0. Introduction

In section one and two I present a general analysis of publishing and, respectively, reading habits. Section three focuses on the data format of a Supplet while sections four and five give more specific details of the Supplets Browser plugin. Finally, section six draws some conclusions.

1. on generic document representations (it is already in progress, lots of published materials in HTML)

Recently you could hear people everywhere praising XML and HTML, which can be said to be a subclass of the former, to the skies. It might even be appropriate to speak of a hype.

Documents in such formats have a flexible format and are based on the simplest representation you can imagine, i.e. plain text. Thus they allow everyone to view the contained data with a simple text editor, e.g. to check what information exactly is included in such a file. Furthermore those formats allow for reuse of the contained data in varying applications which especially facilitates transport across platforms.

So one might imagine that at least those people writing about XML would actually use those formats for their documents. Unfortunately present habits are very different; searching the web, a lot of papers on the topic XML can be found - which are availabe in binary data formats (i.e. Adobe Acrobat or Postscript). These formats require additional software packages to be installed which then force the user to get used to one more user interface, normally leading the user to use those applications just for printing the downloaded documents.

It seems that the various available editors are not conceived as text document editors, probably the overwhelming possibilities of such editors are too much for a standard user.

While it will probably just take some time for those bad habits to change and eventually HTML editors will come up which are more specifically designed to write papers, the printing of documents which are already in HTML is a topic that can be addressed now.

2. reading an article (or paper as an artifact - also nicht bloss als informationsträger) LESEN ALS AKTIVER PROZESS

How do people read an article (e.g. form a newspaper) of high interest when it comes in the traditional, paper based form?

Most readers, especially when they are frequent readers used to efficiently filter the huge amounts of information they are confronted with every day, will mark the most significant parts of the text and maybe adding some remarks to the side of the text (fortunately there is in general some space left for such remarks). A piece of paper 'read' like this contains no longer only the information printed on it; it has additional, individual, information on it - what parts are of higher importance to you and/or remarks, e.g. clarifications. This additional information is the reason for the reader to archive this piece of paper instead the unannotated original text.

This becomes even more evident under reviewing or reporting conditions. Here the reader wants to collect his thoughts and ideas while reading the text and lateron share it with other people (e.g. the author of the original text or other parties interested in the readers opinion on the text)

It is long accepted that reading is an active process where the reader filters and interprets the information input she gets. But as outlined in the above examples various kinds of reading exist that are of an interactive nature; here the reader not only gets input but produces some output, too. This output reflects her perception of the information for individual later use ore to share such thoughts with others.

Since for these behaviours currently is not accounted for people tend to use their computer as an extended printer when they seriously want to 'read' a document. When the printed, annotated, paper is then to be shared with others faxes or copies of the paper are send. An even worse alternative is to send emails: 'The third paragraph could be more explicit'.

3. meta information

The information spoken about in the last section is so called higher order or meta information, meaning information about information.