Application Migration vs. XPage enablement

In a recent customer discussion a claim was made: " If Notes client application don't automagically can be converted into XPages applications, then we very well can migrate to another platform, it is just the same". An interesting claim, I'll subject it to a reality check. In any case it is a good idea to revisit your investment in your existing applications first.

XPages Enablement

• A reasonable estimate for the required effort is up to 30% of the initial effort. It very much depends on the quality of the existing applications and familiarity and skills of your team
• There is no data migration, so existing users using a Notes client can continue to use the application until you retire the Notes clients
• You can look at your application from a functionality and user perspective. A typical first XPages application is an approval module that spans across all your existing applications and provides a single view into anything that needs approval. A next level is forms and tools for mobile personell
• Domino's security model works in XPages too, so no changes are required
• If a function in an XPage shows a bug, it doesn't stop executing the function, since a user can fall back to a Notes client until the bug is fixed
• Mail notifications just work, it's Domino. Also you can build Attention Central

So in a nutshell: low risk, incremental availability, no data migration, freedom to innovate

Application Migration

• Your investment in rewriting applications probably will be the same or more of the initial investment - that's why you did the CoCoMo analysis in the first place
• Data needs to be migrated, so at some point in time you have a hard cut-over. Shut down (or put to read-only) the existing app and start using the new application
• Before releasing the new application it needs to be feature complete including reports, admin UI, data management. Quite often the time doubles from core feature complete to all features - delays and costs increase
• Since applications need to be feature complete (!), you will need to pick an application by application approach. This makes simplification and synergy effects more difficult
• A bug can be a show stopper for the affected function since there is no fallback to the existing application
• You need to build your own notification mechanism and eMail interface
• When migrating data from Domino's NoSQL to an RDBMS, you might be in for complexity especially around access control. Complexity equals risk

Of course YMMV