boards for prioritizing issues (for each installation of Dataverse)

[Philip Durbin]

Dear Dataverse enthusiasts,

If you attended my "crazy ideas" talk* at #dataverse2019 you heard me encourage installations of Dataverse to create a "board" (technically, a GitHub "project") to communicate which issues are most important to you.

Thank you to all the Dataverse installations who have created a board! Here's the list so far:

- FZJ: https://github.com/orgs/IQSS/projects/4

- Odum: https://github.com/orgs/IQSS/projects/5

- QDR: https://github.com/orgs/IQSS/projects/6

- Scholars Portal: https://github.com/orgs/IQSS/projects/8

- UVa: https://github.com/orgs/IQSS/projects/10

- WZB: https://github.com/orgs/IQSS/projects/11

- SciencesPO: https://github.com/orgs/IQSS/projects/12

- DataverseNO: https://github.com/orgs/IQSS/projects/14

- Inra: https://github.com/orgs/IQSS/projects/15

If you have not yet created a board for your installation, please go ahead! The person creating and maintaining the board for your installation of Dataverse must be a "member" of the IQSS organization on GitHub. We are happy to add you to the "dataverse-readonly" team so that you can create your board. You can open an issue at https://github.com/IQSS/dataverse/issues about this or email sup...@dataverse.org with something like "add me to dataverse-readonly team".

We don't have a formal process for reviewing these boards but I can tell you that I take a look before our sprint planning meetings on Wednesday afternoons. Installations are encouraged to review each others boards as well, of course!

If you have any questions about any of this, please let me know!

Thanks!

Phil

* My crazy ideas talk: https://groups.google.com/d/msg/dataverse-community/lktk-artKjQ/xzCpfGf7BwAJ

--

Philip Durbin
Software Developer for http://dataverse.org
http://www.iq.harvard.edu/people/philip-durbin

[Steven McEachern]

Hi Phil

I really like this idea and ADA will be setting one up ASAP. (How do I get the relevant permissions?)
Just one thought - is there any means of aggregating these boards - so that there is a joint board which the sum of each installation’s boards?
It would require we all use the same board headings, but that seems doable.

Cheers
Steve

[Philip Durbin]

Hi Steve,

I'm glad you like the idea! :)

The relevant permissions to create a board for your installation of Dataverse are tied to GitHub accounts. Let's try this workflow:

- Create a GitHub account if you haven't already.

- Open an issue at https://github.com/IQSS/dataverse/issues with a title of "permission to create a board for my installation of Dataverse"

- Someone from IQSS will add the GitHub username that created the issue (after some vetting, if necessary) to the "dataverse-readonly" team at https://github.com/orgs/IQSS/teams/dataverse-readonly

- Optionally, the new member can make their membership in the IQSS GitHub organization public at https://github.com/orgs/IQSS/people

- The new member creates a board by clicking "New Project" at https://github.com/orgs/IQSS/projects and provides a link in the GitHub issue.

- Someone at IQSS will update the internal spreadsheet* to add the link to the Dataverse installation's board under the "Project board under IQSS" column.

- The GitHub issue is closed with "all set" and "thank you!" or whatever. :)

I hope this is clear enough. I'll attach a screenshot of what it looks like from my side when I invite a GitHub user to the "dataverse-readonly" team.

With regard to aggregating the boards and standardizing column names, yes I've been thinking along these lines as well. For now I'm fine with and quite enjoy the creative column names Dataverse installations have come up with ("on fire on fire put it out" comes to mind) but yes, some alignment in the future would be great. Anyone reading this is welcome to create an issue at https://github.com/IQSS/dataverse/issues about future alignment of column names and aggregation. :)

If there are any questions, please let me know!

Thanks,

Phil

* Lately I've been adding a lot of new columns to IQSS's internal spreadsheet of Dataverse installations and I'd like to open as much of it as possible (probably a fresh spreadsheet) so installations can learn more about each other ("installation personas") and so that we can crowdsource the maintenance of the data. Here's a sense of the columns right now:

- Installation Branding
- Institution or organization
- Description
- Link to installation
- Country
- Continent
- Launch date
- On dataverse.org/metrics
- Project board under IQSS
- DOI authority number
- GDCC member
- Number of CoreTrustSeal subdataverses
- Version (* for fork)
- Fork URL
- Custom about URL
- Custom guides URL
- Contact person name
- Contact person email
- Security contact
- IQSS blog post about installation
- Open to harvesting
- Open to all researchers worldwide to deposit
- Persona
- Interested in big data
- Interested in sensitive data
- Focused collection
- Multi-tenant (consortium)
- Interested in extended archiving
- Primarily restricted
- Number of languages
- Public only

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/8124d994-9443-4ecc-b39c-b157f81b72e5%40googlegroups.com.

[Philip Durbin]

Steve, thanks for creating https://github.com/IQSS/dataverse/issues/6221 and a board for ADA at https://github.com/orgs/IQSS/projects/17 ! I hope the screenshots I added to that issue and the back and forth we had help explain the process for other installations of Dataverse who want to create a board to highlight with of the ~900 open issues are important to them.

With regard to a join board that sums up the issues and achieving some alignment on column names, this morning I spent a little time hacking around in https://github.com/pdurbin/druthers which is my old old idea for "crowdsourced issue ordering". In Druthers, as a developer, I ask myself, "If my users had their druthers, what issues would they have me work on?"

As a side note, I think "druthers" is an Americanism. It means "would rather". Mark Twain wrote, "Any way you druther have it, that is the way I druther have it." The name "Druthers" doesn't matter. I just needed a short name. What matters is the idea of crowdsourced issue ordering, which is what we're talking about. :)

It turns out the GitHub API for pulling out the data of these boards is somewhat tedious (and you have to have a somewhat privileged auth token). For now all I have are the column names (below) so we can at least start thinking if we want to all start using the same column names or not. (To protect the innocent, I did alter one column name a bit.) I also added the IQSS/Harvard Dataverse board to be a little more democratic. Plus IQSS seems to use the most emojis, which is fun. :) Also, I'm only showing below boards for installations that are on the map of installations on the Dataverse project website. The order is random (from the Unix `find` command). Here are the column names per installation:

dataverse.scholarsportal.info
To do
In progress
Done

dataverse.lib.virginia.edu
Smaller Wishes
Bigger wishes (More important to UVA)
Real Bugs
Closed

dataverse.no
Needs triage
High priority
Low priority
Closed

catalogues.cdsp.sciences-po.fr
Interest
Needed
Done

data.qdr.syr.edu
Papercuts (Smaller issues)
Larger issues
IQSS!!!! (Annoying bugs) ✊
Implemented at QDR
Closed

data.inra.fr
To do
In progress
Review in progress
Reviewer approved
Done

dataverse.ada.edu.au
To do
In progress
Done

dataverse.harvard.edu
Community Dev  💻❤️
UI/UX Design 💡📝
Ready 🙋
IQSS Sprint 9/25 - 10/9
IQSS Team Dev  💻
Code Review 🦁
QA
Done 🚀

dataverse.unc.edu
would like to do
should do
on fire put it out put it out
Closed

As time allows I'll work on getting the "cards" (which are GitHub issues), out of each column. Once was have the issues, we can see which are the "hot" issues, or at least which issues installations of Dataverse have added to their board.

I hope you all find this interesting. Please stay tuned and let me know if you'd like to hack on Druthers, which Python. You can follow my progress at https://github.com/pdurbin/druthers/issues/1

Thanks,

Phil

[Philip Durbin]

Ok, I hacked on the code* a bit more. Please find "issues.tsv" attached. I'll also attach a  screenshot.

We could definitely used some standardization about columns like "Done" and "Closed". :)

I'd love to hear thoughts on all of this. :)

Phil

* You can find the code in https://github.com/pdurbin/druthers/issues/1

[Philip Durbin]

Hello! I'm warming up this thread because Geneviève Michaud asked about these boards this morning during the GDCC talk. (Thanks, Geneviève!)

I still think this crazy idea is a good one. Again, the idea is that each installation of Dataverse is welcome to create a "project board" on GitHub. Then, you put the issues that you care about on your board.

Finally, IQSS/GDCC runs a report periodically to see which are the "hot" issues. I just ran such a report this morning: https://docs.google.com/spreadsheets/d/1Dx-MN-gyCWCo2Lv9AQW_2yhWLD9LuO1nDnVhBroUOoQ/edit?usp=sharing

(For the developers out there, here's the script that creates the report: https://github.com/IQSS/dataverse-installations/commit/e87b64c )

If your installation does not have a board, please feel free to open an issue at https://github.com/IQSS/dataverse-installations/issues

Feedback and questions are welcome!

Thanks!

Phil

[Geneviève Michaud]

Thanks Phil! My concerns was how to inform priorizations decisions efficiently (avoiding additional tasks on active developers / maintainers).

See you at DCM

Geneviève