The last session of the first day was given to presentations from SoftQuad and Microsoft.
Bruce Sharpe, Vice President of Product Development at SoftQuad spoke on XML: enabling knowledge publishing on the Web. He identified the problem currently faced by Web users today as getting at the right information and doing something useful with it once found. Sharpe opened by discussing the rise of the knowledge-based economy, citing the exponential growth in information and the fact that the number of knowledge workers in the service sector is also rapidly growing (some office employees spend 60% of their time with documents). Web technologies can provide the 'plumbing' for knowledge sharing.
Sharpe talked about the astounding growth of intranets (80% of current web development is intranet related; 90% of all web servers and software by 2000 will be for intranets; 42% of fortune 500 are planning intranets). Sharpe defined knowledge publishing: draws on power of corporate intranets to cost-effectively create and deliver to customers and knowledge workers, an organization's knowledge bases which reside in documents. He sees an evolution that started with email, progressed into HTML, and is now pushing toward XML. Sharpe echoed Sterken's premise that HTML is not enough, but also noted that SGML is too much for many applications. His view: XML: all the taste, none of the calories. It provides a richer target for document conversion and a richer medium for interchange of database information. In other words, it is just right for information-rich documents.
XML enables precisely targeted content: the delivery mechanism can know who you are, and what you want to know. In this way, Sharpe identified a theme that was to recur throughout the conference, that of personalization in publishing. He went on to provide a number of ways that personalization could be implemented using XML on the Web. He stated that personalization not only includes bringing you content that is targeted to your needs/likes or that which has changed since the last time you accessed the information. It also specifically filters out that which you don't want to see, or perhaps that which you have already seen and acknowledged.
Sharpe concluded with predictions:
His advice: plan for XML-based documents and stay tuned.
Jean Paoli, Data Product Team Manager, Internet Explorer, Microsoft spoke on Client-centered computing for the Web. Paoli started by striking a cord in the audience's collective hearts. His statement: The Web is slow, Slow, SLOW. One way to address the slowness of the Web is to enable more client-side manipulation. He spoke briefly about Internet Explorer 4's Dynamic HTML based on an object model and data awareness in the browser. The object model makes it possible to effect:
Data awareness in IE4 means bringing more information to the client, 'pouring' it into HTML documents. However, Paoli pointed out that even though text is an important data format on the WWW, today, it does not contain enough information to enable rich client-side manipulation. Paoli went on to talk about what data and markup is needed to support client-based applications on the Web. HTML is too flat, does not contain semantical constraints, and is not the foundation for a solid architecture on the Web. Conversely, SGML can express semantic information, but is not implementable for the masses on the Web as it is.
Paoli, a member of the XML ERB, sees XML as the solution. Why? Because with XML we can create, validate, present, manipulate, and generate data, much of it at the client side.
NB: As an example of the appropriateness of XML to the Web, Paoli stated that an XML parser can be created in 2 weeks (rather than the estimated 10 man-years for an SGML parser).
Paoli's vision: a world of interactive, structured fragments used by Web technologies. His conclusions: we need to increase the quantity of semantically marked data; we need to use an open standard; we need to have this standard easy enough to be implementable by and for the masses; in fact, we need XML.
[Next]
Contact Robin Cover with corrections and updates, or to submit contributions to the ISUG online document database.
