Authors write and publish because they want to be read. Many authors also want to earn a living from their writing, but for some, income from publishing is not an important consideration. Some authors, particularly academics, publish because of the status, prestige, and professional advancement that accrue to authors of influential or groundbreaking works of scholarship. Academic publishers have historically taken advantage of these motivations to create journals and monographs consisting largely of works for which they pay minimal royalties, or more commonly, no royalties at all. In return, authors' works receive professional review, editing, and formatting. Works that are accepted get placement in widely circulated journals and monograph catalogs.
In the late 1970's and 1980's academic libraries became acutely aware that an expansion of research activity had resulted in the growth of both the numbers of journals and the numbers of articles published in the journals. The combination of increased subscription prices and the number of journals needed to support research resulted in a so-called "serials crisis". Libraries were forced to cancel subscriptions. The reduction in circulation forced publishers to raise subscription prices further to make ends meet, and the resulting cycle of cancellations and price increases led to a fear that the whole system would collapse. If few libraries could afford subscriptions, fewer scholars would be able to read the articles, diminishing the attractiveness of publishing.
The advent of web-based publications in the 1990's led many to believe that the solution to the serials crisis would be a shift of the scholarly publishing industry to so-called "open-access" business models. Open-access publications are those that can be read at no cost to the reader or the reader's institution. The traditional model of publishing supported by subscription fees was thus styled as "Toll-Access" publishing. It was hoped that the combined cost reductions from digital distribution and automation would stop the cycle of rising expenditures.
Perhaps the most successful implementation of open-access has been ArXiv, at http://arxiv.org. ArXiv is a database of digital preprints and reprints ("e-prints"), originally focusing on the particle physics community. Originally started by Paul Ginsparg, a physicist at Los Alamos National Labs, ArXiv is now located at Cornell University and hosts more than 670,000 scientific articles in e-print form. Authors deposit articles they've written into the repository, and other scholars are free to search, browse and download articles without needing any sort of subscription.
One reason for the success of open-access archives has been that they have grown up in a parallel coexistence with the traditional academic journals, which have mostly shifted onto the web. In the so-called "Green" model for open-access, many journals allow versions of accepted articles to be made available via repositories. Authors can thus submit their articles to high-prestige subscription-supported journals without worrying about colleagues' access, because scholars that need to read their works can always access versions from free sources.
Meanwhile, the shift of traditional journals onto the web has allowed the rise of secondary distribution channels. Most academic libraries today enjoy access to a much broader range of journals compared to 20 years ago because of the availability of article databases that aggregate content from large numbers of journals.
The past decade has also seen the rise of "gold" open-access journals. Some of these journals finance themselves by charging publication fees to submitting authors, such as Biomed Central at http://www.biomedcentral.com or PLoS at http://www.plos.org. Others cover expenses with private or public funding, such as SciELO at http://www.scielo.org. They build prestige and avoid becoming "vanity" presses by establishing rigorous review processes.
The success of open-access journals and articles has for the most part not yet been duplicated in the word of books. There are a number of possible reasons for this. The first is the matter of cost. Publication fees for open-access journal articles are in the range of $600-$3000; editing and production expenses for a book published by a university press are estimated to be a lot more. While a book that's mostly text might cost as little as $10,000 to produce professionally, a book with figures, photos, equations and cover art will cost a lot more to edit, design, and produce. Author-funded publication fees this large are unlikely to be practical, even with significant institutional subsidies.
Another factor holding back open-access books may be a preference for print books over e-books. Books are much longer than journal articles, and many readers are uncomfortable reading a book on a computer screen. It's only in the past two years that dedicated reader devices such as the Kindle and tablet computers such as the iPad have improved the e-book experience enough to gain wide consumer acceptance.
The business environment for book publishers is another possible factor. The university publisher loses money on much of its catalog, but compensates for this by having one or two titles that cross over to be successful outside the academic environment. Amazon -- http://www.amazon.com -- has bolstered this pattern, by providing wide distribution for small print-run titles that would never have been available in bookstores before. In contrast, journal articles almost never cross over into non-professional markets.
Nonetheless, there have been a few notable attempts to publish open-access e-books. I'll cover these later in a section on business models for open-access e-books, but it wouldn't be right to omit mention of Project Gutenberg, http://www.gutenberg.org, at this point. Project Gutenberg (PG) produced not only the first open-access e-books, it produced the first e-books, period. Started by Michael Hart in 1971, PG aimed to take the text of public domain works and make them available via the Internet. To date, PG has put over 34,000 works into its collection, entirely through the efforts of volunteers.
Distribution of open-access e-books can be thought of as an enterprise separate from their production, since the costs involved are of a different nature. The scaling laws of Internet distribution favor centralization, and as a result, organizations such as the Internet Archive are able to distribute appropriately licensed e-books on a vast scale; businesses such as Google are able to search and organize them; libraries, blogs, and portal sites are able to select and "curate" them. To some extent, this type of distribution depends on the self-contained nature of the book; it shouldn't require the context of a specific website to retain and accumulate value.
Open-access for e-books provides many benefits in addition to allowing people to read for free. Access to the full text of books makes for more complete indexing. The utility of Google Books, and the effort Google has put into digitizing books from libraries, even when they are unable to make the books available because of copyright, is testament to the value of indexing the full text. Long-term preservation of our cultural heritage is another public benefit of open-access to e-books.