[petsc-dev] Known outdated sections of the User's Manual?

[Patrick Sanan]

That makes sense - is " the modern way" to get the data into an
environment with robust visualization, as fast as possible? For small
things, personally I am a fan of dumping PETSc binaries and opening
them in MATLAB (doesn't require any special configuration on the
library anymore, as far as I can tell), and I think that also works
with python. That also works well dumping to VTK. This is a topic
which might be very welcome for beginners in the tutorials, as there
is often an expectation of being able to scope things.

On Fri, Jun 24, 2016 at 3:45 PM, Matthew Knepley <knepley at gmail.com> wrote:
> On Fri, Jun 24, 2016 at 1:58 AM, Patrick Sanan <patrick.sanan at gmail.com>
> wrote:
>>
>> I'm working on some groundwork to improve the PETSc documentation, and
>> the next thing I'd like to look at is the User's Manual, adding
>> slightly-prettier code listings (as in the update to the dev manual).
>>
>> Before doing that, however,  it would be very helpful to know if there
>> are any sections of the manual which are known, by the relevant
>> experts here, to require deletion or heavy rewrites; it would of
>> course be a waste of time to format these.
>>
>> Specifically, I'm wondering about the following sections, which have
>> in common that they are things concerned with friendly external tools.
>> I've seen new users get very frustrated when they expect these sorts
>> of things to "just work," so it's probably constructive to remove any
>> outdated information here:
>>
>> - Chapter 11: Using MATLAB with PETSc . The support here has changed
>> quite a lot, so I'm not sure what currently works. Is the MATLAB
>> Compute Engine still supported?
>>
>> - Sections 15.10-15.14: Eclipse/Qt Creator/Developers Studio/XCode
>> users. This is likely not all current. Is this information helpful
>> here?
>>
>> - Section 15.15 : Graphics. I saw that there were some updates to the
>> drawing tools recently by Lisandro, so if any of this material is
>> known to be out of date, that would be helpful to know.
>
>
> I think graphics is not so much out of date now as incomplete. We really
> want
> to be telling people to do things the modern way, but the old ways still
> work.
>
>    Matt
>
> --
> What most experimenters take for granted before they begin their experiments
> is infinitely more interesting than any results to which their experiments
> lead.
> -- Norbert Wiener