[Esip-documentation] ACDD open topic: publisher definition(s)

[John Graybeal via Esip-documentation]

Other ACDD topic was around the publisher definitions. Summarizing the emails:

Nan wrote: 

> …we're back to a non-definitions (self-reference) for this term: 'The name of the person (or... ) responsible for publishing the data file or product to users, with its current metadata and format.'
> 
> I think this should be the person who assembled the data and takes responsibility for the current metadata and format.

proposing to change "responsible for publishing the data file or product to users, with its current metadata and format" to "responsible for assembling and providing the data and metadata in the file or product, in its current form."

I agree with Nan's suggestion, which to my mind is just a rewording of what was originally intended. (In our discussions, we meant "with its current metadata and form" to incorporate the idea of assembling.) 

This post has both Ken and Nan's follow-on comments: http://www.lists.esipfed.org/pipermail/esip-documentation/2014/000425.html. To Ken's point, it seems to me that the publisher is quite distinct from the author/creator, and the ACDD attributes support that distinction in pretty good alignment with the data citation/DOI context. (The main difference being you don't often see DOI's being re-published by other publishers, which happens with data all the time. I think "in its current metadata and format" indicates the most recent publisher is the one who should fill out this attribute.)

Reading Nan's last points, I want to make sure the publisher is the one that put the data online in its current form. If software aggregates and publishes a data file exactly as it got it from somewhere else, the metadata doesn't need to change. But if a publisher converts a netCDF file from monthly accumulations to yearly accumulations, or the format from SOS to netCDF, then the most recent publisher is the one in the publisher* attributes. Previous publishers should be identifiable from source, history, or previous_* attributes. (Don't worry, I made the last one up, it isn't in ACDD. ;->)

John




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.esipfed.org/pipermail/esip-documentation/attachments/20141120/047719f0/attachment.html>