On Thu, 6 Dec 2012, Kirk, Benjamin (JSC-EG311) wrote:
> On Dec 6, 2012, at 3:44 PM, Derek Gaston <friedmud@...> wrote:
>> We're a bit closer to "code users" (ie, non-code developers) than
>> you guys are though... so that policy might make more sense for us.
> Really what we should probably be doing is make 'devel' mode the
> default - kinda like PETSc, you can turn off additional error
> checking but it is opt-in instead of the default…
I actually like the METHODS="dbg devel opt" default you've got set up
now. Makes it clear that multiple versions are being built, whereas I
get the impression that half of the "my program crashed" reports on
libmesh-users are from people who didn't dig deeply enough into docs
or web pages to know METHOD=dbg exists.
If we just switched to devel as the default we'd merely replace "my
program crashed" with "my program was way too slow", which may even be
worse in that it's less likely to elicit an email allowing us to help
fix the problem.