XML for Readers instead of Authors
When you look at the types of applications that people anticipate for XML, you see that the common assumption is that XML will be used by webpage authors in order to seperate content from presentation and allow machine processing of documents.
While these applications will certainly be the dominant type, and are good enough reason to develop XML, there is another type of application which could be possible - one which allows the use of XML by the readers instead of the authors.
If the reader of a webpage could add XML markup to the page, then the page could be processed independent of the original author. The reasons for doing this include:
- Pointing to document parts. The reader may wish to extract specific data from the page. This selection may not have been foreseen by the author. Indeed, different readers may wish to extract mutually contradictory selections of a document, so it would be impossible for the author alone to accomodate them.
- Indexing web documents. A reader could describe numerous similar documents using his own descriptive language - he wouldn't have to wait until a community had agreed on a standard markup language for that type of document.
- Upgrading HTML pages. The number of documents written in HTML continues to grow at a very fast rate, so if and when XML documents become the norm there will be a vast legacy of HTML documents. Some method for converting these documents to XML (that is, converting the data in presentational markup to appropriate semantic markup) will be needed.
Basically, the idea is that web documents - because they are in digital form - could allow a reader to more easily act upon the information they contain. By creating tools that go beyond the traditional aids (margin notes, highlighting pens, copy machines) we could see a dramatic extension of the usability of all documents, of the meaning of any document.
XML Marklets enable the reader to add XML markup to arbitrary webpages. Admittedly, they don't work very well just yet... the hope is that they will inspire others to think about placing XML in the hands of the user.
Steve Kangas
May 24, 1999