Daniel Arndt

We can decide if we want to try to patch the bundled `Kokkos` or rather keep the `Boost` patch.

> If one assumes it is more frequent to configure deal.II with an external Boost than with an external Kokkos, patching the bundled Kokkos would probably cause less trouble. See...

> n FindDEAL_II_BOOST.cmake, when the if() at line 69 is entered, variable Boost_LIB_PREFIX should be restored to the state it was before the first find_package(Boost) call (say, at line 55):...

> Are we on 13.2 even for the newest Ubuntu? 🥴 Looks like it: https://answers.launchpad.net/ubuntu/noble/amd64/trilinos-dev.

> So perhaps we postpone requiring a later version until after the release? Yes, if we want to branch off by the end of the month, I would wait until...

Related to https://github.com/trilinos/Trilinos/issues/11921.

@nliber This needs to be addressed in the `parallel_scan` implementation, not in the `Sum` implementation (since that isn't used in the backend implementations). The latter is a separate discussion.

> Should `init(value_type&)` implementations be `static`? I didn't see any dependent on member variables. If we call it as a static member function internally that might break downstream code using...

> Given that, the logic should be as described above and not use reduction_identity. I propose to implement something matching the documentation like: We just recently clarified the current status...

> I think this goes into the right direction. Did you check what this does to the Tpetra test? No, I haven't. I would only expect an impact when using...