing to pay for should be the prevalent issue for telecommunications companies to address.

What Consumers Want on the Infobahn

Sample Information from MacWorld October 1994 Poll

RankingOn-line CapabilityWould Pay $10/month or more
1Vote in Elections 20%
2Search Reference Books 22%
3Distance Learning 34%
4Obtain Local School Information 18%
5Search Card Catalogs 19%
8Participate in electronic town hall 13%
10Video on Demand 19%
18Video Teleconferencing14%
25Gambling 4%

The MacWorld data suggests that the consumers surveyed are willing to pay for Information on Demand and consider it a higher priority than additional entertainment services. The Benton Foundation's research illustrates similar public opinions. A survey of 1000 adults to evaluate public opinion on government policy with regards to the National Information Infrastructure revealed similar attitudes as to what use of the new networks should be.


Interactive Mass Communications Trials

Television based interactive media has been around in the United States since 1977. Of the interactive media intended for use by large and diverse groups, two way cable television has been the most visible. The first two way cable television was installed for a trial by Time-Amex in Columbus Ohio in 1977. The Qube system was very crude compared to what can be offered today but it does offer an context in the use of interactivity in a mass media market. The technology was very simple, two household cable lines were installed, one to receive and the other to send transmissions. Offered as an extension to basic cable television, Qube offered users subscribing households an electronic keypad. Viewers could tell the cable television operator which programs they liked and participate in opinion polls. The polls could be rapidly processed and viewers could see some of the results. One other aspect of Qube that was very controversial was it's ability to obtain instantaneous information to compute program ratings. This issue raised concern about people's right to privacy and with the abilities of today' s technology it has only heightened concerns as it could allow private industry, the government and others to invade peoples homes. Qube ultimately failed because of a lack of content and services that were of value to consumers.
Theories of Mass Communication

Another type of interactive television which has had limited success in the United States is Videotex." Videotex is any interactive electronic system which allows users to send data to and receive data from computers or other videotex users by means of a terminal capable of displaying text and pictures." Theories of Mass Communication The basic hardware that users of this system require is a client device, like a cable television box, or personal computer, that are connected via telephone or cable television lines to a server. This concept lies as the fundamental premise of the arrival of another form of interactive mass communication, the Internet. The World Wide Web provides a higher degree of interactivity than early video-tex systems provided and has enjoyed phenomenal growth. It is likely that a major telecommunications provider or cable television operator will offer full Internet capability with TCP/IP access to a large base of consumers within 3 years.

The necessary industry alliances are coming into existence to facilitate these services. Microsoft Corporation recently announced a strategic alliance with Sony, and TCI cable for interactive television services. The first of these service will likely provide consumers with computers high speed access to Microsoft's new information service. Sony, Microsoft near interactive-TV alliance. Consumers can now reach information providers like Compuserve with new modems that allow computer users to interface with an information provider via cable television. This new approach allows the user to take advantage of high speed communications that can be facilitated by the cable operators existing hybrid coax-fiber plants. LANcity Intros Cable TV Modem

This concept could soon go a step further in a directions that will bring these functionalities to consumers who do not have computers. Advanced set-top boxes, similar in concept to those that descramble signals now, can provide computer like capabilities that allow users to interface to these networks with their existing television receiver. Power TV Point and click capability common to multi-media platforms can be facilitated with a remote control giving users access to Internet type services like e-mail and the World-Wide-Web. Commercial services could potentially pay for these capabilities as it will be possible to conduct banking on-line, and engage in commerce from the home.


Economic Opportunity

The incentive for industry to invest in the infrastructure needed to support Information on Demand lies in motivation to capture revenue from other markets, protect existing markets, and potentially lucrative opportunities in new markets. According to Dr. Ben Bennington of the Information Networking Institute, the telecommunications industry stands to gain revenue by tapping into existing markets with broadband networks and is motivated to make the investment to protect their traditional market, telephone services, from competition.

Bandwidth to the Home

Potential Service Markets for Households
ServiceRevenueOther MarketsNew Markets
Plain Old Telephone Service$96.5 Billion?
Long Distance$40.2 Billion?
Local Wireline$40.2 Billion?
Wireless$10.0 Billion?
Cable Television $30 Billion?
Broadcast TV $27 Billion?
Home Video$11.0 Biilion?
Theatres $7 Billion?
Broadcast Radio $8.5 Billion?
Records/Tapes/CD's$7.5 Billion?
Print Advertising $37 Billion?
Postal Services $25.2 Billion?
Video Arcade Games $7.0 Billion?
Video Games Software $6.0 Billion?
T.V. Shopping $2.5 Billion?
Information Services $2.0 Billion?
© B. J. Bennington 1994

Broadband networks will give the telecommunications industry an opportunity to capture a portion of a potential $140 billion annual market that already exists. Getting dollars flowing through the networks is the primary motivation. Companies with broadband infrastructure can generate revenue from providing connectivity to subscribers, information-service suppliers, and by charging overhead by capturing a percentage of commercial transactions over their networks. New services yet to be conceived also offer potential to generate substantial revenue.

Perhaps the most lucrative market for the telecommunications industry to tap lies in the nation's educational industry. "The nation's school and industry together spend between $400 and $600 billion per year on education and training, an activity that is 93% labor-intensive, with little change in teacher productivity rations since the 1800's." Informedia Digital libraries will allow independent self-motivated access to information for learning, exploration, and research. Potential new technologies could bring about a revolutionary improvement in the way education and training are delivered and received.

Hypothesis

The data illustrates that there is an economic case for telecommunications providers to consider the primary motivation for building broadband networks to be a delivery system for information on demand. Previous experiments in Interactive television has demonstrated that content is the key motivation for consumers to utilize information on demand services. Vast digital libraries will soon have to become available on this infrastructure in order to meet demand. The telecommunications providers will find the most challenging aspect to be building content for the digital libraries that consumers are willing to pay for.

Informedia

Informedia is a research project at Carnegie Mellon University funded by the Federal Government and private industry to explore the development of digital libraries. Informedia has been commissioned to deploy a digital library which will encompass 1000 hours of raw and edited program assets from public television stations, schools, and other universities. The content of the library can be enriched with resources drawn from the Internet and other sources of information. There are several significant problems to overcome in the development of a digital library. Informedia must prototype tools to automate the process of building digital libraries, and deal with the problems of content storage, retrieval, and distribution over broadband networks.

Content Development

A unique aspect of Informedia is the utilization of automated, intelligent mechanisms, that are under development to populate digital libraries and structure data in such a way that the library can facilitate a full-content, knowledge based search. Tools under development for Informedia are focused on a process for converting analog video signals into digital video format.

The process goes beyond existing systems for re-mastering video into digital format, as intelligent software can analyze the video to derive additional information about it's content. Video consists of images and audio which in context is a communications process. With Informedia, computers converting materials to digital format, explore the information for content. Audio components are converted by speech recognition systems into a transcript, and are accordingly indexed to each frame of video. Thirty frames of images per second are also explored for Information. Information hidden in the images like words on the screen or within objects are also captured and related to the frame index. Other artificial intelligence research may be able to go a step further, utilizing technologies like semantic nets, to recognize objects like cars, people, or computers adding to the depth of the index.

Image understanding techniques are also employed in the segmenting process. Segmenting is a technique for organizing the data so that it can be stored in a manner from which it can be efficiently retrieved. Video sequences are arranged like paragraphs, Informedia mechanisms automatically locate boundaries for shots scenes and conversations. Storing the information in segmented format allows for additional utilities to be deployed which add considerable functionality with regards to retrieval of the data.

Content Retrieval

Experiments that involve indexing the contents of servers on the Internet have produced powerful search utilities such as Lycos. The indexing technology behind Lycos is a search engine called Pursuit. Pursuit has been employed in Informedia to index the contents of digital libraries and provide users with an efficient means to retrieve information. User controls provide for interactively requesting the corresponding video page or video volume, to "skim" the returned content. The material may be reused in diverse ways and integrated with additional information resources like the Internet.

Interface

Intelligent mechanisms are also being deployed in the Informedia user interface. The current prototype utilize common multi-media computer platforms, Macintosh and Windows. Conventional keyword query composition gives the interface the ability to call information on demand. However technology is being implemented which will go beyond conventional means and Informedia will demonstrate that space age technology can be put into the hands of consumers. The user interface could be driven by speech recognition technology and include controls for context sizing, skimming control and media playback rate. Additional functionality could be provided in the ability to manipulate the content into a multi-media presentation.


Delivery Systems for Information on Demand

Delivering a system like Informedia to a mass market will require extensive integration of technologies and implementation of the most theorectical aspects of broad-band technologies. Several trials are underway which will are implementing the broadband technologies and interfaces needed to support information on demand. The projects take advantage of broadband networks and allow consumers to interface to services with smart cable television boxes.

The Time-Warner project in Orlando, is the nations first full service interactive network. FSN trials are offering consumers a broad range of services, including 500 cable channels, and true video-on demand. Consumers will be able to take advantage of Information on Demand Services in the near future.

             Many future applications that will become available to customers 
             of the Full Service Network next year were also shown for the 
             first time.  These include: educational applications such as 
             access to a local public library; Shoppervision grocery shopping 
             service; Sports Illustrated's sports on demand service; HBO's HBO 
             on demand service; Time Inc.'s The News Exchange, the world's 
             first news on demand service; US WEST's Go TV and U S Avenue 
             services; the Warner Magic Music Store for the purchase of 
             recorded music; and Warner Bros.'s Omnio navigator, which also 
             will be tested as a gateway to the network.
Time Warner's network in Orlando is based on fiber-optic transmission technology, and is the first large scale implementation of broadband technology in the context of information on Demand. Other broadband technologies are under exploration which take advantage of existing telecommunications infrastructure, compression and transmission protocols. According to the acessment in the Mac World, most telecommunicatons companies plan to build interactive-TV systems over the next decade based on the concept of super-cable TV, offering 300-500 channels and limited video on demand services. One fiber-optic cable will be connected to 500 or more homes via coaxial lines. Unfortunately this model lacks the switching mechanims and dowstream capacity needed to faciliate information on demand.

Information on demand, which emphasizes communication over entertainment services, could be implemented with broadband technologies like Integrated Services Digital Network (ISDN). This takes advantage of existing telecommunications capacity as it could be implemented on copper mediums. Unfortunately, ISDN lacks the ability to deliver full motion broadcast quality video, required by the multi-media component of true Information on Demand models.

It is unlikely that the telecommunications companies will build the fully switched fiber-optic networks that would supply enough upstream and downstream bandwidth to support elaborate Information on Demand Services in the immediate future. Several technologies that are being explored show promise. Delivery Systems for Information on Demand


Information on Demand
Copyright © 1995
Carnegie Mellon University