Re: [h5md-user] box and observables

[Peter Colberg]

On Mon, Sep 23, 2013 at 11:48:27AM +0200, Konrad Hinsen wrote:
> Why not take the same point of view for "observables"? We'd then have
> only subsystem-naming groups at the top level, and all observables, 
> including "box", inside them.

I like the idea of replicating the subsystem structure from "particles"
to "observables". But given that the observables are not specified in
detail, would this proposal allow for further subgroups within the
subsystem groups?


As for the box, I don't like the idea, I don't like it at all.

This sets false expectations (that all observables within a subsystem
group are sampled on the same time grid). Further, it only shifts the
problem of data sampled at different times one level deeper, from the
root groups (observables) to the subsystem groups.

I prefer to have a single location for the box, <H5MD root>/box.

Peter