Dear fellow Fedora users and contributors,
I would love to interact more systematically with Fedora folks who are interested in social science, including, but not limited to, social science software. I am keen to think about what it would take to make Fedora workstation a more promising distribution for social scientists to use.
Anybody interested in co-thinking? Looking forward to your thoughts…
That’s a great idea. I’m open to any collaboration. However, I have myself no additional need for specific software on Fedora, although I am open to support it. Do you have already requirements/objectives to tackle/achieve in mind? Social sciences about Fedora/Linux/technology or on Fedora/Linux/technology?
Looking forward to collaborations!
I will also think a bit about ideas, concepts, approaches in the coming days. Especially what we could contribute to the research of socio-technical systems/societies (which seems to be a focus for both of us).
I think an easy way to create historical/statistical maps would help.
I don’t have much time to spare, but as an ex-social scientist, I’m definitely interested in this. Documenting the apps that are available as well as any gaps would be really useful.
I’m a half ex, half in of a social scientist, working at a German university. I majored in social science and minored in applied mathematics. I worked on the requirements and commissioning of IT equipment for social scientists for some years (until I choose some other field of research). And until recently, I was the head of the IT department for several research institutes (as a part-time job in addition to my research work). Therefore I’m very interested in this topic. I would like to include servers in the discussion as well.
Would love to extend that conversation out from Workstation and Server to the Cloud Base Images as well. Keep us in the loop.
Social scientist here in psychology and neuroscience.
I am definitely willing to dedicate some time to this!
it is great to see so many voices and so much interest in this. I have not yet much experience with the SIGs, but I propose we form a SIG at this point, following Creating a Fedora SIG - Fedora Project Wiki. That would imply having a wiki page, to I am happy to add the various ideas and requests voiced in this discussion, point to some further resources of likely interest, etc.
Please react with if you like translating this discussion into a SIG.
Generally, I would love a SIG. But maybe we should discuss mission, goal, objectives first? There are already some great ideas we can build on.
- creating software for historical/statistical maps easily (this would be a bigger goal)
- documenting what Fedora already provides to (or issues there are for) social scientists (maybe also document the packages that can be used through rpmfusion, flatpak or pip?)
- social science-related preparation of server/workstation/cloud (what to adjust for social scientists in each? only packages?) Here, I would argue that Silverblue/Kinoite’s immutable nature can be an advantage, too. It can avoid changes to the research environment that could corrupt/affect the result/outcome, creating consistency throughout, although this seems to be less critical to most social science research.
We should consolidate a mission/goal/objective that we can reliably achieve with the number of people that will contribute. A clear benefit may also facilitate others to join
Btw, it is a bit offtopic, but as you are all from social science-related areas: The FOSDEM 2022 will have a Legal and Policy Issues devroom. Maybe that’s interesting for you, too. The event schedule will be published in the coming days on the page.
Regarding a vision – I think it would be great if Fedora would be a distribution that can be publicly well recommended for social scientists interested in using Linux. At the moment, I feel that is not the case yet (because of missing packages).
So, in terms of packages: What would we need to clarify (building on @py0xc3 post)?
- What software is already available as ready packages, in which ways?
- What software needs to be packaged?
- What software needs to be programmed?
Answers to that can be collected in a wiki.
I am not well positioned to comment on “social science-related preparation of server/workstation/cloud”.
A “work programme” might look like this
- Collect information on packages on wiki
- Package relevant software that is not yet easily available on Fedora.
- maintain a wishlist of software (but I think the task to programme new software is too big, at least for me.)
- Revise work programme
Complementary, an issue board on pagure can be added, e.g., as used by the QA team: https://pagure.io/fedora-qa/issues
To facilitate async. issue handling, discussions and schedule of what/when/who to do on the documentation/wiki. Also, to clarify things in a comprehensible manner before adjusting the documentation (or generally, clarify questions that may also be relevant to other social scientists). E.g., one misses a package while someone else knows that the package is already provided in some repository (maybe it’s no science-dedicated package), or knows an alternative or so; or how to adjust Fedora for a specific scientific task. This can be open to any social scientist and add indication about what to tackle in the docs.
Much documentation content may start as a ticket:
- new issue: I need package A, has anyone an idea where to get it?
- no, it doesn’t exist yet. We have to add it, as something that exists (so that scientists are aware of it in general) but that remains a need, to the wiki (task A) and schedule to package it in future (task B) - who to assign?
- once packaged, adjust docs (task C).
Alternatively, as indicated above, an issue could also be used about how to adjust a Fedora for a specific scientific purpose/analysis (if interesting for others as well, the outcome may be also added to a doc?)
Just a rough example.
So, does this refer to creating a group on pagure? I am happy to do that.
I like the idea of using a ticket system for tracing progress and inviting interaction with people outside the SIG.
I’d suggest, we wait for a day whether some arguments emerge that nuance or problematise the idea of a SIG with pagure and wiki.
Absolutely. Just added a rough example of a ticket above. Your ideas apply the same way of what a ticket system may achieve. Its a complement for organization and external contacts for the SIG. Looking forward to comments.
Btw, just to avoid misunderstandings, your account you use here or on the wiki also enables pagure access.
As it ain’t a big issue to create/delete it, feel free to check out and try a potential pagure group with a repo for issue handling:
group: Group SocialSciencesFedora - Pagure.io
repo for issues: Overview - SocialSciencesFedora/SocialScienceIssues - Pagure.io
issue board in the repo: Issues - SocialSciencesFedora/SocialScienceIssues - Pagure.io
→ maybe a second repo for socio-technical research collab? Just to facilitate a discussion about that as well
Hey, I would be happy to be co-maintainer on these - group, repo, board
And I have now created a link at Category:SIGs - Fedora Project Wiki to collect/maintain the links and resources discussed here.
Sorry, forgot I have passed the admin of the group to you (as far as I can see, it can have only one admin, although this is not relevant for the repos). Also, I added you (and the group) as admin to both repos (just added a socio-technical repo; we can delete it later if there is no interest).
See: Group SocialSciencesFedora - Pagure.io
I added the remaining participants to the group as well (hope that’s ok). @giuliogabrieli and @umbrado seem to have not yet signed up on pagure.
I suggest to use this thread in discussions.fedora as an interim thing, to setup the SIG and determine objectives and such, not for permanent/ongoing discussions within the SIG (the SIG page links to here). That thread would become very ugly over time
I have no experience with the groups; let me know if it is necessary to add you additionally to the repos. I have added already the group itself to the repos, so I hope group members will be in automatically (let’s find out : )
I know it currently doesn’t fit the “least privileges” approach, but we are still in the discovery phase