[NeXus-committee] is NeXus a "data exchange format"

[Herbert J. Bernstein]

Dear Colleagues,

  In my discipline we have been unable to resolve the question of what
is the "real" data that should be archived/exchanged.  Different
people have different requirements.   I think it is a mistake to make
the
decision at the level of the format specification.  I would suggest
that the format be as liberal a container
as possible, able to absorb any combination of raw, partially
processed, fully processed or completely
fictional (i.e. pure conjecture models) as possible and let the
beamline scientists, the experimentalists,
the end users, the theorists and the archivists decide which parts of
which files are useful to them.  If
somebody has something they consider useful, we should provide a
convenient way for them to store it
that is sufficiently well-documented that they have a chance of
reading it back later, and, better, that
there is even a chance that a third party might also be able to read
and understand it.

  Regards,
    Herbert

On Mon, Sep 1, 2014 at 9:38 AM, Mark Könnecke <mark.koennecke at psi.ch> wrote:
>
> Joachim,
>
> Am 01.09.2014 um 15:17 schrieb Joachim Wuttke <j.wuttke at fz-juelich.de>:
>
>> Does the manuscript reflect current consensus:
>> NeXus aims to be "a common data exchange format",
>> i.e. not a primary raw data format, nor an archival format?
>>
>
> This is the consensus; I will will check if this reflected properly.
>
>> Why then all the application definitions at raw count level?
>> Why should I wish to _exchange_ event-mode logs prior to
>> reducing them?
>
> * Application definitions at raw data level facilitate collaboration on data reduction software. And especially
>   with event mode data you may want to rerun the reduction with other binnings etc.
> * Application definitions at the processed data level facilitate collaboration on data analysis software. And there are processed
>   data application definitions.
>
> Regards
>
>     Mark
>
>
>>
>> - Joachim
>>
>> _______________________________________________
>> NeXus-committee mailing list
>> NeXus-committee at nexusformat.org
>> http://lists.nexusformat.org/mailman/listinfo/nexus-committee
>
>
> _______________________________________________
> NeXus-committee mailing list
> NeXus-committee at nexusformat.org
> http://lists.nexusformat.org/mailman/listinfo/nexus-committee
>