On Mar 12, 2012, at 2:20 PM, Sean Farley wrote: > But we don't have this model. We have petscvec.h petscsys.h etc etc. then the > model breaks down. Hence I like the petsc-private. Jed gets what he wants and > I get what I want. Plus it doesn't break the current PETSc application code. > > Then what is petsc.h? You just need to include the right directory with -I in > the petsc makefiles then everybody wins.
The problem is we dump all the petscvec.h etc crap also at /usr/include PLUS there would be stuff in /usr/include/petsc/*.h this is confusing. Why is some inside and why some outside? This is why I said Jed's model of standard usage didn't work for us.
