Re: [h5md-user] Modules

[Konrad Hinsen]

Peter Colberg writes:

 > Then, how about using one subgroup per module, as follows:

Fine as well with me. The subgroups would thus be empty initially,
unless the module itself specifies that something should go inside,
right?

 > where "system" contains, e.g., the string "SI" for SI units. But I
 > will write no further on the topic of units, since I am not qualified
 > as a user of Lennard-Jones units…

Even that is a unit system, whose use should be noted somewhere so
that readers know what the numbers mean.


Peter Colberg writes:

 > The specification will not be revised as often as a software program,
 > thus single-level version numbers are practical. The benefit is
 > simpler H5MD version comparison by a computer program.

Sounds reasonable. It makes sense (to me at least) to have a
single-level version number for the data and a two-level version
number for the specification. A minor version change in the
specification would indicate changes to the text (clarifications,
corrections) that have no impact on the data format.

Konrad.
-- 
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: research AT khinsen DOT fastmail DOT net
http://dirac.cnrs-orleans.fr/~hinsen/
ORCID: http://orcid.org/0000-0003-0330-9428
Twitter: @khinsen
---------------------------------------------------------------------