WWW9 Devday: Publishing Tools Track

Organizer: Dale Dougherty, O'Reilly Network

9:00-10:30: Perspectives

The Once and Future Web Publishing Platform -- Dale Dougherty, O'Reilly

The Web publishing platform has changed significantly since its inception. However, it is still nowhere near where it needs to be. In this talk, I'll review important trends, standards and technologies that are shaping the Web as a publishing platform. I'll compare the Web publishing platform of GNN (circa 1993) with that of O'Reilly Network (2000).

Open Source Tools for XML-Based Web Publishing -- Edd Dumbill, XML.COM

Open source platforms are among the best-of-breed for creating dynamic publishing experiences. XML is becoming the best-of-breed way to create and exchange content. Bringing the two together, this session will explore XML-based web publishing with PHP, Zope, and tools from the Apache Project.

10:30-11:00: Break

11:00-12:30: Content Development Techniques

Content Negotiation Techniques -- Koen Holtman, CERN

Content negotiation is the process by which the 'best' version of a web page is automatically chosen and displayed when the user follows the link to the page. I will discuss the current state of web content negotiation standards, together with currently applicable negotiation techniques, inside and outside the standards. The focus is on techniques for site designers who want to support 'standard' browsers. Techniques for browser designers, by which they they can better support site designers who want to put up negotiated content, are also discussed. I will review the necessary measures which must be taken to make negotiation work correctly if there are web caches between the browser and the server. I will also review recent developments like the creation of a feature tag registry by the IETF, and the implementation of the transparent content negotiation standard in the Apache server.

Reusable HTML Components - Behaviors -- Dave Massy, Microsoft

This presentation covers the use of Behaviors as supported in Internet Explorer 5 and specified in the working draft specification at http://www.w3.org/TR/becss. As well as the W3C draft specification there is full documentation and samples of Behaviors on the Microsoft site at . These Behavioral components allow additional dynamic functionality to be added to any element in an HTML document meaning that script on an HTML page can be placed in easily reusable separate files that can then be applied to content by a document author who need have no knowledge of scripting. The talk will demonstrate the technology and explain how it makes managing and publishing content easier.

Database Publishing with XML -- Michael Rys, Microsoft

Everybody knows by now the usefullness of XML in publishing information on the web. Ever wondered how to get your data from your relational database system on the web efficiently and securely? This presententation will demonstrate how to utilize SQLServer's XML capabilities to build XML driven web sites. The presentation consists of an explanation of the architecture, the different XML related features and how they play together to build a simple XML and database based web application that will be demoed live during the talk. The discussed features will include: HTTP access to the database, security issues, generating canonical, heuristical and userdefined XML for SQL queries, the usage of schemas and XPath to query the database, and how to elegantly provide rowset abstractions over XML to shred XML into relations.

12:30-14:00: lunch with Tim Berners-Lee

14:00-15:30: Content Management Systems

Cross-media Publishing -- Michael Wechner

A couple of months ago I started a content management and publishing system based on XML and XSLT. The system is now developed as an Open-Source Project.[http://www.wyona.org] The WAP-service of the newspaper "Neue Zurcher Zeitung" is already powered by this system. [http://www.nzz.ch/wml/index.wml] [http://www.nzz.ch/wml/index.html] The interactive edition of the same newspaper will also be powered by this system in the near future. This presentation will offer a brief demo of the system and show how cross-media publishing is very simple with XML and XSLT.

InfoSite Content Manager -- Lars von Olleschik

InfoSite is a web-based content management system that we are developing as a ColdFusion application. All aspects of administration, content management and content contribution is done through the browser. InfoSite is able to assist in distributed publication for a large group of content creators. It assists in international site care as the interface can be switched to the language the current user prefers. Soon we will also be able to publish one web site in several different languages.

In contrast to many other systems the content management of InfoSite is not based on a two-dimensional tree chart. Instead the contributor himself adds knowledge about the information he is responsible for to the central database. The content description is made by the selection of content specific keywords within a common vocabulary shared by all contributors of a specific site. The content manager of a site is able to specify whether information is available that fits certain boolean criteria. In that way he creates subsets of information that will dynamically grow or shrink. This is extremely powerful as there is no longer a need to place links to a piece of content statically on each page that should point to it. Instead a document can be placed on one or many different places within a web site.

Weblogs in Manila: Tools For Writers -- Dave Winer, Userland

Manila is an Internet server application that allows groups of writers, designers and graphics people to manage full-featured, high performance web sites thru an easy-to-use browser interface. This talk will look at weblogs as a distinct form of publishing and how Manila is used to run sites such as weblogs.com and editthispage.com.

15:30-16:00: Break

16:00-17:30: XML Publishing Applications

XML at the Koninklijke Bibliotheek -- Simon Bosse, National Library of the Netherlands

The Koninklijke Bibliotheek (KB) is the National Library of the Netherlands and collects all published material in the Dutch language. Besides the main catalogue (OPAC) there are a lot of specialized bibliographic databases about manuscripts, pamphlets, watermarks, electronic journals and others. They consist of different DBMS systems from sybase and oracle on unix systems, to inmagic and access on PC's.

The KB wanted to provide its users with one interface to search through all these databases. One possibility would be to connect a search interface with Z39.50 to all databases, but some of the databases do not support Z39.50. The solution the KB has chosen was to export all databases records in XML and store them in a new database, index them with the Alta-Vista Search Development Kit and create a web interface to search. First only a subset of the most common fields (author, title, abstract, etc.) was used, but it was found this approach was not flexible enough. Therefore a complete DTD was developed which contained all the different fields from all databases, including the OPAC.

This talk will deal with the design of the total system, the data model, the choice for XML as common record format and for the communication between the servers. The choices for the different tools and programs. The development of the DTD and the use of XSL for the presentation of the records. One of the points made is the easy adaption of the interface to search in a specific subset of the database or to create a different user interface for specific records.

Title Open eBook - A Standard for Publishing -- Ben Trafford, Yomu

The Open eBook Publication Structure 1.0 is designed to allow print publishers to convert their texts once, and republish them with any provider of electronic books. This specification is enabling the publishing community to make the jump to distributing their material via the profitable new channel created by the World Wide Web.

OEBPS is an XML application, that has an HTML-like base, and also supports a minimal CSS subset. However, it can be extended to support virtually any tagset in a robust, interoperable fashion, and can also support extended styling as well. It also has an interesting package structure, that not only describes the objects present in the electronic book, but also reading orders and metadata. This talk will discuss OEBPS 1.0 in some depth, and will also offer an overview of the Open eBook Forum, a new body dedicated to developing and maintaining open standards for electronic books.

Using XML for news aggregation and delivery -- Dave Galbraith, Moreover

Moreover aggregates news from 2000 sources on the Web, taking unstructured data and delivering structured results in various flavors of XML. This presentation proposes to focus on the specific problems and experiences associated with using XML in the harvesting and delivery of news on the web. This will include: harvesting (dealing with sources that are constantly changing both in terms of layout and content); categorization and indexing (building a dynamic categorization system for news on-the-fly); delivery (how to deliver structured news content using XML technologies for use within other applications, ranging from websites to search engines). We'll talk about RSS and its importance as an open syndication mechanism. We'll also look at how XML-RPC and SOAP fit in.

Updated: April 21, 2000
webmaster@www9.org