Intuitive SGML: database integration in SGML authoring		Table of contents		Indexes		Hyperlink semantics for standoff markup of read-only documents

Stadler Thomas

Publishers wanted, authors needed! The new information age is waiting for your works

Abstract:

The new paradigm of information objects has recently emerged that replaces the old one of documents. The new view on information concentrates on smaller bits of information which may be connected in different contexts and that are linked and webbed together under multiple perspectives.

This paper focuses on the techniques and applications that are available already to produce and maintain information webs. We discuss the fact that many authors and publishers are writing books as they have been doing for the last 500 years. Partly it seems to us to be the publishers and authors turn now to redefine their methods, their products and their markets. What are the new opportunities, what abilities and skills are needed, and what are the problems in the shift to a new way of writing and publishing?

Introduction

One or two generations ago, in 1945, Vannevar Bush sketched his Utopia of the upcoming information age: “The Encyclopedia Britannica could be reduced to the volume of a matchbox. A library of a million volumes could be compressed into one end of a desk.”

NOTE:

Vannevar Bush, As we may think, July 1945, The Atlantic Monthly

Bush called the device “Memex” and Bush's grandchildren built it – more or less:

“As he has several projection positions, he can leave one item in position while he calls up another.” The answer were windows.

“He can add marginal notes and comments”. The answers were notes, annotations and bookmarks

“The process of tying two items together is the important thing. When the user is building a trail, he names it, inserts the name in his code book, and taps it out on his keyboard.” The answer was hypertext.

And some of the questions still need answers: “The prime action of use is selection, and here we are halting indeed” and so do WebCrawler and Altavista. Or this“new profession of trail blazers, those who find delight in the task of establishing useful trails through the enormous mass of the common record” , which leads us into the middle of the discussion: Where are thetrail blazers showing us the way through the jungle of information? If not “finding delight” at least finding an entrepreneur willing to enter new and uncertain market segments?

What were the changes since then in the last half of the century?

Digital information in many formats is wide spread. Computers are in every office, in living rooms and even in the rooms of our children. The “Memex” exists as World Wide Web, as CD-ROMs and as large storage devices. Technical documentation and some products of the publishing business are in the middle of the shift from paper to screen. There are new reception environments. The business world, the scientific world and the private world are ready to pay for the trail blazer's work. But the industry gives us on-screen paper.

The technique is ready

Concepts

There are all kinds of theories about information and information webs. The markup standard is given as SGML or quite as good by XML. HyTime can represent some interdependencies. There are tons of articles, proceedings and books concerning hypertext.

There are good proofs that not every problem in the transmission of the information and its relationships to the receiver can be solved – but many can be solved and many are.

Not every academic idea seems to be feasible; some approaches are preliminary, provisionary. But, e.g., having a look attopic maps , an application of HyTime, one gets a quite good idea where applied theory is heading.

Applications

The speed of the development ofgood SGML applications is too slow. In some areas there are too many monopolies on applications and too few competitors.

Nonetheless most of the common requirements are met by current applications. There are some good SGML-editors, good viewers and browsers. There are excellent conversion tools and transformers. There is fulltext retrieval and database support. The production is supported by workflow management systems. In more complex environments some SGML based document management systems are available which administer information units, their relationship, indices, multiple use, aspects and perspectives, DTD changes etc.

We would not claim that things are really adult there – but they are waiting for actors ready to play around, to test, to make experiences, to develop requirements, to help the vendors to make progress and further development.

Market

The market asks for printed paper, still. The market asks for good electronic products as well – and doesn't get them. There are three areas to be considered: a) the industries which need digital documentation from their suppliers for further use or for combination with internal and other information; b) the people “outside” which need fast access to always up-to-date information (independent of its format); c) the mass market of the publishers clients.

In theindustry theLeidensdruck is tremendous, but there are many legacy problems as well. There are huge problems in most companies to change the documentation environment. Not taking care of those internal problems the customers ask for digital documentation in acceptable revision cycles. The customer's requirements are not clear yet but there seems to be a request for branch-wide formats. Civil aircraft are documented in ATA documents; the news agencies have defined a DTD; the telecommunications, automotive and pharmaceutical industries are in the process of defining their own communication structures. Company DTDs are needed internally, but they are too narrow and sometimes secret, so there has to be an abstraction to a branch DTD and sometimes another one to HTML.
The people outside, the mechanics, the branch managers, the trainers need access to newest information in many cases translated to foreign languages. With “people outside” are meant as well the millions of users of an application who expect to get a good technical on-line documentation of their WordPerfect or Excel. These readersdo not want SGML , but they needprecise, extensive, navigable, up-to-date and nice documentation, which gives them fast and easily all the information they need in a given situation, and not more.
In the publishing world the market wants good applications which ease the access to the information and the navigation within it. It is a quite interesting and a typical observation, that the most praised encyclopedia on CD-ROM is not made by Encyclopedia Britannica Inc. and not by Grolier and not by Brockhaus or LaRousse but of all publishers byMicrosoft . Did all those great publishers of yesterday decide to hand over the business of information brokering to others? They have the knowledge, they have the infrastructure, they have the profession, they have the staff, they have a name and they have customers. They seem not to want to be “those who find delight in the task of establishing useful trails through the enormous mass of the common record”!

The authors aren't ready to work; the publishers aren't ready to invest

There are by far not enough information products which make use of the existing possibilities; not in the technical documentation, and not in the publisher's spheres.

In the middle of the 15th century the technical conditions were met, to democratize the distribution of the written word byprinting it in editions greater than one. Within some decades the publishing industry was born and became adult. The monopolies of the churches and the feudalism faded away. The roles of the publisher as an information brokerand as a paper maker were synonyms over some centuries.

I would claim that we are in a time where the unity of paper and information is broken by a new technology which is even more important than the invention of the printing press. In the development of our days the content and its arrangement is touched, and not only the technique of multiplication and distribution.

These are the times where new branches of industry seize the opportunity, when the old ones miss the chance being asleep or too lazy or too anxious. The first step for the new branch is drawn by some software help systems and by some CD-ROM products. Insufficient first steps are shown by hundreds of companies and thousands of products mirroring the paper world onto the electronic media and calling it multimedia by adding more or less stupid movie clips.

As far as we have reached now the assertion is, that there are applications and integrated products which support

definition of neutral and semantic structures, descriptive markup for improvement of quality, rendering, navigation and linking

creation and maintenance of small information units (lexia) which may be connected in different contexts or productions

definition of quite complex linking schemes: multi-directional links, weighted links, link roles

preparation of multiple media and integration of multimedia

administration and support of text generation, maintenance and production.

distribution, update methods, payment modes. All is clearly to see in the development within the Internet and the WWW.

Use these possibilities. Look at information as a network of lexia and build your works a bit more appropriate to the structure of thinking and of concepts. It is as simple as that. The way into the new information age can be done step by step only. These are the elements of step one.

Go for the first goals, publishers! Do not walk! Sprint!

The author's dilemma

This last section in this paper discusses the topic of the realNew Information Age's challenges, and the real problems.

In order to be able to plan how information should be presented in an electronic way, it is very essential to understand how people read, how people choose to go a certain way, how they decompose the sequence of letters and words into concepts and finally into memory.

Traditionally authors and their assistants (editors, copy editors) are the active parts, the information makers. The marketing department might have the idea, the publisher gives the order and the money and sometimes the environment. But the author is the one who translates the concepts into words and pictures and sound etc.

Books have been static by definition. What was written could not be changed dependent on the reader's actions or retrieval requirements. The only degree of freedom for the reader was, not at all to read the book, or to read it in parts only, or to follow a non-sequential path guided by references, indices, table of content or by chance. But the book would not change.

The completely new with electronic media is that the computer can react to and interact with the reader and her environment.

Tomorrow's information has to be positioned in a multidimensional space that has the following main categories: a) interconnections between information units, b) objective environment and c) subjective environment. The receiver's space is in b) and c), which have to interact with the reader or user and is dynamic by nature.

Under ahypertext aspect we have to find out how to model the associative way of thinking as a network of lexia, of information atoms; how to avoid a too coarse or a too fine granularity of units; how to give consistent guidance to the reader what she might expect in following a link and never make her lost. The linking system has to be weighted, qualified, consistent, helpful, and it might depend on receiver's parameters.
Situation and installation parameters have as well an influence on variants of information. If there is no sound card a link to an audio event makes no sense, if the ship has no radar the advice to the sailor lost in fog to have a look on the radar screen sounds strange to the poor captain; if the last service bulletin is not implemented at the aircraft, the fault isolation procedure cannot refer to it. If the customer did not buy the error diagnosis module, there might be a proposal to buy it and why this would make sense in the given situation, instead of starting it. Then thegeographical environment plays a role: The information depends on first of all language, then regions and countries (the Catalan is by far more interested in Barçelona than in Madrid or Sevilla; the U.S. mechanic needs the 110V table and not the 220V)
The information is presented dependent onproperties of the reader : skills, experience, education, history of the reading NOTE: The libraries of tomorrow will be at the hands and the desktops of the readers. So the reader's ways through the jungle, their methods to find something of interest earlier or later, their errors, their strategies, their annotations, their bookmarks, their ways of remembering are important topics for the applications of tomorrow – but out of the scope of the single author. , process (the tenth time that she enters a unit), interest (knowledge, example, maintenance, repair, definition, relations etc.), time of day (“it is 04:00 a.m.; have a drink, call at home and have a nap!”). Whatever one might think about the quality or usefulness of Microsoft's help assistant, this feature is an astonishing example for an interaction between user and application! After a while hidden watching, the documentation becomes active and asks “Sorry for disturbing, but it is the second time that you deleted a paragraph with the backspace key. Are you interested to hear about a better way?”

We know quite precisely how the information of tomorrow should present itself to the receiver:customized and reacting to my needs, my situation, my knowledge and my requirements and being in defined, understandable relations with the information universe , dynamic and not static, tailored, fast in the three dimensions given above.

If we agree that the one who builds up those many dimensional networks still is calledauthor – what great many skills do we expect from her and how can she be supported by an application and new standards? Or will the author's role be subdivided in a whole new branch of professions, e.g., information designer, architect, webber, imaginator, scenario builder, receiver investigator etc.

The main problem in my opinion is the problem of thesynchronization of concurring concepts . The information structure we talk about is highly overlapping. Dependent on the perspective, the aspects, external conditions, and dynamic situations content and relations will flow in order to adopt themselves to the reader's requirements and situation. The author's task is to plan and fill all these levels and to foresee what might happen at the reader's side.

There are technical problems. With overlapping hierarchies for example we will reach for sure the limits of SGML or XML

NOTE:

There is a good analysis of this topic in: Refining our Notion of What Text Really Is: The Problem of Overlapping Hierarchies: Allen Renear, Elli Mylonas, David Durand. Draft version, January 6, 1993

! The maintenance of such systems will be a challenge on its own. But, technical problems can be described and solved sooner or later.

What makes me shivering is that I have no idea what sort of brain and what methods of working or cooperation the authors need. For paper books it was a sometimes empty demand for the author to understand how people read. For the building of the multidimensional information space it becomes theconditio sine qua non to write together with all virtual readers, to have clear concepts about all the many axes in the information universe and to fill the many spaces with coherent content.

So this paper ends with an uncertainty. We are able to describe theWHAT and theWHY . We should begin to discuss theHOW and theWHO .


Intuitive SGML: database integration in SGML authoring		Table of contents		Indexes		Hyperlink semantics for standoff markup of read-only documents