Axel Huebl

Thanks for documenting this. I had it on the radar and thought I made an issue for it already, but this seems to not be the case yet. Done now!...

not clean. better way: - keep 1d arrays - `particlePatches/numParticles` - `particlePatches/numParticlesOffset` - create directories - `particlePatches/offset/` - `particlePatches/extent/` allow arbitrary **records** already defined in the species in these new...

We should combine this with a solution to store boxes of fields (cannot find issue right now for this), so that we can reconstruct domain decompositions even if no particles...

If we find enough documentation (and if the performance is not as bad as "parallel"-xdmf reading in tools), we can add a small script that automatically adds VizSchema markdown for...

Hi @cosmonaut-ok, Thank you for the idea! We plan something very similar, currently drafted as "variableBased" encoding related to #221. In that, we will use the intrinsic capabilities of e.g....

:+1: Also, instead of "stored", "represented" would be more accurate. The standard does not care about how a file format "stores" physically, we care about how it is *represented*, e.g....

yes, that is the first point: methods could be simple striding or more sophisticated methods with various kinds of averaging.

> I'd like to suggest to separate those two cases more clearly in the standard. ("A speciesType is either a single species or multiple species. Multiple species are a semicolon-separated...

Thanks for the suggestion. Another more compact solution we use in other places is to use the same field, but only defined the prefix `other:`. For example: `particlePush` value `other:somethingThatIsUnspecifiedButCanBeAnythingUntilTheNextVersionGivesItAName...

Not that I am aware of, that's the idea with the prefix.