To edit this file or the web pages, please do:
svn checkout https://github.com/pencil-code/website/trunk/ www


Topics for Pencil code User Meeting in Graz 2016:
=================================================
% Notes: "-" means that this item is still open (nothing has happend)
%        "+" means that it is already in progress (people working on it)
%        "*" same, but high priority
%        "/" postponed
%        "x" means that it has been dealt with
%            [assigned Persons] feel free to add yourself
%
# http://pencil-code.nordita.org/UserMeetings/2016/

+ state of the conductivity module?
  Wlad, Boris, Anders, Sven => a session is needed to actually do the moving of code.

- state of the particle modules (documentation, avoiding overlap, etc)
  Dhruba could not report

- re-organization of the particle modules with "swarms" being a sub-module?
  Dhruba could not report

+ general re-initialization of particles
  inject particles during the run.
  Nils, Axel, Alex, Jorgen

+ PC paper?
  Fred: compilation on the webpage on existing papers, RAS
  Philippe: find out about Fronties in Science: instrumentation & methods
  use just a *compilation* of individual contributions
  multigrid (there are now 2 versions)
  link README file into manual.
  review hightlights or write methods paper
  Frontiers in Science

+ hightlights to update on the webpage
  People are informed

- filter out sound waves during the run with an adaptation
  of the anelastic module

+ reduce size of compiles src-directories
  Matthias: dynamic linking of Fortran libraries?
  Philippe: keep compile files on DATA partition
  Jörn: linking on request

+ IDL auto tests, status report & workplan
  what is actually tested and when/where is this test actually run?
  ask Wolfgang for status and next steps

- 3-D visualization using radiation transport

+ Yin-yang mesh, status report (Matthias)

+ F2003, status report, Matthias & Philippe
  A test code is available in "src/experimental/test_f2003.f90".
  Matthias has to add one more check, first.
  After that, everyone should try out this test.

- post-processing and many other things always compiled.

+ Boundary conditions, Matthias
  A theoretical discussion lead to no suggestions for improvement,
  but there seems to be some over-determination with boundary conditions.

/ cuda, GPU, status report
  Nothing was done since the last PC meeting in 2015.

+ Parallel IO with HDF5
  HDF5 is the new "portable" file format.
  We agreed on developing (if feasible) a switchable HDF5 module, which
  should on the long run replace all existing file formats. So that a lot of
  code (and hence complexity) can be removed after some years of transition.

x github status: recent increase in response time
  Philippe asked GitHub about the slow response time and master branch merges.

- messy initial conditions folder 
  (WL: I disagree that the initial condition folder is messy, 
       unless it is said that the special folder is messy too)
   AB: yes both are messy; for the time being a proper README file would help

x better documentation of the alternatives to IDL (Python and Julia)
  (WL: I've been telling newcomers to avoid IDL and use python instead but 
       there's no proper documentation, either in the manual or the subroutines themselves.)
  (Philippe: Julia is the better choice over Python and IDL. If you need
       some Python/IDL modules, you may remote-control them from Julia.)
  It is the obligation of the responsible developers to document their routines.
  For example, one could add a hint in the quick-start guide, after the Python
  routines and procedures have settled.

+ Combining special modules
  Wladimir and Matthias proposed to link special modules dynamically.
  Philippe would maybe ask a computing center, if this is fine for them.

- The code is getting too multipurpose, is it a good thing? Colin described it as "walking into someone's attic". 
  Pencil is an engine and a toolbox, and right now they're amalgamated together. Ideally we should separate them: 
  a clean engine and an extended toolbox. The engine should be magnetic.f90 with the induction equation and little
  else, and all the "battery term", "lambda term" and other user-specific things should go in specials. 

- wlyra@caltech.edu
    SMTP error from remote mail server after RCPT TO:<wlyra@caltech.edu>:
    host mail-incoming.caltech.edu [131.215.239.181]:
    504 5.5.2 <norlx51b>: Helo command rejected: need fully-qualified hostname

+ next PC user meeting: Newcastle, 2017