[Petsc-trilinos-discussion] Status update, summary remarks

[Heroux, Michael A]

There have been some good discussions on the list so far.

Ross, thanks for bringing up the challenges with CASL.  Others, thanks for
the detailed exchange.  There is a lot of good information to mine from
the conversation.

Lois McInnes and I are talking with Thomas Ndousse-Fetter today to see if
we can get a better sense of what he wants from these discussions.  He has
some app teams (unnamed) who are apparently driving the request for easier
access to DOE library capabilities.  Lois and I are arguing that we need
direct access to these people in order to make sure we are focusing on the
right issues.

The exchanges we have had on the mailing list are good, especially since
they provide tangible direction for CASL, but it's not clear we are
addressing what Thomas has in mind.

One last comment:  Early in the conversation there was a comment that
PETSc tends to keep interfaces and modify the implementation underneath,
while Trilinos comes up with new interfaces when a new approach is used.
While it is true that Trilinos develops new packages (e.g., Tpetra vs.
Epetra) when the design change is disruptive (in particular, when we
switched to exposing templates at the user interface level), it is also
true that we continue to evolve the implementation of existing packages.
Epetra has been around about 12 years.  In that time the *.cpp files have
been completely rewritten. The header files have been very stable in order
to enhance backward compatibility, only adding default arguments to
existing methods, and adding new methods as needed.  I just wanted to
clarify that point.

Mike