Pages for networks / participant organizations, scoping

[ahahn-gbif]

(NB: this is not identical to the use of the term "network" in e.g. issues 381 and 69, where we refer to regional sections of the GBIF network)

Rationale: Data publisher registration and data publication through GBIF happens via GBIF Participant nodes, who i.a. endorse new publishers in their domain. For country participants, visibility and credit is given to the node in the country pages. Typically, all datasets of a publisher registered under and endorsed by a country Participant will be assigned to that country. Associated Participant organizations, on the other hand, are frequently thematically focussed, and are involved with the mobilization and publication of data across a wider geographic area. They also typically are involved with only specific datasets from a wide range of publishers, who publish other data as well that are not relevant to the associated Participant's thematic focus. For this type of Participant, appropriate visibility and credit needs to be given at a similar level as for country Participants.

To date, a very basic representation is given, including a short descriptive text for the network and a list of relevant datasets. The information which datasets are relevant to include in that list is manually maintained. Example: https://demo.gbif.org/network/2b7c7b4f-4d4f-40d3-94de-c28b6fa054a6

Elements to include (simple version / advanced stage)

• title (s)
• description of the network (s)
• network logo, if available (s)
• contact information (s)
• basic statistics (count / sum of datasets per dataset type (incl. event core datasets, metadata-only) and occurrences / checklist / sampling event records) (s)
• a list of relevant datasets (s) -- including publisher name -- including basic statistics (type, count of records)
• map of occurrences from all datasets in the network (a)
• literature citations of reported data use / publications (DOI) (a)
• more advanced statistics / breakdowns (a)
• ranking of datasets by the size of their contribution (to be defined) (a)
• latest dataset published (a)
• countries, territories and islands contributing data to the network (a)
• IPT installations contributing data to the network? or list of publishers (a)
• news and events tagged with the network title (a)
• downloadable report, similar to the country reports (a)

Related issues to address / discuss:

• Terminology: we use "network" for different entities, which is confusing: -- GBIF participation as a whole (https://demo.gbif.org/the-gbif-network) -- an Associated Participant who acts as an endorsing node and acts on behalf of a network of data publishers on their side (https://demo.gbif.org/network/2b7c7b4f-4d4f-40d3-94de-c28b6fa054a6) -- potentially: a bracket for a thematic collection of datasets, without explicit Participant status (e.g. country-slices of eBird)
• main menu structure: where to include?
• network visibility on individual dataset pages ("is member of: ..."/"contributes to: ...")
• clarity on different hierarchical entities across data publication and GBIF participation, e.g. OBIS as a data publisher / an associate participant / a network
• workflow/mechanism to register a dataset to a network from the side of the publisher instead of centrally
• internal definition of a "network" as a decision basis for setting up a network page for a (requesting) entity, y/n

[dschigel]

Nice! I would add

• split by publisher & stats
• definition of network (at least for internal use)
• related to visibility - formula to autogenerate the citation for the dataset belonging to a network "proudly delivered to you via..." or a more serious version, I propose

to change

Chatzigeorgiou G, Faulwetter S (2016). Polychaetes from two subtidal rocky shores of the North coast of Crete, collected for the NaGISA project 2007-2008. Version 1.2. Hellenic Center for Marine Research. Sampling_event Dataset https://doi.org/10.15468/b7dt2n accessed via GBIF.org on 2017-07-21.

to

Chatzigeorgiou G, Faulwetter S (2016). Polychaetes from two subtidal rocky shores of the North coast of Crete, collected for the NaGISA project 2007-2008. Version 1.2. Hellenic Center for Marine Research via OBIS. Sampling_event Dataset https://doi.org/10.15468/b7dt2n accessed via GBIF.org on 2017-07-21.

The shortest version I can imagine is to add " via [abbreviated network name]" after the name of publishing organisation.

[timhirsch]

The simplest approach to this as a starting point would be to say we give it the same elements as the current publisher page on demo.gbif.org, which gives us all of the content described by Andrea plus stats (datasets, occurrences). I do not agree that the network should be included in the auto-generated citation for the simple reason that a dataset can belong to more than one network, and 'via' implies an exclusive association. It should be enough to ensure that dataset pages include a list of which network(s) the dataset belongs to. I am also not so sure about splitting by publisher - how would that work? We can't list publishers associated with the network since some publishers will have datasets in several networks or none. As long as the listed datasets include publisher name in the information displayed on the network page, that should be enough.