2019:Technology outreach & innovation/Tools for partnerships – developing the technology our partners need/Notes

SESSION OVERVIEW[edit | edit source]

Session link: https://wikimania.wikimedia.org/wiki/2019:Technology_outreach_%26_innovation/Tools_for_partnerships_%E2%80%93_developing_the_technology_our_partners_need

QUESTIONS:

Magnus: Quite a lot of museums are very small with minimal government funding. I would find it hard pressed to find someone to work with. I think a lot of work needs to go via local affiliates to find local GLAMs with little/no internet presence.

John: This is what FindingGLAMs is useful for. Large part of the research is figuring this out - what tools do these organisations need? What if they don't intend to digitise?

James: I can see the sense in centralising this work, but how will you keep an international perspective?

Sandra: Investigations & discussions will be happening over this year. Don't have a concrete idea right now, but will be keeping this in mind.

John: re: Governance, this is new to the movement, taking on technical responsibility outside WMF, except maybe WMDE. We still need to discuss funding opportunities in Sweden. We will have to actively invite and host events to publicise this. In this first year we'll do research, expand technical team, and in parallel talk to affiliates to figure out best structure and prioritisation.

Magnus: Besides Commons, i think the big dumping ground for museums is Flickr. Flickr has recently introduced limitations etc., potentially an opportunity for Commons to establish itself as a new home for that content. Especially for medium sized institutions, they need assurances e.g. for protection from file deletion, ensuring they are respected. Also, Commons upload is too formal for simply uploading e.g. personal collections. Any thoughts on a staging area for file uploads?

John: re: Commons being a hub in place of Flickr - this is something we've observed. Definitely a need among smaller institutions, might have money for digitisation but not upload/website hosting. Lots of digitised content that hasnt been uploaded. Commons has potential for this, but need to think about how to visualise this to show museums it's the right place. Also need to think about deletionism and see what the community thinks about things like a staging area. Needs a big commitment especially from WMF.

Puik: Regarding the workflow, the first task (clarifying copyright status) is one of the most complicated. Ideally it should be in the enrichment sections towards the end. Copyright status is so messy, so it's a limitation to have to clarify copyright status as step 1. Agree with the need for a space for unclear copyright status material.

Magnus: It would be easier to say 'we upload everything here', with some of it copyrighted and only visible to us (and maybe WMF), and there we can annotate it and move over the copyright-free/CC material as we uncover it.

Magnus: Other people have advocated for a non-free Commons. So that material could be hosted - with permission.

Mark Graham: If you do want to store stuff, Internet Archive offers free unlimited storage. There are APIs, people upload millions of items/GBs of stuff. APIs going in and out. Happy to talk about it.

John: Interesting workflow too - other actors with responsibility in the pipeline.

Gerard: One of the things we do in Wikidata is for images with unclear or negative copyright status, we still document the data in Wikidata. So data isn't owned by Commons, rather on Wikidata. Wikidatafication of Commons. Similar need for Wikidatafication of fair use images on Wikipedia. On en.wiki there are a large number of images with clear copyright status (expired etc.).

James: Support idea for the staging area. Controlling who has access. On a separate note, WMF efforts in the past have been top-to-bottom solutions, e.g. GLAMWiki Toolkit. Whereas the things with most impact are levels of infrastructure e.g. pywikibot, quickstatements. So with research and assessment, how do you see the balance of those things. Full monolithic process vs. separated couplings to make each layer pluggable. Are you looking at both aspect?

Sandra: Not in that framing. We want a modular way of working. We don't build one monolithical thing that solves all problems, but rather smaller modules that fit together in a flexible way.

John: Make sure that these modules have a consistent style with documentation and can fit together.

James: The most useful [?] don't have a look and feel -- it's been an API or an infrastructure component.

Magnus: Best of both worlds if the monolith application uses an API that is documented. Official solution and people working against the API with their own solutions e.g. fringe users or separate projects.

John: This isn't an attempt to take over all the things. Thousands of edge cases.

Asaf: Regarding the staging area idea. A single tool-builder couldn't accomplish this, it takes fundamental changes to Commons itself. Every Commons contributor could use a dashboard with useful notifications.

Magnus: For small/mid-size contributors, they can get an internet presence for free that is multilingual/discoverable. More than a category tree is necessary for this. Dream scenario should include files that are made available publicly but not under a free license, bringing them together in one view. Is MediaWiki the right platform for that kind of presentation? Crotos (http://zone47.com/crotos/?l=en) is separate from MediaWiki but relies on its infrastructure. Maybe scope for WikiMuseums or WikiGLAMs or something which doesn't run on MediaWiki.

John: What is the role of MediaWiki in hosting images? Are we the right owners of that infrastructure for non-free images? It's a bigger discussion. Lots of discussion and community consultation to figure this out.

Gerard: When you think about collaborating with Internet Archive, they could use Wikibase.

Simon: Talking about decentralised tools. A threshold where tools need to be adopted more centrally.

John: Requires a lot more stakeholders from the movement. Figuring out that process is a focus of the next year.

John: Yes, if a tool is absolutely required for GLAM partnerships. Bring it in as a module.