With the release of Office 2003 Microsoft have confirmed their total commitment to XML. The impact of this is that we can at last start to escape from "word processing" and progress into document systems, with all the advantages that come with structure. This will be a hard task and will take a long time. As a result few normal users will use the Office 2003features until those organisations that can gain immediate advantage have generated some experience and knowledge and have tackled the migration nightmare.
Needless to say Office 2003 employs a thick client architecture, the only way Microsoft can hope to continue to increase their revenue. However the servers are also upgraded and along with SOAP and Web Services the new system is clearly aimed beyond document systems at business applications. With products such as Biz-talk Microsoft has already established a toe-hold in business-to-business e-commerce. These introductory servers are gradually being rationalised, together with office oriented servers such as SharePoint. The objective is to provide an environment which will allow office applications to be integrated with business applications, with a common workflow. It has proved difficult to build robust integrated applications with the unstructured Office tools and the more formal data processing applications. There has been some success with exploiting Browser front ends with Application Servers to achieve application integration with legacy systems. This is far better done with Java based systems from BEA, IBM, et al than with Microsoft's Windows-only products. But if Microsoft can provide the fabric to integrate business and office applications then they will be in a strong position, courtesy of their stranglehold on the office side.
This then is where XML comes in, since structured documents and business data can both be defined in XML packets. XML units are defined by a schema, so that while XML is an open standard, the schemas are proprietary, many owned by Microsoft!
There are new versions of Word and Excel, which add XML capabilities to the older formats. A Word document for instance can be saved as XML, using the WordML schema; nobody expects this to be simple or reliable, but it is a necessary step forward. It is unlikely to be a good idea for building clients which can extract, manipulate, displayand post data into the XML packets in the workflow systems. Such client applications are an integral part of the business applications and need to be maintained and controlled as any other business application client. This is far more complex than the current portals which can map data extracts from multiple data sources into a common client; it is the difference between a transaction system and a query system.
The basic requirement is for a "forms" tool and this is where Microsoft has produced the most interesting add-on to Office 2003, InfoPath (formerly called XDocs). Infopath will live for a long time alongside Word 11 and Excel 11, although it would be nice to think that it will mature into a true XML document editor to replace Word eventually. For the present keep the involvement of Exchange, Word and Excel to a minimum and find time to investigate InfoPath. In the first phase it should be treated as a development tool, controlled by the IT department, as a means of delivering integrated applications. At a later date it will be worth considering how much it can offer as an end-user tool but it is always dubious to let users do their own thing with tools that can update data!< BR>
Martin Healey, pioneer development Intel-based computers en c/s-architecture. Director of a number of IT specialist companies and an Emeritus Professor of the University of Wales.