Issue 11514

Organizations without any node associated to them at all

11514
Reporter: jcuadra
Assignee: ahahn
Type: Bug
Summary: Organizations without any node associated to them at all
Priority: Major
Status: Open
Created: 2012-06-28 10:47:00.01
Updated: 2012-11-20 16:40:09.891
        
Description: We have several Organizations that are not linked to any node at all.

To get them:
------------
SELECT * FROM agent where agent_type_id=2 AND deleted is null AND id NOT IN ( select to_agent_id from agent_relation where relation_type_id=1 and deleted is null)


List
------------

http://gbrds.gbif.org/browse/agent?uuid=79175520-c240-11dc-ab55-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=d48116e0-281e-11d9-8436-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=ec469dc0-9530-11d9-8902-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=80388cc0-ea72-11d7-8a6f-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=4f160e00-329d-11d9-8439-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=79acbdd0-f405-11dc-ae6d-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=af3a7b70-8fe1-11da-956e-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=1eadc140-5b3c-11da-9b50-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=f2965760-c5d3-11d8-bf61-b8a03c50a862
http://gbrds.gbif.org/browse/agent?uuid=f4ce3c03-7b38-445e-86e6-5f6b04b649d4
http://gbrds.gbif.org/browse/agent?uuid=72da316f-62b4-451e-9238-cf72289e6372
http://gbrds.gbif.org/browse/agent?uuid=1ce482ab-14e3-48f0-8b70-b58b14625902
http://gbrds.gbif.org/browse/agent?uuid=e196c8d6-f795-463c-80c4-310dd14ee50b
http://gbrds.gbif.org/browse/agent?uuid=fbca90e3-8aed-48b1-84e3-369afbd000ce
http://gbrds.gbif.org/browse/agent?uuid=d44af9a3-e779-40c0-a186-79e7717c6d2b
http://gbrds.gbif.org/browse/agent?uuid=7879e569-4a13-4643-b833-d1a564675b86
http://gbrds.gbif.org/browse/agent?uuid=6d2a7654-1ed0-4924-b22e-9bc221ab2124
]]>


Author: ahahn@gbif.org
Created: 2012-06-28 15:41:30.493
Updated: 2012-06-28 15:42:44.366
        
remaining issues after first cleaning round:

1) probably to be removed:

- http://gbrds.gbif.org/browse/agent?uuid=79175520-c240-11dc-ab55-b8a03c50a862
62.147.188.158
Likely erroneous registration (see name, never touched since first registration in May 2010. Suggest to flag deleted, but contact publisher on possible new start:   http://biologie.univ-mrs.fr/view-data.php?id=1

- http://gbrds.gbif.org/browse/agent?uuid=80388cc0-ea72-11d7-8a6f-b8a03c50a862
Finsiel
UDDI leftover? Needs checking, but Finsiel used to be a development contractor rather than a publishing institution. Depending on scope of registry, can probably be removed.

- http://gbrds.gbif.org/browse/agent?uuid=79acbdd0-f405-11dc-ae6d-b8a03c50a862
Mongolia Natural History Museum
See GBIF Helpdesk Case 1429: endorsement selected to be under Korea (17.3.2008), but refused due to data issues. Mails with technical suggestions went unanswered, no further follow-up.
Relationship now set in registry - should not be validated.

- http://gbrds.gbif.org/browse/agent?uuid=f2965760-c5d3-11d8-bf61-b8a03c50a862
Thematic Networks
A bogus entry that seems to have served as a memory aid for values serving as "Thematic Network flags" in the UDDI. Needs to be translated into the new structure (create networks and links to datasets), and then removed

2) currently orphaned, new Node being negotiated:

- http://gbrds.gbif.org/browse/agent?uuid=d48116e0-281e-11d9-8436-b8a03c50a862
Dept. Of Biology, University of Trieste
Used to be endorsed by "EU-BioCASE". Current discussions with CETAF whether their node will be operated by EU-BioCASE; otherwise, find alternative node

- http://gbrds.gbif.org/browse/agent?uuid=4f160e00-329d-11d9-8439-b8a03c50a862
Israel Nature and Parks Authority
Used to be endorsed by "EU-BioCASE". Current discussions with CETAF whether their node will be operated by EU-BioCASE; otherwise, find alternative node (possibly Germany)

- http://gbrds.gbif.org/browse/agent?uuid=ec469dc0-9530-11d9-8902-b8a03c50a862
European Environment Agency
Used to be endorsed by "European Commission", but status unclear. Pending Nodes decision.

3) Partners we are in collaboration with outside the endorsement workflow framework:
(?) do we need to establish a dummy node to connect those to, or can they remain floating?

- http://gbrds.gbif.org/browse/agent?uuid=f4ce3c03-7b38-445e-86e6-5f6b04b649d4
The Catalogue of Life Partnership

- http://gbrds.gbif.org/browse/agent?uuid=72da316f-62b4-451e-9238-cf72289e6372
The International Plant Names Index Collaborators

- http://gbrds.gbif.org/browse/agent?uuid=1ce482ab-14e3-48f0-8b70-b58b14625902
Index Fungorum Partnership

- http://gbrds.gbif.org/browse/agent?uuid=e196c8d6-f795-463c-80c4-310dd14ee50b
The Global Biodiversity Information Facility

- http://gbrds.gbif.org/browse/agent?uuid=fbca90e3-8aed-48b1-84e3-369afbd000ce
ECAT development publisher

- http://gbrds.gbif.org/browse/agent?uuid=d44af9a3-e779-40c0-a186-79e7717c6d2b
International Union for Conservation of Nature

- http://gbrds.gbif.org/browse/agent?uuid=7879e569-4a13-4643-b833-d1a564675b86
The Knowledge Network for Biocomplexity (KNB)

Author: ahahn@gbif.org" rolelevel="10001
Created: 2012-09-17 16:06:43.15
Updated: 2012-09-17 16:06:43.15

Internal mail 21/08/12:

Hi Mélianie, hi Olaf,

There was some discussion coming up last week around the current endorsement procedure for new publishers. As you know, any newly registered organisation currently has to be formally endorsed by a participant node, from which point onwards their datasets are indexed. The explanation we send to node managers in the process includes the following paragraph:

"According to the current procedures, GBIF Secretariat policy is to ask the GBIF-Participant Node Managers whether they want to endorse new data provider installations in their domain. This is a simple quality control step required by the GBIF Participant Node Managers Committee. We understand that it is the Node Manager who best knows the institutions and databases in their country/organisation. GBIF Secretariat does not have such knowledge for individual countries/organisations.

The "[Organisation name]" registered as a new organisation. I would like to ask whether you endorse them to be listed under your country in the GBIF registry and at http://data.gbif.org/datasets/.

Please answer YES, NO, or request further details."

This process dates back to before my time with GBIFS, and was established at a time where a) we did not have procedures for indexing taxon datasets, only occurrences, and b) where we could actually provide the node manager with some kind of data preview, at least the link to the already configured dataset, so that they could take a look at the data. A potential new data publisher would install the publishing software, configure their dataset, and then self-register the access point through a "register me" web page, from which point the endorsement process interaction would pick up.

Since then, several things changed. "Registerme" functionality temporarily disappeared, though it will be re-integrated with the new portal version. The IPT setup workflow requires that an organisation is registered before any datasets can be published through it, so that very often, node managers are asked to endorse an organisation entry with no data preview available to aid their decision. Even in the old procedure, we would ask endorsement for an institution based on a single dataset, while subsequently published datasets are then in the responsibility of that organisation, without intervention of the endorsing node. Last not least, we currently have an inconsistent state where organisations publishing occurrence datasets go through the endorsement procedure, while those starting with a taxon dataset do not.

While we can rework workflows and possibly services (self-registration, dataset previews, automated messaging) to clean up a procedure that slowly crept into the current state, we feel that it might be the time to revisit the endorsement requirement as such. The procedure is based on a decision of the participant node managers some time around 2003/2004. Since then, our landscape changed, and before implementing around old requirements, we would like to check whether the assumptions are still valid, or whether the requirements have changed and a new procedure should be discussed with the participant nodes. Questions include:

- if the endorsement workflow is maintained, should it treat taxon dataset owners the same as occurrence dataset holders? (according to Markus, David Remsen (rightly?) always said the endorsement rules only apply to occurrence datasets and it's a completely new game with species data)

- should the current procedure of endorsement of institutions be maintained at all? Or: is there a good reason to hold back a dataset that the publisher wants to publish? (there have often been discussions around publishing of, e.g., datasets from institutions based in Russia, China etc. The right to publish data through GBIF is probably not the strongest argument when recruiting new participants, and usually some workaround gets them endorsed by a thematic network anyway. Maybe data quality should be handled as a completely separate topic, rather asking for a "rating" from a Node?)

Is this something worth re-opening with nodes (or the steering group)?

Thanks,
Andrea