XML'99 Expo Update

December 8, 1999

Simon St. Laurent

Secure Database Connectivity Appliance Coming Soon

Killdara's XML-based network appliance

John Ogilvie, President and VP of Technology for Killdara, eyes the Cobalt server, which his company customizes to create a secure XML server.

Killdara is developing an XML-based network appliance for transmitting information between computers securely across networks. The Paraphrase Engine can be used to transmit information between databases where both parties are using a Paraphrase Engine, or in conjunction with other applications that support public-key security and XML processing.

Killdara, of Almonte, Ontario, have built their appliance on a Linux and Java foundation, building a device that act as front-ends to separate database servers. Digital signatures and encryption provide a security layer to XML documents traveling over open networks.

The Paraphrase Engine constructs XML documents on the fly to meet requests for information, and can also receive XML documents for addition into the database or further processing as email. DTDs are a core part of the Paraphrase Engine's strategy -- documents created by and used with the Paraphrase Engine must be valid XML documents.

Killdara's partners include SoftQuad, whose XMetal XML authoring software is integrated with the Paraphrase Engine, and Entrust, which provides the public-key infrastructure. The appliance itself is built on Cobalt server appliances.

The Paraphrase Engine will be entering beta testing with systems integrators in the new year, and should be available commercially (for US$9995) in the second quarter of 2000.

Percussion Software Releases HTML to XML+XSL Tool

Percussion Software demonstrated XSpLit, a tool for converting HTML documents into XML and XSL equivalents. XSpLit allows developers to use extra annotations in ordinary HTML documents to indicate XML equivalents and build XSL style sheets that would convert the XML back to HTML.

By adding semantic information to the raw HTML through attributes, developers can take advantage of HTML formatting tools to create templates, and then run those templates through XSpLit to create reusable XSL. XSpLit recognizes features like repeatable lists and tables, making it useful for situations where content is generated from databases and other list formats. Percussion's Rhythmyx software provides additional tools for the database connectivity.

XSpLit will also produce the XML half of the XML+XSL combination and a DTD for that XML, making it useful for converting the content of HTML documents to XML as well as HTML template creation.

XSpLit is free for personal use, and can be used with or without Percussion's Rhythmyx Server. It should be available in January 2000.

infoShark Demos XMLShark Alpha, Beta Next Month

infoShark is demonstrating an alpha of its XMLShark tool for data transfers to and from Oracle databases, and will be starting a beta program shortly.

XMLShark uses infoShark's Commerce Accelerated Relational Data (CARD) schema to create complete representations of relational data, preserving referential integrity information. CARD is publicly available at the BizTalk repository, and could be used on its own or to integrate XMLShark with other applications.

The "Commerce Accelerated" portion of the CARD acronym refers to components in the schema that allow vendors to indicate information costs, building infrastructure business can use to buy and sell database information as well as exchange it.

While the current version of XMLShark is focused on Oracle, Barbara Bouldin, President and CEO of infoShark, plans to bring XMLShark to other database platforms as well to "liberate data" and encourage interchange among database systems.

STEP X2X XML Linking Engine in Beta

STEP UK has begun a beta program of its X2X linking engine and made a technology preview available. X2X provides support for external links, building connections between documents without modifying those documents.

X2X is based on the W3C's developing XLink technology for extended links, though it uses a database store for more scalable storage and processing. X2X accepts XLink information for both input and output.

X2X stores its linking information in relational databases, using JDBC to establish connection. The X2X Server provides an application interface to connect with other applications that need to specify links -- X2X itself doesn't provide tools for creating documents or links.

To generate output, an X2X Datafetcher component retrieves information from document repositories, like Web and file servers. The X2X Linkmerger component then combines the linking information with the document information to send recipients a document complete with linking information.