GeoNode Workshop protocol 05.06.2024 - 09:00 - 12:00

[mwallschlaeger]

Participants: @specka , @mwallschlaeger , @svoboda-zalf , @krmeierzalf , @KevinUrbasch , @silvadealmeida

1) Organizational Topics

1.A) Writing Language

I stumbled on german and english documentation, discussions and issues along my way through github and other documents. I suggest only writing documentation, issues etc. within the zalf-geonode repository in English so that it’s easy for everybody to adapt to our progress. Additionally our github repositories and issues are publicly available and others may be interested in discussions and conclusions we made along the way. So try to make all discussion publicly available so we as a team but also other contributors can understand why things are the way they are.

The best book on how to run a successful open source software project is written by karl fogel, also interesting for non developers: https://producingoss.com/en/producingoss-letter.pdf

Discussion results:

• All discussions and communication takes place on github
• Central point of coordination is the github zalf-geonode project board https://github.com/orgs/zalf-rdm/projects/6
• Info panel guidelines people to the right repository to create and discuss issues
• We use “@” syntax to address people directly
• Results and agreements of physical or virtual meetings are posted into a public project wiki
• Within zalf-geonode we do not distinguish between fdm and dis employees, all are treated the same

TODOs:

• @specka will check if we can buy a physical book (https://github.com/zalf-rdm/zalf-dis-operations/issues/996)

1.B) Use and role of Github, issues, projects and users

How to discuss issues and how to and where to open an issue. Because there are potentially so many ways to notify each other on a task or a problem with a software we must take some time to discuss and agree on how we want to organize the work on zalf-geonode project.

Discussion results:

Github GeoNode Scrum/Kanban: https://github.com/orgs/zalf-rdm/projects/6/views/2 (please use this to create issues)
create new issue like:
Tab Complete Board -> New Issue -> click -> Convert to Issue (right side) -> Label issue -> TF Geonode
Github GeoNode Repository Issues: https://github.com/zalf-rdm/geonode/issues

TODOs:

• @NikoleiSvoboda and @KristinMeier have to sign up with Github
• @mwallschlaeger will add accounts to our organization

1.C) Role of TF Data Publication

We have this Task Force “Data Publication” and for the future development of GeoNode I (@specka ) would prefer to have regular meetings fixed scheduled (bi-weekly), to have a place to discuss GeoNode specific stuff. Right now, this TF seems to be sleeping, or the task / objectives are not clear.

Discussion results:

Regular meetings? → every two weeks; preferably on Tuesday every two weeks at 1pm;
Who is in charge of organizing it? Topic selection? → @specka Xenia und @mwallschlaeger (TF GeoNode)
next dates: 11.06.2024 11am, 25.06.2024 1pm and from there on every two weeks thuesday 1pm

2) Publication Structure

2.A): How to structure data / data collections inside GeoNode (maps, datasets, documents, stories, metadata)

GeoNode supports several object types to represent geodata, non geodata and metadata. We never decided on how a publication is structured inside our GN repository. Everybody in our working group must be familiar with the geonode object types and know our guidelines on how to handle different data formats within this structure.

Discussion results:

• A Map (renamed to "landing page*) is the center of each data publication. To the map all metadata and belonging objects are attached via the linked resources functionality.
• Several Datasets containing geodata or non spatial data are created, one for each table or geodata object. The datasets contain the same metadata as the map, extended by the attribute description (table description)
• Documents (may be renamed to supplemental material) are used to upload supplements like pdf, jpg, zip etc. if the datatype is not supported by geonode documents out of the box, we zip them and add them as documents. Usually each supplement will be uploaded by themself. If it makes sense to group them they can be zipped and uploaded. The supplemental material will be attached to the map (data collection) as well. Documents have the same metadata as the map (data collection)
• GeoNode also allows you to create geostories or dashboards. If we or users are creating such a GeoApp we’re gonna attach this to the map as well. GeoApps have the same metadata as the map (data collection)
  Visualized (https://docs.google.com/drawings/d/1GOT749dDZJSiW3NUsHFdD4Kzdn_XskPtlg1tG-Wy0Ts/edit?usp=sharing):
  

TODOs:

@KristinMeier: 1-n Beispiele (Use cases) für bisher publizierte Datenkollektionen mit Maps beschreiben (Beispiel: Jochheim Datensatz);
Use Cases:

• Mehrere Karten für einen DOI
• Datensammlungen mit versch. Autoren für einzelne Tabellen?
• Datensammlungen mit versch. Keywords für einzelne Tabellen
• Geostories mit DOI publizieren?
  @KristinMeier: Abgleich “Wunschdokument Repo” mit angelegten Issues
  Task: 1-n Beispiele (Use cases) für bisher publizierte Datenkollektionen mit Maps beschreiben (Beispiel: Jochheim Datensatz); geonode#141

User and Group Management

Discussion: Users

Basic problem is that all users related in GN metadata must be users inside the geonode repos. But users come from different sources: ORD, BonaRes Repo, ZALF AD, BonaRes AD, ORCID.

Users must be identified as themself no matter where they come from, and if we create them beforehand e.g. by ORD upload, they still need to be connected to their already uploaded datasets when uploading for the first time after zalf geonode launch.

Further we must discuss how new users enter the system? Under which circumstances users can change passwords etc.

References:

• https://github.com/zalf-rdm/OpenResearchData_Migration/issues/11
• Question: authors and user accounts in GeoNode geonode#97

Discussion:

• Users can use different authentication systems (ZALF AD, BonaRes AD, ORCID) to authenticate; user accounts can be merged if the same email address is used; than different user
• Question: how to deal with “further authors” → in G, an user account must be created;
• Example: ORD data → 300 persons/authors; some still existent at ZALF
• Should email address of user be changed if he changes organization?
• Differentiation between “corresponding author” and “simple author”
• “Corresponding author” entspricht Data Cite Contributor -> ContributorType = ContactPerson
• Bei ContactPerson kann sich die Institution ändern (wird aus ORCID genommen)
• Bei Autoren darf sich das nicht ändern= “Simple author
• We need to avoid duplicates “duplicated users”;
• Question:
• Do users need to login to GeoNode?
• For publishing of datasets no login is required; only Data Stewards will interact with GeoNode
• What about GeoStories?

Discussion results:

This discussion is not finalized but we agreed on the following Workflow for now:

• customers login into upload tool, add contact information (by typing all information or provide orcid id) for all different contact role individuals into the upload tool
• we create contact role individuals as local users in geonode, if they not exist already if they exist we reference to the existing one
  • referencing ether by email addr. or orcid
  • this contact role individuals will exist in geonode as local django users but their unable to login
  • the author who has logged into upload tool beforehand will be the owner of the landing page and linked resources will be able to login to geonode afterwards
• geonode admin users (staff users) are maintained manually by us.
• different groups could be created for different user groups e.g. external publishers like kassel could have their own ecosystem inside our geonode where they publish their datasets with their admins users. (under discussion)

TODOs:

@specka : we need a different name for “Sign in with BonaRes Acoount” → instead of BonaRes a more generic wording should be used; → should be formulated as an issue with label “TF GeoNode” (zalf-rdm/geonode#129)
@specka BonaRes, ZALF, ORCID → TODO: needs new issue in the Upload Tool development (https://github.com/zalf-rdm/bonares-toolbox/issues/963)
@specka What kind of authentication mechanism do we implement in the Upload Tool (UT)

[mwallschlaeger]

Who is in charge of organizing it? Topic selection?
@specka I will take care of the next TF GeoNode meetings and prepare them

[mwallschlaeger]

regarding data protection regulations, here are the responses of forian melzer regarding our plans:

2024_06_13_Datenschutzhinweise Webseite und Mail.docx
2024_06_13_Stellungnahme Datenschutz.docx

ref zalf-rdm/geonode#96

[specka]

@mwallschlaeger How are we managing the TODOs of the discussions? I would suggest to create an issue for each TODO so we can track it and link it here to the discussion. No TODOs would be lost, and we have a nice way of tracking the progress.

[mwallschlaeger]

yes i thought to do the same after writing dicussion, I rejected for now as some of us still miss an github account