understanding was incomplete.
Thanks to an excellent presentation, and demonstration, delivered by a
colleague, I really do get it.
In my language, using WebSphere Portlet Factory to write once, deploy
twice makes so much sense - WPF gives me the user interface ( a JSR 168
portlet ), Expeditor gives me the mechanism to create an offline,
replicated database model.
I'll have to try it but, as far as I can establish, I can create a
portlet using WPF that pulls data from a relational database e.g. DB2
UDB via a JDBC datasource.
Rather than "hard-coding" the JDBC datasource name in code, I have my
WPF model refer to an externalised resource e.g. a property file.
When I deploy the portlet to Lotus Expeditor, I update the property file
to point at a local DB2 Everyplace or Apache Derby database ( running
within the client ) and the portlet continues to function. I then use
the Lotus Expeditor server ( -OR- DB2 Everyplace on a server ) to
sychronise the data from the back-end DB2 UDB server to the client's
datastore.
Sounds good to me - will give it a try and report back.
Stuff rocks :-)