Bloggers: Kevin Kampman and Gerry Gebel
At The Experts Conference last week, Microsoft disclosed the latest delay in the shipment of Identity Lifecycle Manager “2”. Originally targeted for the second half of 2008, customers will now have to wait until at least the first quarter of 2010 – a year from now.
Microsoft says it needs the extra time to address feedback from early adopters (including Microsoft’s internal deployment) that focus on three key areas:
- Usability improvements for handling Management Policy Rules
- Database performance and scalability improvements
- Implement self-service password reset for Windows XP Service Pack 2 clients (currently only supported to SP 3)
Another Release Candidate is scheduled for this fall, which will include the above updates.
This is a tough balancing act for Microsoft. We applaud their willingness to delay shipment of ILM “2” to address customer concerns, but the wait has been extensive to say the least. Additional delays impact customers and prospective customers in many ways, such as issues with budget allocation, availability of implementation resources, limitations of existing implementations, and so on. Microsoft is also missing a market opportunity as user provisioning remains one of the most active IdM market segments where organizations are continuing to invest, despite current economic conditions.
As Microsoft deploys capabilities that extend to every user, their responsibility to do the right thing, that is, test and verify, grows exponentially. They can't afford to be wrong, that is, test as they go, when it could impact large segments of their user population in a negative way.
The group/role/policy capability enhancements are probably more problematic than they anticipated. There are implications in this area that a number of Microsoft identity management vendors have struggled with; it is unrealistic that Microsoft should expect this one to be easy or straightforward. We might even expect these to be downplayed in the first availability. As we've learned in our current research on roles, retrofitting something to the AD groups that already exist in the field will not be smooth or easy
Comments