Telco 20220225¶
Date¶
Friday, 25th February, 14:00 UTC
Connect¶
ZOOM VC link: https://psich.zoom.us/j/68711884689?pwd=YlF5VmtNbDdrOHJkYXc2YTlGWWMxdz09
Meeting ID: 687 1188 4689
Passcode: 183302
Agenda¶
Welcome new members
Review previous meeting minutes and actions: January Telco
Recent issues on Github (especially those labelled with “telco”)
-
NXxpcs is a new contributed definition (to be an application definition)
add
NX_ARBITRARY_UNITS
: Does this need a NIAC vote?
-
NeXus ontology and namespace usage
?add to agenda for next NIAC?
Spring virtual NIAC
Autumn hybrid NIAC
Present¶
BW, CZ, HB, PJ, FA, SB, LG, RB, HG, AB, RO, WDN
Minutes¶
NIAC on March 3rd, 4th, with special effort to have a chorum on the 4th.
SB to make pull requests
Ontology email aacount - special email address that forwards to nexus-committee? Or is it just a login ID?
NXxpcs app def - want xpcs data in a common format - is ready for contribution
New type of units “arbitrary units”
SB prefers to calculate combination units on the fly
RO asks what is supposed to be the validation consequences of arb units?
BW says that au is problematic anyway because it is a nonstandard description that is commonly used (add to proposal #921)
SB thinks arb units is equivalent to unknown units.
RO and PJ can agee on “unkown” units as a compromise (NX cansas can consider using this)
PJ will make the changes in NXxpcs
LG wants to be able to state photon flux differently from NXmx since XFELs provide photons differently
Problem is that a pulse has unknown time
Other problem is that NDXL currently doesn’t easily allow requiring one of a set of fields.
AB and HB think this is just a bug and short name should be optional - AB will fix it
SB wants to have a nexus workshop in Germany 17-18 March to bring people and projects together
would like someone to introduce the NIAC, sample environment, validation, data processing/visualisation
BW volunteers to talk at the meeting.
AB thinks that he could also speak but needs to check beamtime schedule.
AB will add the NIAC vote needed tag to #942
SB asks about inheritance - HB says we should look at UML, which adresses this question. HB thinks there is value in having specific inheritance rules. BW would like to see a specific proposal to avoid endless ideological discussions.
SB asks about linking between fields and how a link can be required in an app def (using NXDL)
AB says that NeXus is meant to be independent of the file format implementation (i.e. HDF5 is not a requirement)
SB asks if someone from ESRF could speak at the workshop about NeXus use or utilities? WDN will discuss availability with colleagues.
BW will publish a NIAC schedule asap