Talk:Movement Strategy/Initiatives/Identify Wikimedia's Impact

Experiment in Portuguese Wikipedia

I like to work in researches and tools to identify topics of impact, but in my experience that is not a easy task. I created this tool for Portuguese Wikipedia more than a year ago, the tool create a priority number for each article based in the size of the article in bytes and the number of pageviews, the idea is that articles that are small and have a high number of pageviews are priorities for editing, because we are providing very few content related to the demand for that information, thus edit that article is more impactful that edit an article that have a big content and a low demand. I disclosed that tool in ptwiki village pump when it was available, the comments were positive, but I realized that the tool is not been used, articles that were at the top of the list a year ago are still at the top, some are "X is Y" articles that can be removed from the top of the list by simply adding a phrase or a source. What I noted with that tool and other similar tools that I have made before was that the editors have some resistance in follow lists of priorities. I don't know why that happens, but that is something we need to take into account when developing methods to list the most impactful subjects to collaborate, we need to make something to make the volunteers interested in follow the list. Danilo.mac talk 21:56, 12 November 2021 (UTC)

@Danilo.mac: Thank you for posting here! You raise some essential points regarding this initiative that perhaps are yet to be answered, but I'll try to address what we know and how your interests may, if you'd like, contribute to it.

This initiative was prioritized previously in the Strategy process, so we know it's an important one, but it's also really complicated to start with. As you mentioned in your own experience, contributors have their own personal interests and may not be typically drawn towards, nor even convinced in, the "impact" that certain topics may have over their own interests. Still, Wikimedia projects have a very wide scope of knowledge that may never be fully covered and, obviously, prioritizing certain areas of this knowledge may get us to increase our "impact" a lot.

As for your great tool, I'd say that you started in the best way: That is, by exploring a method to prioritize the gaps on Wikimedia projects. This is exactly what this initiative currently aims to address. In a later step, we will work on figuring out how to bridge those gaps (maybe through editing competitions, for example). But we don't have to worry about it at the moment, because there will be probably one way or another to do it.

Unfortunately, there is currently no ongoing discussion about how to "Identify Wikimedia's Impact". There might be one in the near future, though, and I'd be happy to ping you then, if you're interested. Also, if you'd like to further explore this topic, please consider applying to a Movement Strategy Implementation Grant to research it. I'm happy to assist you in applying or anything else you need --Abbad (WMF) (talk) 18:54, 18 November 2021 (UTC).

Indeed, it is hard to convince editors to edit something out of their own personal interests. When we start to try some approach to do that maybe we can test it in ptwiki first as we already have that tool.

I have explored the initiative of find articles for prioritize in ptwiki for a long time, in 2011 I worked in an algorithm to evaluate the article quality automatically, in 2014 we converted it to a Lua module to make the evaluation in real-time, and I developed a tool for show the quality x importance table also in real-time (e.g. table for physics). That works well and help to find articles to prioritize, but it has some limitations, it depends on volunteers to put a template in the article talk page and fill the subject and the importance number in it, and the precision is restrict to 6 levels of quality and 4 of importance. Some years later I tried to create something that work for all articles without extra work for volunteers, to do so I made this research to automatically discover the topic of the articles (based on the pattern of the first phrase of the articles that usually is "<article title> is a <article topic> ..."), and I made a tool to show the table pageviews x article size (e.g. table for "book"). I disclosed that tool in the ptwiki village pump, but it seems to not being used, I thought that was because two parameters (pageviews and size) makes it too complicated, so I create a formula to merge those two parameters into a unique "priority" parameter, that is the tool I mentioned in the beginning of this topic.

After so many attempts without see a significant use of the tools, I am not so excited in continue that work as I was some years ago. I hope what we did in ptwiki can help other wikis, at least it helped to prove that volunteers don't like to follow priority lists. Danilo.mac talk 22:45, 18 November 2021 (UTC)

"Map topic areas" bullet item

In the recommendation, "Map topic areas in close collaboration with projects, partners and local communities to benefit from their expertise and ensure their autonomy and independence" is a sub-bullet of the "Conduct and support research and analysis to identify topic areas that have a high impact on the world and on knowledge consumers" bullet item. However, it ended up in the Bridging Content Gaps initiative instead of this one. Was that intentional? I don't think it makes any sense, neither structurally (as a sub-bullet item, it related to its parent, now that context is lost) nor topically (mapping topic areas is about identifying impact, not about fixing the identified gaps). Tgr (talk) 06:19, 1 July 2023 (UTC)

Articles that are popular in many big wikis

I create this list of articles that are popular in many big wikis. My idea was identify topics importance and impact, but I not sure how useful this list really is. Can someone give me a feedback about that? Does someone has a better idea to generate a list with topics of impact? Danilo.mac talk 20:44, 14 July 2023 (UTC)