Release Conference Call - May 4, 2009
Kepler Release Conference Call
May 4, 2009
Participants
Jones, Berkley, Welker, Brooks, Barseghian, Crawl, Wang, Riddle, Leinfelder
Agenda and Notes
1) Status updates and Estimate of task completion timeline
-- Berkley
-- mostly working on installer; working on 3 paltforms; DLLs in proper dirs
-- working on getting R install working again (end of day...)
-- need volunteers to test the installer
-- on track for time except possibly for documentation
-- Welker
-- working on refactoring the build; improving the workflow tests
-- stopped using junit ant task, using our own unit test task now
-- will start on prototyping of mudule renaming for publication soon
-- looking good on timing for release
-- Brooks
-- merge exp files; merged, but weird highlighting problems
-- ant report overrides is failing with class cast exception (David will fix -- fixed)
-- Barseghian
-- his bugs are coming along. June deadline ok
-- fixed the JLabel truncation issues, and search button
-- come across a few more problems, with new bugs -- associated with to actor library tree -- all are tractable
-- bigger taks is to create WF run object to encapsulate provenance
-- tracking bug to be sure kepler 2.0 has what's needed for modules to build in
-- Wang
-- globus actors into new module, trying for 2.0, bug 4040
-- Crawl
-- DBConnection removed, so now compiles and runs with Java 1.6!
-- add 1.6 target to NMI build and have it test there
-- Leinfelder
-- event module: took event model from Morpho, using as base for tab views to communicate with each other
-- evaluated Ptolemy event model, but it was limited to workflows
-- need to extend some TabPane and ViewPane interfaces for clearer comm mechanism, more formal
-- will produce a design diagram
-- R module is its own module now; can be released on independent schedule
-- will ship R, just not sure if it will use JRI yet
-- try to include R dist in R actor module or its own module, rather than the core installer
-- this allows us to sync version dependencies between RExpression, JRI, and R
3) Documentation
-- probably a large job; how can we accomplish it?
-- request to LT to find way to pay Kirsten Menger-Anderson for updates to the docs
-- chad will generate list
4) will announce next release to the kepler-dev list, and include the Build and Release team