Wednesday, September 24, 2008

Some initial Oracle Beehive impressions and projections

This post outlines some of my initial impressions of and projections for Oracle Beehive, based on an Oracle analyst briefing last week and some Beehive sessions at Oracle OpenWorld this week. For some additional context-setting, check the following posts:

Oracle Collaboration Suite: RIP

Some initial Oracle Beehive details

Somewhat paradoxically, Oracle has not yet made the OpenWorld presentation files available for download (and I didn't bring a camera to capture OpenWorld session demo images), so I can't include screen shots in this post -- that's unfortunate, in part because many of the screen shots reporters may run with in their Beehive-related articles this week are likely to be misleading, focusing on Beehive/Outlook integration rather than the broader and deeper Beehive capabilities available through the full/purpose-built and web-centric Beehive client.

A list of Beehive features, from a Beehive introduction session yesterday (presented by VP and Chief Architect, Oracle Collaborative Technologies, Terry Olkin):

  • Contacts
  • Messages (email)
  • Wiki
  • Tasks
  • Conferencing
  • IM (transcripts)
  • Calendar
  • Documents
  • Discussions

The Beehive model is workspace-oriented, and the high-level Beehive object model includes:

  • Entity
    • Artifact
      • Document
      • Calendar entry
      • Task
      • Conference
      • Message
        • Email
        • Discussion
        • Voicemail
        • Instant message (transcript)
    • Actor
      • User
      • Person
      • Group

Terry Olkin also briefly displayed a super-detailed object model diagram (one I'm sure easily gets into the braincramp complexity range, when the various adapters and interfaces are included) -- the Beehive model appears to be very deep and intertwingled. Oracle has not published the underlying/corresponding Beehive database schema, but the developer-oriented object model is documented in the Beehive Application Developer's Guide.

Several of the Beehive scenarios described at OpenWorld over the last couple days (in presentation sessions and in discussions at the Oracle Beehive stations in the cavernous OpenWorld exhibit halls) are similar to IBM Lotus Connections in some respects, e.g., attempting to exploit relationships among information items and people, not just basic capabilities such as document libraries and discussions. I definitely got the impression, over the last couple days, that the initial Beehive user experience is just the tip of the iceberg, in terms of fully leveraging the underlying conceptual model.

I'll post more details on these topics if/when I get access to the related OpenWorld presentation files. For now, some initial projections:

1. Oracle Beehive is likely to be popular with many Oracle-focused enterprises, especially those using one or more Oracle applications, as Beehive services will be integrated into future releases of Oracle's myriad enterprise application offerings (and are already included in a new Oracle Social CRM product). I doubt Beehive will broadly displace Outlook/Exchange and Notes/Domino for enterprise messaging in the immediate future, however (although it could become a competitive issue for Oracle customer organizations currently using Novell Groupwise, for example).

2. Oracle Beehive's DBMS-based infrastructure is, imho, a leading indicator of likely future communication/collaboration products from IBM, Microsoft, and other vendors. SharePoint is already SQL Server-based, of course, but its object model is not captured in SQL Server in a high-fidelity model mapping today (SharePoint treats SQL Server as more of a property bag manager, and SharePoint doesn't currently use any of the XML management services in SQL Server). Beehive doesn't fully exploit the native XML management capabilities in Oracle Database 11g, however, since Beehive is for relatively lightweight document management (and Oracle's Universal Content Management, designed for more complex content/record/etc. management scenarios, and based on Oracle's Stellent acquisition, also doesn't fully exploit Oracle Database in this context).

3. Beehive's wiki-centric model -- pervasively supporting "the wiki way", rather than treating wikis as a distinct workspace capability, is another useful example of beyond-the-basics hypertext support.

Of course, it is not a foregone conclusion that Beehive will be broadly successful, as Oracle failed in at least three earlier attempts to compete in enterprise communication/collaboration domains, but success in these areas is in many ways a strategic imperative for Oracle at this point, and Beehive appears to have a very clean and integrated application model, so I'd be willing to bet $.07 it will be much more successful than OCS was (BTW I ran into Mike Gotta in the OpenWorld analyst room this morning and paid him $.07 for the OCS bet I lost, as explained in this post...).

More to follow in subsequent posts, as I learn more about Beehive...

Post a Comment