Introduction

Many people have had similar visions of the future whereby people will have access to vast amounts of information via an enormous world-wide network. While this vision is quite compelling, the question arises, "what's holding things up?" The answer appears to be that the technology to implement this vision exists today; however, it will take a great deal of effort to pull it off and no organization appears to have stepped up to the task. The overall goal of this paper is to convince you that electronic publishing is both feasible and desirable.

This paper breaks with the tradition of discussing related works first and defers such discussions until after the proposed system has been thoroughly presented. The overall electronic publishing architecture is presented first, followed by a fairly detailed discussion of each component.

The Electronic Publishing Architecture

The basic proposed electronic publishing architecture is shown in the figure below:

Electronic Publishing Architecture

The electronic publishing architecture consists of five basic components:

Repositories
A repository contains information to be shared. It is anticipated that there will be a variety of repositories providing differing service levels (e.g. faster access, greater availability via duplication, differing storage costs, etc.) Indeed, it does not take too much imagination to visualize repository services competing amongst one another for customers.

Information fetched from a repository incurs a combination of access and royalty fees.

The network
The network is simply the existing world-wide telephone/information network. This network already has the ability to support an enormous number of simultaneous point-to-point connections that can be established and broken on short notice. In general, the cost of a connection is roughly proportional to the distance traversed. Over time, the network will evolve from a system that is primarily geared to providing point-to-point voice circuits into a system that provides higher bandwidth point-to-point digital connections. The network already has multiple vendors vying for customers for long haul services.

Caches
A cache is located relatively close to a number of users and its primary purpose is to reduce communication costs. A cache reduces communication costs by permitting a number of users to multiplex information across a relatively more expensive connection between itself and a repository. In addition, commonly accessed information is cached locally, thereby eliminating the cost and delay of repeatedly accessing it from the repository. As with both the repository and network, a user may have the option of selecting amongst several competing cache service providers.

Viewer
There is one viewer per user of the system; it is a PC-class computer with keyboard, display, pointing-device (i.e. mouse), primary memory (i.e. RAM) secondary memory (i.e. disk), and a network connection (i.e. modem.) In addition, a viewer may have other I/O capabilities (e.g. microphone, speaker, image scanner, printer etc.) The viewer software also provides the ability for people to author information for publication in a repository. Again, users will be able to purchase viewers with differing capabilities on the open market; for bootstrapping purposes, the first viewer software will have to be essentially free.

Banks
A bank is used to both store and exchange money between users on the network. Again, there may be more than one bank, the only requirement is any two users who want to exchange money mutually agree on a bank to broker the money exchange.

There is a great deal of competition built into this architecture to both reduce costs, improve service, and increase innovation.

Paper vs. Electronic Publishing

The economics of electronic publishing are not an obvious win over paper publishing. The goal of this section is to provide a realistic comparison of electronic publishing versus paper publishing.

The way that paper publishing currently works is as follows:

  1. An author writes some material and submits it to a number of publishers to see whether they are interested in publishing the material. While authors with established track records are actively solicited by publishers, new authors are typically presented with a slew of rejections slips. While a new author can choose to self-publish, the large established publishers have already developed the channels for efficiently distributing material that they publish.
  2. A publisher reviews the submitted material and decides which material to publish. Publication contracts are signed with the authors that specify the royalty amounts.
  3. The material is prepared for production; this includes entering it into a machine, proof-reading, editing, etc.
  4. The material is printed on printing presses. The material costs are quite simply paper and ink, neither of which cost very much in the quantities that a publisher purchases them in. The material cost of a thick paperback book is probably less than $.25.
  5. The books are warehoused prior to distribution.
  6. Retail stores place orders for books, the orders are assembled into a single shipment and transported to the retail store. While paper is quite dense, the transportation costs are quite minimal as well. In the area of book-stores, the B. Dalton and Walden bookstore chains are beginning to dominate retail distribution of books, particularly in shopping malls. These two chains add an extra level of warehousing and distribution.
  7. The retail store places the books on the shelves (usually for list price.) The retail store has to have enough mark-up on the book to pay both salaries and store rental fees and still have enough left over for profits and taxes.
  8. A customer browses the shelves looking for interesting books and purchases the ones that look interesting. If the customer pays by credit card, roughly 1% of book cost goes to the credit card company.
  9. For paperback books, any books which do not sell have there front covers ripped off and returned to the publisher for a refund; the rest of the book is disposed of, since it is not worth the hassle of transporting it back to the publisher.

Thus, costs of paper publishing can be roughly broken into:

Given that a typical paperback book currently costs approximately $5.00 and both the material and transportation costs are quite low, there is sufficient money left over to pay for the rest of the costs. The paper publishing model has been finely honed over the centuries and will not be easily displaced by electronic publishing. Luckily, it is not the immediate goal of electronic publishing to displace paper publishing; instead, the goal is to provide a viable alternative to paper publishing.

Electronic publishing works as follows:

  1. The author writes some original material.
  2. The author chooses a repository to store the material in and self-publishes it. The author selects the royalty rate to be charged. The repository cost for storing the amount of material in a thick paperback book is pretty minuscule. There is a tension between the author wanting charge a high royalty rate and the users wanting to pay as low a royalty rate as possible.
  3. The material is entered into a variety of indices (i.e. author, title, subject, etc.) The author submits the material to critics for review in an attempt to get the word out that their material is worth reading.
  4. A user browses indices and/or review columns and follows a hypertext link start reading the author's published material. If the material is interesting, the user will read it in its entirety and the author will collect a complete royalty; otherwise, partially read material only results in partial royalty payment.
  5. The user may choose to print out their own personal copy that is both more portable and probably easier to read.

The costs of electronic publishing can be roughly broken into:

While it is quite possible that a user will be able to read a book for less than the cost of purchasing the same book in a bookstore, only the most voracious readers will be able to save enough money reading electronically published material to entirely offset the current cost of a PC; instead, electronic publishing will initially appeal to those people who have already purchased a PC for other reasons.

Paper publishing differs from electronic publishing in the following additional areas:

Obviously, both forms of publishing have both advantages and disadvantages over the other.

Electronic Publication Legal Issues

In the United states, the following legal issues surround electronic publication:
Common-carrier vs. broadcast law.
Currently, there are two broad categories of law concerning communication - broadcast and common carrier (point-to-point) law. Examples of broadcast communication are book publication, newspapers, radio, and television; similarly, examples of common carrier communication are the mail system, air freight system and telephone system. For broadcast communication, the laws hold the broadcaster entirely responsible for the content. For common carrier, the law attempts to ensure the privacy of the communication (i.e. the phone company is strictly disallowed from listening in on a conversation without a court order.) Despite the fact that electronic publishing is not a very good fit for either common carrier or broadcast law, the legal system will attempt to apply these laws to electronic publishing. Ultimately, new laws will be needed to deal with the legal issues that are unique to electronic publishing.

Interstate commerce and banking laws.

Since electronic publishing requires the use of electronic banking for royalty and service payments, the federal and state laws concerning interstate commerce and banking are applicable to electronic publishing.

Copyright law.

Copyrighted information can be stored in a repository. However, since copyrights disallow either electronic storage or copying, a waiver is required to allow the repository to electronically store and distribute the copyrighted information. Non-copyrighted material (i.e. public domain) may also be stored in the repository without requiring a waiver. Whenever a court of law determines that a copyright has been infringed upon, the repository must be prepared to follow court instructions which are likely to require removal and/or royalty reassignment.

Libel law.

All information that is stored in a repository must have someone who is responsible for its content in the event of a libel suit. Some repositories will act as publishers and bear full responsibility and other repositories will require that the authors bear the full content responsibility. In the event of a libel suit, it must be unambiguous whether to sue the repository or author. Whenever a court of law determines that information is slanderous, the repository must be prepared to follow the directions of the court, which will frequently insist upon removal, public retraction, and/or damages.

Even though an author and repository may have mutually agreed that the author has sole responsibility for some published information, a court of law is not legally bound to accept such an agreement. Indeed, it is quite likely that plaintiff's lawyer will decide to sue both the author and repository for damages in a libel suit. Unfortunately, it is quite likely that some jury will eventually find the repository at fault and thereby establish case law that repositories are always responsible for their content, irrespective of any mutual desire between the repository and the author to exclusively assign content responsibility to the author. It is quite likely that legislative action will be necessary override any such case law.

Community standards.

Historically, the United States has permitted communities to exercise some control over the information that is published in the community, primarily in the areas of pornography and religious beliefs. Since the telephone network is world-wide, it will be very difficult to identify any world-wide community standard. However, there needs to be a mechanism for organizations and/or communities to identify information that is unacceptable. For example, parents should be able to restrict and/or disallow their children's access to material that the parent deems inappropriate.

How the legal issues surrounding repositories are eventually resolved in the courts will have major ramifications as to how successful electronic publishing eventually is.

Political Issues

Electronic publishing is likely to annoy and/or threaten existing publication channels. These channels are likely to exert significant resources to impede and/or stop the viability of electronic publishing. Two examples where existing industries expended significant effort to impede a new technology are the movie industry's resistance to television and the recording industry's resistance to digital audio tape (DAT); for these two examples, television ultimately established a synergistic relationship with the movies and digital audio tape has yet to gain a significant foothold in the United States.

One of the most effective strategies to reduce the political risks associated with electronic publishing is identify those organizations that are likely to be threatened by electronic publishing and work with them to address their concerns with electronic publishing. For example, it may be possible to convince the newspaper industry that electronic publishing might provide an extra source of income if the newspaper is electronically published in addition to being printed on paper. As another example, the audio recording industry might be able to distribute samples of music via electronic publishing and thereby increase overall music sales.

In these days of political action committees and special interest group lobbying in Washington, it is naive to think that a new technology such as electronic publishing is going to survive a concerted attempt by any special interest groups interested in its early demise without a significant counter effort to protect it. Any organization that undertakes to push electronic publishing forward must be equally prepared to undertake the political battles that will ensue in Washington.

Will Electronic Publishing Actually Happen?

Is it likely that electronic publishing will happen? People who are already predisposed towards electronic publishing will conclude that electronic publishing is the wave of the future and entirely replace paper publishing in short order; conversely, people are predisposed the other way, will conclude that electronic publishing does not have a chance and is not worthy of further consideration. Neither of these extremes is likely; instead, electronic publishing is more likely to be adopted gradually. Electronic publishing's first users will be out-right enthusiasts, followed by a core of professionals (e.g. engineers, doctors, scientists, etc.) needing access to material that is usually only found in university libraries, followed by public library users, etc. It is likely to be a long time, if ever, before electronic publishing is actively competing with paper bookstores.

A useful analogy for the adoption of electronic publishing is to think back about the adoption of he personal computer. The original personal computers (e.g. the Apple II) was quite primitive; it had a lousy keyboard (upper-case only), lousy display (a television), lousy storage (a cassette deck), a lousy processor (an 8-bit 6502), limited memory (less than 8KB), and was quite expensive as well. Initially personal computers were only used by enthusiasts, followed by early adopters in the business community (e.g. VisiCalc users), and eventually over time, the personal computer has evolved into an indispensable tool used by business and industry. It seems likely that electronic publishing will follow some sort of quirky path to general acceptance as did the personal computer.

You may read the next chapter discussing viewer software or return to the table of contents.


Copyright 1992-1995 Wayne C. Gramlich All rights reserved.