Last week, I discussed Microsoft's confusing server message and the need for the company to consolidate its many server products. Well, ask and you shall receive: One day later, Microsoft responded with an announcement at MEC 2002 that it was consolidating its BizTalk Server, Commerce Server, and Content Management Server software into one product, code-named Jupiter, that will ship in two stages over the next 18 months. And the company is also consolidating its management products—Microsoft Operations Manager (MOM) and Systems Management Server (SMS). Perhaps these moves are an indication of simpler times ahead. This week, I discuss Microsoft's e-business server consolidation and some other MEC announcements.
Microsoft Senior Vice President of .NET Enterprise Servers Paul Flessner announced the company's Jupiter plans during his MEC 2002 keynote address last week. "Oftentimes when customers have a choice of an integrated product and standalone products, they feel like they have to make a choice between a pure play best-of-breed application, or the OK functionality you might find in an integrated product set," he said, explaining Microsoft's decision to integrate its e-business servers. "With Jupiter, our customers won't have to make that choice. We have the advantage of building this integrated solution on top of award winning, best of breed e-business products, so we're going to be able to deliver an integrated experience that's also best of breed functionality."
Microsoft will deliver Jupiter in two stages. In the second half of 2003, Microsoft will ship a Jupiter server product that includes process automation, workflow and integration technologies, Business Process Execution Language support, and an integrated development platform based on Visual Studio .NET. A second release, due in the first half of 2004, will add many Web site management features and an integrated environment for knowledge workers, probably based on Microsoft Office.
To extend Instant Messaging (IM) to the enterprise, Microsoft will release its Real-Time Communications (RTC) server, code-named Greenwich, in early 2003. The company originally planned to release this server as part of Windows .NET Server (Win.NET Server) 2003. And although Greenwich will still have strong ties to Microsoft's next server OS, divergent development schedules and increasing capabilities caused a split. According to Microsoft, Greenwich will provide a central means to manage all real-time communications within a business, supporting IM, video conferencing, and voice communication. And Greenwich provides the security, manageability, standards-based architecture, and extensibility that Microsoft says will set the product apart from competing solutions.
The next Microsoft Exchange Server version, code-named Titanium, is due in mid-2003 and will deliver a vastly improved Outlook Web Access (OWA) client that virtually duplicates all the functionality in the Outlook 11 client, also due in mid-2003. New OWA features include spell checking, task management, and antispam technologies. Titanium will also extend the use of mobile devices such as Pocket PCs, thanks to support for industry-standard mobile Web technologies and new mobile services. At MEC, Microsoft touted new Titanium features, such as support for Win.NET Server Volume Shadow Copy, Active Directory (AD), and Outlook 11 cache mode, so that offline users can still access their inbox information.
Microsoft also used MEC to introduce a new member of the Microsoft Office family, XDocs. XDocs is the code-name for a new application or service that will work inside other Office applications—Microsoft isn't sure which ones yet—that will bring XML Web services technologies to the client. Building off the investment that companies have made with XML on back-end systems, XDocs provides a client-side interface for accessing information. Previously, companies might have used Web pages for this purpose, but Web pages have two huge limitations. First, from a UI perspective, Web pages aren't very rich environments. As a result, they can't support many Office features that users have come to expect, such as spelling and grammar checking, WYSIWYG drag-and-drop functionality, and conditional formatting. Second, Web pages aren't persistent, and are ineffectual for entering huge amounts of data over time.
To address these and other limitations, XDocs will present the user with a familiar Office application interface, complete with all the niceties found in other Office applications. XDocs will support three modes: In Design mode, developers and technical users can create forms-based interfaces that interact with XML back-end systems. In Editing mode, an end user can access an XDocs form, and edit existing back-end data. In View mode, an end user can query back-end data, perhaps presenting it in various visually attractive ways.
XDocs functionality consolidates heterogeneous data sources into one interface, which reduces training and support costs. From a developmental standpoint, creators of XDocs forms can build in client-side data validation, exception handling, and other advanced features, further minimizing costs and data-entry errors. Scott Fisher, a program manager with the Office team, told me this week that XDocs is a direct result of customer feedback. "Customers need this technology," he said. "Consider the Education segment: Higher education wants to simplify the application process. You can't use a Web form to fill out a medical student application, because it's a 70-hour process. You can't do it in a browser, it's not possible. But XDocs works offline. You can fill out the form over time, and submit it when it's done. We've gotten a very positive reaction from users."
I asked Fisher about the viability of XML data, and he said that most back-end servers are already XML-enabled, whether they come from Microsoft or not. "Any industry with stringent requirements for data will want XDocs," he explained. "The insurance industry will use it for doctor referrals."
XDocs will probably ship with Office 11 in mid-2003, and Microsoft expects to ship a beta by the end of the year—likely in November.