Tik-110.501 Seminar on Network Security
"Electronic Commerce"
November 20, 1996
Petri Wessman
Akumiitti Oy
Petri.Wessman@akumiitti.fi
The emergence of the Internet (and other large-area networks) as a new publishing medium has opened up more than one can of worms with regards to traditional publishing. Since anyone who has access to the network can "publish" a document (which is not limited to text) at virtually no cost, the concept of "publishing" must be examined in a new light -- as must the concepts "copyright" and "library", among many others. This document examines the history of online and electronic publishing, some of the issues involved and possible future trends.
Things used to be simple. A writer would have an idea, he or she would write a draft of the work, offer it to a publisher, and finally -- if he/she got lucky -- the publisher would publish the work. Of course, the story doesn't end there. The published work has to be shipped to the distribution network, there might be marketing expenses in order to make potential buyers take note of the new work, and finally, the book store on the other end of the chain has to make decisions on how many copies to order. Too few, and he risks running out of a hit title. Too many, and the unsold piles will accumulate dust and take up valuable storage space.
Admittedly, the above traditional scenario is not exactly "simple", but in contrast to the new emerging ways of publishing information -- not just books -- and the changes in the publishing infrastructure, it is at least easy to understand. Things are changing, though, and changing fast. The unforeseen explosion of the Internet both as an service and as a phenomenon leave little room for publishers to cling to their old ways. In some quite fundamental ways, publishers must adapt or risk being left behind as new forms of media steal away potential customers.
One could make the point that the Internet is not really that big a deal as regards to publishing. That would be true if the Internet is seen only as technology, but it is really a question of much more than that. In a way, the Internet is a meme, a "thought virus", that changes the way the people who use it view information and the world around them. National and cultural boundaries lose meaning, and the pervasive globality on the Net, together with the increase in multimedia products for inexpensive home computers, helps create an infrastructure for creating globally accessible multimedia. Is something still a "book" if is exists only in electronic form and consists of a spider-web of links to text, images and sound? What does the word "publish" mean in that environment? And, of course, where's the financial profit in all this? [Fill93]
Traditionally, before the age of the inexpensive personal computer, printing a book (or some other publication) has been a difficult and technical process. After the source text has been produced, it must be displayed in a form that is pleasant to read and suitable for the publication in question. This means, among other things, that pictures must be laid out near the text segments that reference them, a table of contents and an index must be generated, a fitting font and text layout style must be selected, and a multitude of other design options decided on. In the best case, the end user would notice none of this -- good typographic layout is seldom intrusive -- but would "merely" find the book to be "a nice read". Bad design, on the other hand, usually sticks out and can totally undermine the effect of the text itself. Traditionally, this layout process has been in the hands of professionals, and the layout itself has been performed with specialized text layout languages with steep learning curves.
The onset of the "Desktop publishing revolution" changed this. Whereas before you had needed extremely expensive equipment and extensive know-how to produce professional-quality output, now personal computers armed with WYSIWYG word processing programs and inexpensive laser printers made it possible for just about anyone to produce commercially viable output. This was not always a good thing, since while the new technology made the creation of high-quality documents possible, it did nothing to advance the artistic qualifications of the creator of the document. The early stages of the "revolution" saw a glut of garish documents which apparently tried to fit every single font that could be produced by the machine onto every page, resulting in documents that were better in inducing headaches than conveying information. As the technology and the skills of the users matured this became less of a problem, most users began to understand that in producing readable documents, less is often more.
The spread of high-quality document processing also raised the stakes, so to speak. Before desktop publishing, a small company could well produce material using an old fashioned typewriter, or a dot-matrix printer. Now, you had to be able to produce professional-looking output in order to appear convincing to potential customers and partners. Suddenly, a computer became a necessity instead of a high-tech toy.
As usually happens with something truly useful, desktop publishing became established technology fairly quickly, and developers started to look in other directions. One of the most interesting new directions came from the fact that traditional text is "flat". In other words, it's designed to be read either sequentially from the beginning to the end, or used as reference text (i.e. dictionaries etc). If you encounter a new concept while reading the text, you must look it up in another book, or possibly on another page, resulting in page flipping and often forgetting where you were in the first page. Wouldn't it be convenient if you could embed something in the document that would automatically link, say, the word "palimpsest" to its dictionary definition (and possibly a picture)? The concept of "hypertext" was born.
The word hypertext was coined by Ted Nelson, and was published for the first time in a paper delivered to a national conference to the Association for Computing Machinery in 1965. Though fairly computer-illiterate at the time, Nelson had always been fascinated by systems for manipulating documents in non-linear ways and for storing "random" information for later access. This interest may have been sparked by his own condition, a psychological syndrome recently named Attention Deficit Disorder. It made him extremely vulnerable to interruption of any kind, and prone to forget things instantly. In his paper, he envisioned a feature called "zippered lists" which would link related or identical elements in two texts with each other at word, sentence, paragraph or document level. The paper received little attention, but Nelson continued developing his vision. A vision which would become "the longest-running vaporware project in the history of computing", Xanadu [WJu95].
Deriving its name from Samuel Coleridge's 1797 poem fragment "Kubla Khan" (which is, ironically, a prime example of literary vaporware), Xanadu became Ted Nelson's obsession and dream. It was to be a global hypertext system, which would eventually encompass the whole of human knowledge and allow easy access to it. One of the key ideas of Xanadu was "transclusion", by which the builders of Xanadu meant the ability to quote all or part of a document without copying. In a Xanadu system, there would be only one copy of each document, and all other "views" of it would be handled by the transclusion mechanism. This would permit the original writer of the document to receive royalties each time his document was referenced. Xanadu was also intended to permit anyone, not only the writers of a document, to create links to and within it. This would permit global annotation, so different readers of a text could add their comments and views (and pointers to further information) without changing the original text. Xanadu was to be a universal library, a global information index and a computerized royalty system rolled into one.
The vision of the Xanadu project was grandiose, but unfortunately it crashed into the brick wall of existing technology. The technology that Xanadu needed doesn't exist now, let alone in 1979 when Ted Nelson, Roger Gregory and others started working on it. Working with what was then high technology (an Onyx with 10 megabytes of hard disk space and 256k of RAM!), Nelson and friends were trying to create a global information infrastructure on a machine that could barely edit a single large document. However hopeless their task, they did develop some interesting algorithms for data storage and retrieval -- some of which Nelson refuses to divulge to this day. Whether they exist and could have been the firm base that Xanadu would have needed to succeed is anyone's guess, but in reality the project (or "saga", some would say) led only to unmet promises of wonderful new technology "just around the corner". Met with lack of cash and failure of the people involved to deliver any sort of working prototype in any sort of reasonable time, Xanadu crept back to being what it had started out as: a dream. In the end, the one problem that the Xanadu developers never could get around was simply the lack of bandwidth. A system like Xanadu would need immense bandwidth, making the current Internet bandwidth needs of the World Wide Web pale in comparison.
Ted Nelson was by no means the only -- or the first -- person to toy around with hypertext-like systems. In 1945, Vannevar Bush published an article in The Atlantic Monthly called "As We May Think". In it, he proposed a system called "memex". Memex would allow the creation of personal indexes from documents and would enable users to link sections of text together with "markers". Memex was at that time a purely theoretical idea, and resulted in no actual development work [WJu95].
Various demos and prototypes of hypertext systems were built over the years, and not all were concerned with text documents. At MIT Media Lab the first hypermedia videodisk, the "Aspen Movie Map", was created. In the Aspen map, the car-level street views of the town of Aspen were digitally recorded on videodisk, so that in the application the user could wander around the "virtual Aspen" and get realistic visual feedback. The Aspen map also represented one of the first experiments in the now-popular (and possibly over-hyped) field of "virtual reality".
The basic idea of hypertext (i.e. that you could create links within a document which would allow the user/reader to navigate through the work in non-linear fashion) slowly gained acceptance in the computer industry. Hypertext-based documentation became more and more common, with 1987 seeing the introduction of the HyperCard application by Apple. HyperCard was the first hypertext product intended for the end user, and various documents soon started to appear as "HyperCard stacks". However useful, HyperCard was only a local application. People waiting for a global networked hypertext system had to wait until 1990, when a small group of researchers at CERN unleashed the World Wide Web on an unsuspecting Internet community [W3].
The World Wide Web started as an experiment in distributed information management at CERN. The intent was to provide a uniform way of presenting documents so that anyone with a suitable client program could view them, regardless of the platform. The documents were to be hypertext documents, so that references could be added and a semi-uniform "web" of existing knowledge could be built. Traditionally, information had either been arranged as a strictly hierarchical tree (directory), or as a mass of loosely organized documents with a keyword-based search engine as a front-end. Neither of these solutions was seen as flexible or easy to use enough to help solve the problem. The problem, and the key reason for the creation of the WWW project, was the fairly high turnover rate in researchers at CERN, creating a need for a single source of information toward which a newcomer could be directed [W3].
It worked. In fact, it worked much better than anyone had anticipated.
By now, most people who know of the Internet are also familiar with WWW. In fact, a large number of people think that WWW is the Internet, an illusion that the current Internet products do little to dispell. Whether WWW has true lasting power remains to be seen, but it is clear that there was a need for something like it. It's not perfect, but it does the job.
But what is the WWW? In short, it consists of WWW servers running on Internet-connected hosts, which fetch documents located in dedicated WWW directories and send them to WWW clients via HTTP, the HyperText Transfer Protocol. Typical client programs in use today are Netscape Navigator and Microsoft Internet Explorer, but the competition between client programs is fierce and it is difficult to determine the long-term winner. The job of the client is to display the document (written in HTML, a simple text-based formatting language based on SGML) in a suitable form for the client system. Typically, this means showing text in a variety of fonts, adding hypertext links to the indicated places, formatting pictures into the document, and much more. In its first incarnation, HTML was a very simple language, but current needs for online animation, distributed programs (Java), more control over the layout process, and support for user feedback have transformed the language into something that isn't quite so simple anymore. Whether HTML will live on or collapse under an ever-mounting pile of "features" is a question for the future, but again, "it does the job". It's still reasonably simple, it enables the creation of hypertext (i.e. linked) documents, and it provides a uniform format for presenting and exchanging documents.
It is interesting to note that in some ways the WWW and HTML represents a step back from desktop publishing. In the early days of the Web (as it is often called) the only way to create documents was to learn HTML, a text formatting language. You needed a separate client program (and a HTTP server) to actually see the result. Various WYSIWYG-style HTML editors have slowly appeared on the market, but writing "raw" HTML is still often the only way to get the desired result. It is to be expected that the need to know HTML will disappear in the near future, at least for normal document creation, and people will simply have word processing programs with the option of "publish this document on the Web". Which, of course, will make it easier than ever for private individuals to publish their work directly for global consumption.
In comparing the WWW to earlier visions of hypertext, it is interesting to note the differences. When compared to Xanadu, in particular, one notes that the WWW is still a primitive "organism". There is no (easy) way of saying "show me all the documents that have links to my document", nor is there any mechanism for including pieces of other documents in your document, other than providing links to them. The WWW works on the same decentralized, anarchistic model as the Internet -- a model that, despite its drawbacks -- works surprisingly well in "the real world". Whether the current-day Web with its often agonizingly slow response times and primitive user-feedback mechanisms can evolve into the information infrastructure of the future remains to be seen. Whatever one's opinion of the Web, it cannot be denied that it has started something.
The "Internet explosion" has caught the publishing world largely by surprise. Suddenly, it is no longer enough to publish, say, a reference work and (perhaps) update it in a few year's time. Users used to having online databases at their disposal are slowly but surely developing a need for information that is actually up to date. An encyclopedia or dictionary does not age much, but a printed reference work on movies, for example, is out-of-date the moment it is printed. New movies and TV films are being made daily worldwide, no publication in book form can hope to keep up with the pace.
Multimedia CD-ROMs, while offering much more versatility than books, are still plagued with the same problems: information gets out of date, and a mechanism for getting new, updated versions of the CDs must be created. Also, most titles billing themselves as "multimedia" nowadays tend to be of questionable quality -- some publishers seem to think that taking the text of a book, adding a few images and video clips, and pasting the label "multimedia" on the package is sufficient. Obviously this is not the case, and the publishers who both realize this and start to actively create innovative multimedia products will have the upper hand in the market struggle to come [Fill93].
Besides "static" products such as books, video tapes and CD-ROMs, a new category of online publications is slowly evolving. Among these are the so-called "online magazines", which may or may not have "hardcopy" counterparts. In some cases these are merely online look-alikes of the printed product, but in other cases they are quite different, with emphasis on making use of the constantly evolving nature of the Internet. For example, some columns may be updated daily, and reader responses to articles added next to the article itself as hypertext links. Examples of current-day online magazines are HotWired, OMNI and JavaWorld.
Of course, all this new-found diversity means that a publisher wishing to offer cutting-edge products must make substantial investments in computer and network technology products and personnel. In some cases this is reasonable, especially when a large publishing house is in question, but in other cases this could well be the straw that breaks the camel's back. The solution in this case is either to ignore the new possibilities and concentrate on the traditional areas, or to find suitable third parties with the know-how to create products with the new technology. While it is tempting to retreat to traditional ways of operation, electronic "books" and online services have a lot to recommend them. An electronic version of a book -- especially an online one -- can be reproduced at virtually no cost (making over-stocked warehouses a thing of the past), takes very little (physical) storage space, and makes realtime updates to the material possible [Raw91].
What are the drawbacks of all this? Besides the need to learn new technology, the online world operates on very different market patterns than the "hardcopy" world. Users of online services are used to getting information for free (or for very little cost), a publisher attempting to charge the same amount of money for an electronic version of a book as for the "normal" version is in for a hard time -- unless, of course, the publisher has added some value to the product (images, hypertext annotations, video clips, etc). If an electronic version of a product costs substantial amounts of money, the temptation for users to just copy the work from a friend etc will probably become too large. After all, copying a entire book with a copier is a slow and clumsy process, while an electronic copy is (in most cases) extremely easy to make and, even more important, identical to the original. Various copy-protection schemes are possible, but they tend to cause more harm than good -- few users want to type in cryptic pass-phrases each time they use an electronic encyclopedia, for example. At present time, the vast amount of information on typical CD-ROMs acts as a fairly effective copy protection scheme, since the hard disk space needed to store the copied CD-ROM usually costs more than the CD-ROM itself [Fill95]. Most true copy-protection schemes of future online products will most likely be based on strong, public-key encryption [Crypt][Raw91].
Online magazines are another problem in the sense that users no longer buy "copies" of the magazine, they browse the issues at will. To generate cash from this, a publisher can either resort to selling advertising space (this seems to be fairly popular at the moment) or create some kind of billing scheme (this usually requires that users register themselves with magazine before they can read it). Before a secure electronic cash system that supports arbitrarily small transactions is widely in use on the Internet, it is unlikely that pay-to-read services will succeed. As another option, some online magazines operate as a sort of advertisement for the printed version, saying in effect "here's a condensed version, for the full version you have to buy the issue".
What about books? The idea of an electronic book is an old one. In 1971, Alan Kay at Xerox PARC (Palo Alto Research Center) developed the concept of a dynabook, a small, portable computer with a high-resolution screen and an electronic pen for input. It would be able to display two document pages at a time, and one could load new data into it via a radio link [Raw91]. At that time, the dynabook as an idea was a mere curiosity, but nowadays it would be possible to make them. The problem is, would anyone use them? The questionable sales of the so-called "personal digital assistants", the Apple Newton for example, have given reason to doubt the market possibilities of electronic books -- at least until display technology improves and the prices come down.
Some companies have started "online bookstores" and are distributing digital versions of books [OBS]. Sometimes these works are available only in digital form, while at other times the printed version is available. Project Gutenberg, a volunteer organization, has texts of numerous (mostly classical) books available online free of charge [Gut]. In the future, the number of books available in electronic form is sure to rise.
All the hype around multimedia, the Internet and other new media forms serves to cloud one important fact: printed books are here to stay for the foreseeable future. A printed book has an extremely high resolution, needs no power source, can be carried along easily, is not vulnerable to shocks or extremes of temperature, and can be browsed far more easily than a typical multimedia product. When most computer users want to read a piece of text stored on computer, more often than not they print it out on the nearest laser printer and sit down to browse the resulting stack of paper. Why? Some of it may just be old habits, but in most cases reading from paper is much more pleasant than reading from a computer screen. Books won't go away until we get technology that offers all the benefits of printed books.
What is considered the "default media" depends on what one is used to, of course. A piece of modern folklore tells of a father showing an encyclopedia to his daughter. The young girl is puzzled for a while, but suddenly brightens and exclaims "Oh, I see! It's like a printed Encarta!" (Encarta is a popular multimedia CD-ROM encyclopedia).
In the old model, the notion of "publishing" something was fairly clear-cut. A publication was "frozen" the second that it went to the printing press, all updates and corrections had to wait for the next edition. With global computer networks, the issue becomes anything but clear. On the WWW, for instance, the phrase "this page is under construction" is so common that it is considered bad style to even mention it -- the assumed default is that the work in question is constantly evolving. In an environment like this, what does it mean to "publish" something? Is it the act of placing the work (which may or may not be "finished") on the Web? Is something published only when it is "finished"? Is anything ever "finished" in reality?
The basic difference is that instead of getting a static snapshot of a product at a certain stage (a printed book represents the stage when both the writer of the work and the publisher considered the work ready), we now get a view of a constantly evolving object. Works are no longer necessarily "write once and forget" projects, and the creator can update and edit the work at will. This is most obvious with online magazines, and in some cases the "publish"-oriented notion of an "issue" has almost disappeared, being replaced by a constantly changing "mass" of articles, editorials, and reader feedback.
This is not always a good thing. For one thing, it makes references to the material more difficult. With printed works you could always refer to the hardcopy book or magazine issue in question, but with networked versions this is more ambiguous. Another, potentially more serious, problem stems from the fact that online publications are notoriously "under construction". How can you tell whether the work you are reading is a dependable source of information or merely a draft outline that may change completely the next day? In some cases, you cannot. To this end, it would be reasonable for commercial entities to only "publish" material that is considered "ready", on the same criteria that is used for printed works.
The idea on annotating and adding to works is not a new one. Numerous "annotated" versions of classical books have been published, with someone's (or many peoples') comments added to the original text (often with the intention of providing background or interpretation). The idea of being able to do this in an incremental, online fashion is tempting, but to actually work the process needs a coordinator (an editor of sorts) to make sure that the annotations are reasonable and accurate. An online work that allows update by the readers of the work begs other questions: what about the ownership of the work? Who owns and has rights to a multimedia work that includes comments (and possibly articles) from the readers? In the future, it is probable that in many cases the authorship of a hypermedia document is not a clear-cut case, with the document having been developed into its present form by an informal group effort. If the document is in the public domain this probably won't be much of a problem (as long as all the participants get noted somewhere!), but with commercial works the problem moves into the quicksand of computer copyright laws [Fill94].
All in all, the emerging global network technology will slowly but surely push publishers into new fields of operation. Instead of selling "things", they will be selling ideas and content. The commodity will no longer be physical, and won't necessarily even be static. Some publishers will probably turn into "data warehouses", getting their profits from the collecting and organizing of information -- be that information text, images, video, or virtual reality scenarios. In this sense, the role of publishers will somewhat converge with that of libraries, one operating on a commercial and one on a "community supported" level [Fill93].
Since one could say that the currency of the future is information, the ownership information becomes an extremely important issue to anyone wishing to make a financial profit. Unfortunately, the laws regarding copyright were born at a more "analog" time, at a time when making exact copies of works wasn't so easy, and distributing the copies worldwide even easier. Nowadays, the illegal copying of information -- be it computer data, music or whatever -- is so commonplace that the old copyright laws seem to have very little impact. Without going into moral issues, the idea of making copies of digital products is very tempting. Instead of paying substantial amounts of money for something, you can get an exact copy of it for free. It's a notion that is very hard to resist [WMa94].
The Internet has created even more problems for the copyright concept. Different countries have different laws regarding copyright, what is illegal in one place may well be perfectly legal somewhere else. Since the Internet is a global entity, the "location" of data found on it is sometimes hard to determine. This has resulted in some amounts of "data offshoring", with users moving their data to "places" on the net where legal limitations are less strict. For example, the outdated U.S. export regulations on cryptographic software has resulted in a lot of U.S. crypto development work being done on Internet hosts outside the U.S. The fact that the laws of various countries have only now begun to take the Net into account has resulted in numerous strange and morally questionable court rulings. For example, people have been charged with distributing obscene material on the Internet even if the host on which the material is located is in a country/state that permits the material in question -- in other words, making something available on an Internet host has been equated with "distributing" the material to every point in the Net. It is to be hoped that the eventual integration of the Internet into modern-day society will result in saner laws about its use.
Whatever happens to the copyright laws, the problem of digital products being too easy to copy remains. What can be done if one wants to make money with digital products in the future? The original "information must be free" ideology of the Internet makes this an especially difficult question [WMa94].
Several strategies have been tried:
Networks and the digitalization of documents is also starting to have an impact on libraries. A library used to be a storehouse for printed books, but nowadays it is often much more. Typical modern libraries have Internet access points, computerized search engines into the catalog of available books, and a variety of electronic documents available on CD-ROM for browsing. The material that you can loan from a library isn't limited to books, either. Music CDs, films on video tape, and other modern media are often available. A library is not about books anymore, it is about providing public access to information -- whatever the form of the information.
With this model, it is natural for libraries to extend into the networked realm. A library could offer interfaces into its catalog of available titles through the Internet, and perhaps allow users to reserve books through the Net. There are very few technical problems with this, but a large number of non-technical ones.
At least in Finland, the bureaucratic nature of library management, combined with extremely limited working budgets, has severely limited the developing of "virtual libraries" to complement the physical ones. Most people in charge of library development and funding are out of touch with modern technology -- a telling example is the fact that until very recent times, no official paper about the future directions of Finnish libraries even mentioned the word "Internet", let alone proposed using it. In part, this is probably due to the traditional Finnish government policy of ignoring the existence of the Internet and desperately trying to develop other "official" networks that (unsuccessfully) mimic the functions of the Internet. This policy is slowly changing, perhaps someone finally realized that the Internet will not go away just because it is officially ignored. A large "fear of the unknown" factor must also be taken into account, the numerous horror stories about the insecurity of the Internet have probably given some people a biased view [Kuro95].
When and if libraries get funding in order to expand into the online world, they are faced with numerous choices. An extreme possibility would be to go completely digital, and offer only a virtual library with digital documents to users. In some cases this is justified (Project Gutenberg is a nice, if limited, example of a virtual library [Gut]), but in most cases there is still need for a physical library. Printed books will not vanish overnight (if ever), and a modern library serves many needs that a mere digital interface would not provide. Pure virtual libraries also discriminate against users who do not have network access -- a citizen should not be forced to learn new technology in order to access information. A virtual interface to a subset of library services is probably the most reasonable option.
Despite problems, Finland's online library services are still fairly sophisticated when compared with most other countries. When asked about possible future trends in the U.S., a group of experts predicted the following time lines [WDe95]:
As can be seen, the situation in Finland is not bad, we already have network access in some libraries. The prediction that virtual reality will become available in libraries before network access is surprising, to say the least. Of course, these are merely educated guesses -- the real world tends to provide surprises.
One of the classic roles of libraries and librarians has always been the classification of information. Ever since John Wilkins tried to create a universal classification scheme in 1668, it has been recognized that classification is a highly non-trivial area. Wilkins' scheme of classifying everything under 40 categories (including "things; called transcendental", "discourse" and "beasts") soon collapsed under the weight of new knowledge. Most modern classification schemes realize the subjectivity of classification (for example, do you store information about movies under "Art" or "Entertainment"?). An ideal system would store it under both, with links to the other categories. A well-known Internet search engine, Yahoo!, does just this. Yahoo! has a team of about 20 people who sort through the World Wide Web and classify interesting pages under the Yahoo! "ontology" (i.e. catalog scheme). Since Yahoo! is a digital entity, the cross-linking of information is easy. The role of the librarian in a virtual library would probably be quite similar, the creation and refinement of classification schemes and filing available material into that scheme. The traditional schemes used by libraries today are starting to show their age, and the term "classification crisis" is a common one in. Classification must be updated to handle modern forms of information, and methods of searching through that material must be provided. This is quite clearly a job suited for the libraries of the future [WMa96][Kuro95].
One could argue that a virtual library is little different from a WWW link collection. In the worst case this might be true, but a library typically has much more available resources for the updating of the information and making sure the links are valid. There is usually no way to figure out the authenticity or dependability of information found on the WWW, whereas documents stored in a virtual library have at least some credibility.
While sounding good in theory, moving to new forms of media and new technology makes huge demands on librarians and library staff. In addition to prior knowledge, they must now at least be familiar with -- if not master -- various sorts of new technology. This requires training, and training requires money. Until the people in charge of library development policy realize the advantages of the new technology, this money will probably not materialize.
While new forms of media evolve, the rules about their distribution and use must likewise evolve. The global shift into a networked paradigm forces many publishers and providers of information -- whatever the form -- to reconsider their way of operation and find new channels of generating income. Doing business on the Internet is still in the early growth stage, nobody knows the directions it will evolve in. Some risks must be taken in order to win.
Technology isn't the only driving factor in the change. The way people communicate with each other and view the world around them is changing, and the very concept of "information" has begun to be equated with "value". While digital media offers great promises, completely disregarding traditional media in unwise. In many situations, nothing competes with a printed book.
The nature of published works is slowly changing from a static model to a dynamic, continually updated model. In some cases, user/reader input can be incorporated into the product. This offers great freedom and versatility, but also holds dangers with regards to the reliability of the information and with ownership issues.
Libraries are also caught up in the change. We are already beginning to see public Internet access points in libraries, in the future libraries will probably offer some part of their material directly over the Net. What separates a library and a typical WWW "link list" in this case? The librarians, who are professionals in information management, retrieval and cataloging. A library of the future should be able to offer reliable search methods into authorative, proof-read data. It is quite possible that some libraries will go completely digital and turn into "virtual libraries", offering only digital documents. Unfortunately, all this requires funds, a resource that public libraries are typically short of.
All in all, the world of online publishing and hypermedia documents is still very much a frontier. There are riches to be found, but only if you brave the dangers of the "wild". Strict rules are few and far between, informal agreements between groups of people is the norm in today's Internet. How the commercial sector will change this picture remains to be seen.
| CERN | The European Laboratory for Particle Physics (Conseil Europeen pour la Recherche Nucleaire), the birth place of the WWW. |
|---|---|
| HTML | HyperText Markup Language, the language for creating hypertext documents for the WWW. In addition to text, documents can contain images, video, sound, and many other forms of data. Hypertext links allow global reference to other documents. |
| HTTP | HyperText Transfer Protocol, the communication protocol used by the WWW. |
| palimpsest | Writing material (as a parchment or tablet) used two or three times after earlier writing has been erased. |
| SGML | Standard Generalized Markup Language. A general-purpose document formatting and definition language upon which HTML (among others) is based. |
| WWW | The World Wide Web, a global hypertext network on the Internet. |
| WYSIWYG | "What You See Is What You Get", an acronym denoting desktop publishing programs that try to show the final printed form of the document on-screen, making visual layout very easy. The opposite of this are text layout languages, such as TeX, which must be run through a program in batch fashion in order to view the final layout. |