Commons talk:Media knowledge beyond Wikipedia

Context and background

This essay / statement has been written together quite fast and dirty, as a response to the Wikimedia Foundation 2024-25 annual plan proposal, by a small (and slightly random) first group of Wikimedians in their volunteer capacity. We warmly welcome anyone to join and contribute. Comments, suggestions, updates, feedback and of course signatures are extremely welcome.

The first draft was written together in this Google document - it als contains more context why this statement was published! Spinster (talk) 07:13, 31 May 2024 (UTC)

Siehe die Diskussion auf com:Forum. Ich werde von dieser Version doppelt ausgeschlossen. Zum zweiten weil es auf Google-Servern erstellt wird (Googles Slogan: "Do evil.") und zum ersten, weil Menschen, die wie ich für Wikipedia auf Demos fotografieren, darin nicht vorkommen. Ein wenig Support - wenigstens mit Worten - für die Gefahren, denen sich Menschen für Wikipedia aussetzen, wenn sie dieses "Wissen der Welt" draußen in der Wildnis einsammeln, wäre schön. C.Suthorn (@Life_is@no-pony.farm - p7.ee/p) (talk) 11:31, 5 June 2024 (UTC)

English pls! Also disagree. Prototyperspective (talk) 12:33, 5 June 2024 (UTC)

Das Wissen der Welt, wenn es in Englisch ist. Was ist denn "pls"? public limited societe anonyme? C.Suthorn (@Life_is@no-pony.farm - p7.ee/p) (talk) 20:49, 5 June 2024 (UTC)

@Spinster Thank you for taking the lead on this. As concise and powerful as it is written, I must admit, that im surprised to only find "collecting" and "sharing" mentioned as central goals of Commons. But what about "creating" and "collaborating"? IMO these need to be included to paint the full picture of the commons community. MB-one (talk) 22:37, 5 June 2024 (UTC)

@MB-one This statement has indeed been written first with collecting in mind (the typical GLAM-Wiki activities), precisely because I have explicitly heard Wikimedia Foundation staff say that they don't think this type of work is useful for Wikimedia's mission - which I find extremely worrying. I personally think the strength of Wikimedia Commons is exactly in the combination of both new photography and creation and combining it with "heritage" materials: we collaborate as a community, for instance, we combine current photos of buildings with their blueprints from GLAM collections, current photos of nature with historical records of them, current diagrams and maps created by volunteers with historical ones... that's where our strength lies and where we create a unique contextualized resource of media knowledge that exists nowhere else. I'm happy to update the text to include creation and collaboration if folks felt left out, which was absolutely never the intention. I only hope we see each others' actions as done in good faith, because (as you can see in responses below) we'll need to unify and work together to make sure Wikimedia Commons will be further resourced and reaches its full potential. Spinster (talk) 07:33, 6 June 2024 (UTC)

I have added extra wording to the original statement to now very explicitly include those activities. Spinster (talk) 07:45, 6 June 2024 (UTC)

I would also add that, here in Seattle, I've seen enormous value in bringing together in one place content from the Seattle Public Library, the Municipal Archives, the University of Washington Libraries, the Museum of History and Industry, and other sources. We've repeatedly been able to help these institutions better understand and document what is in their own collection by way of comparison to content of other collections that they apparently had never looked at. Even from a single institutional source, our category system repeatedly had revealed connections (and in many cases, corrections) of which these institutions were previously unaware. - Jmabel ! talk 14:46, 6 June 2024 (UTC)

Missing section on Commons in annual plan

Please note also Commons:Village_pump/Archive/2024/05#Feedback_period_about_WMF_Annual_Plan_for_2024-25_is_open! on how to include it. Enhancing999 (talk) 17:39, 5 June 2024 (UTC)

Response from the Selena Deckelmann, CPTO at the Wikimedia Foundation

Hi there,

Thank you for your engagement and attention to the Wikimedia Foundation’s Annual Plan and thank you for all the attention given to this open letter. I am encouraged by the interest, and feel that now is a good time for us to dig into an important conversation.

One aspect of the Annual Plan that I’ve heard remarked on is a focus on Wikipedia. While the support of free knowledge will always be broad in scope, I’ve made supporting Wikipedia an explicit focus of our planning for the coming fiscal year because we must use the finite resources we have on the core infrastructure for assembling content, serving it to the world and fundraising, and then make difficult trade off decisions. Main factors in prioritization of work for Wikipedia include: supporting the key workflows of volunteers who are assembling knowledge, support for the ways and scale at which the world accesses and consumes the knowledge, supporting the safety of all volunteers, and covering prioritized knowledge gaps. I’m going to name a few of those tradeoffs I have already made and described in the previous annual plan below, but more decisions remain which I would like to make with you. I’m always trying different ways of explaining these choices, because I want everyone who is interested to be able to engage and participate. Hopefully this adds to your understanding.

Just to review the work that has been in flight during the current fiscal year (July 2023-June 2024): in direct response to community requests and consultations, we took on projects with the aim of relieving Commons’ administrators from some of the moderation burden they currently have. In this effort, we revamped and improved the current user experience with UploadWizard, and fixed some persistent bugs related to it. In order to further reduce the number of bad uploads through UploadWizard, so that in perspective we can reduce the number of deletion requests, we are experimenting with a tool to automatically detect logos when uploaded on Commons through UploadWizard, in order to facilitate their evaluation by the community. A need for machine detection tools was raised in several discussions and user interviews we had in the past with the community, and logos are the second largest reason for media deletion.

Going forward into future years, there are two key challenges on my mind:

1. Commons is large and growing, and has many needs, tools and advocates. Despite some challenges, I think it is a lively, fascinating and creative project worthy of greater support. While we do not fund project work based on the dollars raised by a project, it is true that according to feedback we regularly receive from volunteers, we don’t yet meet the needs of our Wikipedia communities. In that environment, what criteria might we use to appropriately fund the work across the various projects? A specific challenge I am wrestling with is reconciling the urgent concerns and challenges Wikipedias face (e.g. stable but stagnant editor numbers, declining adminship, increasingly sophisticated vandalism and threats from the growth of generative AI systems and fundamental changes in internet search) against the urgent needs of many other projects. With this in mind, I have prioritized improving the state of our fundamental platform for all projects, MediaWiki, because that investment will improve both Wikipedia and the other projects. This work began in FY23-24, and will continue for multiple years. The fundamental architecture of MediaWiki has not substantially changed since its creation, and due to the scale of Wikipedia and other projects today, as well as a changed internet environment with higher regulatory, privacy and security standards that must be met, we need to rethink a few things and make development with MediaWiki easier in the future. This tradeoff decision was multifaceted – we reduced investment in several teams, eliminated teams, and formed a larger MediaWiki cluster of teams: MediaWiki Core, MediaWiki Interfaces and Content Transform. In a similar vein, I’ve also prioritized Trust and Safety investments that impact all projects, such as areas that Stewards manage. Once these areas are in good shape, we will have an opportunity to get more specific about whether and how to shift significant resources to specific Wikimedia projects. But these will be tradeoff decisions, as for the foreseeable future, the budget for the WMF will be mostly flat, while at the same time we grow grants to community-determined projects, and maintain our investments in staff and infrastructure like data centers to keep projects available and properly maintained.

2. Currently, Commons’ self-described mission has to do with Commons as a good in and of itself. In practice, this means it is hard to leverage Commons in the places where our audiences go to consume knowledge, principally Wikipedia, because images that are highly relevant to Wikipedia can be difficult to make part of the Commons collection. WMF has an obligation to invest resources in a way that furthers not just knowledge collection, but knowledge dissemination. To what extent is the Wikimedia Commons community invested in the dissemination of knowledge via images on Wikipedia articles? To date, my observation is that the primary focus of the Commons community is the collecting of free content, rather than its dissemination. If that’s the case, we should talk about these differences more openly together to plan a way forward.

And you all have many much more specific questions and issues you want addressed. I think we have many committed volunteers who would like to see us come to conclusions together and move forward with action, and I want to work with you all to do that.

In the next few weeks, I’d like to have an open call on these topics to further address these questions and then take next steps. If you are interested please reach out to me or look for an update on this discussion page. SDeckelmann-WMF (talk) 02:18, 6 June 2024 (UTC)

@SDeckelmann-WMF: is something forthcoming to indicate what are the needs of Wikipedia that Commons is failing to meet? The largest thing I see is that Commons cannot host non-free content. As I am sure you know, per foundation:Resolution:Licensing_policy, WMF expressly forbids Commons (unlike all other WMF wikis) from having an Exemption Doctrine Policy that would allow any non-free content. (foundation:Resolution:Licensing_policy dates, as a page, from 2007, but I'm pretty sure that at least with respect to Commons it documents a policy that is about three years older than that.) If WMF were to change that policy, I can't predict the outcome of a discussion within the Commons community about what policy we might adopt, but the first move here would have to come from WMF, not Commons. There is no point in the Commons community discussing a direction WMF has expressly forbidden: WMF controls the budget and the servers. If there is something else that the Wikipedias want from Commons and are not getting, then it really needs to be spelled out. - Jmabel ! talk 05:07, 6 June 2024 (UTC)

M.E. unterstützt Commons alle anderen Projekte (Wikidata, die Wiktionaries, Wikivoyages und die anderen WM-Projekte) sehr gut. Die Einbindungen von Medien in alle anderen Projekte funktioniert praktisch fehlerfrei. Eine bessere Unterstützung anderer Projekte ist jedoch auf zwei Wegen möglich: Die Medien in Commons könnten besser such- und findbar sein. Und das Hinzufügen von Medien könnte leichter sein. Zum ersten Punkt. Mit SDC (Strukturierte Daten auf Commons) wurde ein mächtiges Tool geschaffen, das aber drei Probleme hat: 1) Die Auswertung von SDC durch die Suche ist verbesserungsfähig (beispielsweise könnte eine Suchanfrage wie "Bilder in verlustfreien Formaten aus Bremen, die zwischen 1. Januar 2019 und 14. April 2023 aufgenommen wurden" möglich gemacht werden. Soweit das derzeit überhaupt geht, dann nur mit genauer Kenntnis einer IT-artigen Such-Syntax, aber weder als Freitext-Anfrage ("Prompt") noch durch (Datums-, Orts-, Sach-) Auswahlfelder in einem Suchformular. 2) Die Eingabe von SDC ist sehr beschwerlich. Vorhandene Tools wie ACDC u.ä. sind für einzelne Use Cases gemacht, für viele andere Use Cases gibt es keine angepassten Tools. 3) Nur ein verschwindend geringer Teil der Medien ist mit SDC aufbereitet. Dafür braucht es viel mehr Leute, die sich darum kümmern. Zum zweiten Punkt: Hier gibt es soziale und technische Probleme. Auf der sozialen Seite: Es ist immer wieder problematisch Fotos für Wikipedia zu machen. Eine Fertigstellung von v:de:Fotorechte DACH wäre schonmal was, was die WMF tun könnte. Zu den technischen Problemen: Hier hat sich in der letzten Zeit wirklich einiges positive bewirkt. Für Videos und webp werden nun Metadaten angezeigt, Uploads mit 5GiB sind nun möglich und funktionieren wegen eines stabileren Uploads nun auch. Mehr ist aber nötig: Beispielsweise der Upload von Dateien, die größer als 5GiB sind und verschiedene Versionen des Uploadwizards für Leute, die weniger als 10 Dateien, 10 bis 30000 Dateien oder 30000 bis 4 Millionen Dateien hochladen (wollen). C.Suthorn (@Life_is@no-pony.farm - p7.ee/p) (talk) 12:37, 6 June 2024 (UTC)

Structured data is not nearly as well-maintained and useful as file categories. Instead of distracting resources away from where they'd be efficient, please improve the search engine and related features...for example by making it easily possible on WMC itself to show files of combined categories in a wall of images and things like that. As for SDC, it doesn't need more people who waste their time adding this data that is already added elsewhere: instead it needs tools that automatically add or adjust the SDC using categories and maybe more. Prototyperspective (talk) 13:01, 6 June 2024 (UTC)

SDC was just another unicorn on which tons of funds were buried by the WMF because ???, producing an unfinished and broken product in the end. And for years it was misleadingly advertised as a "replacement" for categories, something it would never be, as it's not its nature (label vs. warehouse concepts).

WMF is so out of touch with reality that back in 2015, when Wikipedia received the Erasmus prize, WMF staff produced statements and interviews saying something to the likes of "now that we have the main stuff done, we need to focus in diversity" or whatever it was - referring to yet another WMF debacle, and one with dire consequences to Commons, Wikipedia Zero. It's so flabbergasting that we have to put up with this kind of declarations and statements when everyone that is here (onwiki) perfectly knows that so much of the basic is yet to be done, either in Commons and Wikipedia, and has to face that reality every day. We produce tons of DIY workarounds, precarious patches and whatever, in an attempt to fix what WMF should be doing, but it isn't - not because of "limited resources", but, above all, because the focus has been elsewhere, and certainly not in the community that actually produces, curates and maintains the content that actually attracts most the funding WMF gets.

And now, per the above declarations, we are reduced to hoarders pilling up stuff, whose main occupation seems to be detect copyvios and pill up more and more stuff with the excuse it's "for Wikipedia". Dear @SDeckelmann-WMF, as chief product for the WMF, I heartily invite you to know more about this platform and our community, so to dispel these many incorrect notions you and so many others at WMF seem to have built over the years. Darwin ^Ahoy! 10:08, 7 June 2024 (UTC)

100% agree. I think it's time to think about what can be done that is effective instead of kindly asking WMF or pointing out issues internally.

Internal open letters don't work and stopping donating to them just means their misleading donation banners are shown for an extra second. If such open letters are of any use it's if @Spinster: and/or sb else contact relevant media to report on it as well as other widely-supported communications about the neglection of improving technical features such as this and this. Prototyperspective (talk) 11:48, 7 June 2024 (UTC)

If it is not on the plan for the next years hat will happen with the changes to the UploadWizard? Will the development stop before being finished like it was the case with the structured data project that is still missing core features? On the "dissemination of knowledge" part: Unlike Wikipedia our target audience are not causal end users. Wikimedia Commons is a platform from volunteers, scientists, journalists and public institutions for volunteers, scientists, journalists and public institutions. GPSLeo (talk) 13:31, 6 June 2024 (UTC)

Adding my 2cents - I feel the claim that the focus of the Commons community is "collecting of free content, rather than its dissemination" is not correct.

We might have different asks and ideas about how we can help disseminate the knowledge contained in the images on Wikimedia Commons. But our main focus is not only on collecting them. We strive to collect them in a way that they are easily findable and useable. Many of us spend hours upon hours on improving descriptions, adding SDCs and categories not only to our own images, but also to images uploaded by others. Negating this work is not a good basis for moving forward.

Many of us are acitve in several projects and make sure images we upload are used in other Wikiprojects. For those images for which we don't see immediate use, we work to add all relevant information so they can be used by other projects.

Furthermore dissemination of the knowledge contained in our images is not only possible via Wikipedia or other Wikimedia projects. I know for example a lot of literary intstitutions in Germany that use images hosted on commons regularily. I taught some of them how to do this, so they also do the correct attribution.

Generally a lot of non-profit orgs use our images. They are an important part of "our audiences". Among the orgs that use our images are the Goethe Insititutes across the globe, who have collaborated with the Wikiworld in various ways. I know that many GLAM projects also work bi-directional - we collect images from these institutions, but we also teach them how to use images from Commons for their usecases. I know it is hard to track this kind of re-use and knowledge dissemination. But just because we don't have hard numbers it should not be discounted. Kritzolina (talk) 15:55, 6 June 2024 (UTC)

@SDeckelmann-WMF: I think that Kritzolina raises an important point: Commons media are used outside of Wikimedia projects in worthwile ways that are hard to measure. Unlike for Wikipedia articles, we can't just count page access stats (which are, of course, also far from the whole picture for Wikipedia - given data uses like Google's -, but even less so for Commons). Like: A GLAM institution that has contributed to Commons might point an academic user who wants to print an image in their book to Commons for downloading, and then the image appears in that print publication, but we can't really measure the impact of the publication (that is, we will most likely not even know of the publication). Or other media that use images from Commons, often not fully attributed, maybe not even to Commons (none of the licenses we use requires that - CC-BY-SA for example requires that you mention the author and the license, but not where you got the picture). In general, I agree with the sentiment that, besides the function of Commons as the central media repository for Wikimedia projects that just has to work reliably, our target audience are GLAMS (not only as contributors, but also as users), other non-profits, and also the media in general; maybe somewhat less so the casual end-user, though it's of course good to have an attractive interface for them, too. Gestumblindi (talk) 13:49, 9 June 2024 (UTC)

Yes. I received an award from DoCoMoMo WeWa for my photos of mid-20th-century architecture in Seattle, which has been very useful to people preparing landmark applications (not so much for the specific building being landmarked, as for having a resource to find images of comparable buildings, including buildings that have now been demolished). I'm sure I'm nothing like alone in having images I took myself and uploaded to Commons used that way. - Jmabel ! talk 16:01, 9 June 2024 (UTC)

I think those examples of uses are bit misleading...WMC is also used in many very or fundamentally different ways.

• A few examples for the former: using it to find pictures for a study that is planned to be released under CCBY or for a CCBY youtube video, making CCBY items available/findable in search engines like Google Images, CC-search portal and DuckDuckGo images, finding free media for low-cost educational media productions, and using the category system for image or concept classification systems such as AI systems.
• Examples for the latter is that you can browse around the site and find educational/useful contents directly – some say an image is worth a thousand words and the Wikipedia approach to have lots of text with a few images is just one approach that can be complemented with other ones. For example, if there are some charts buried in some report somewhere and a diagram in some particular study that doesn't mean people find it when they search for such an image here or elsewhere. (imo there's enough photos now and their importance is overstated compared to illustrations etc)

Maybe there should be a list of ways WMC can / is useful. I think the semantic categories are key there along with the open collaborative approach where things are organized into one integrative structured system as the logo suggests.

However, I think the walls of text should distract from the key problems and now may repel users to get involved in this discussion (maybe subheaders could help): I don't people should need to argue about why WMC is important and useful to WMF – if WMF people don't see the usefulness of the site themselves then they are simply inappropriate for their job and should probably do something else. The relative priority of WMC is a different subject so I think that's the issue here and it would be raised if WMC content and pages were a) better indexed in search engines resulting in greater pageviews, b) better integrated into Wikipedia (rather than a link nobody clicks at below large References sections, c) was designed so that good-quality content is easily found and can be viewed well such as greater use of wall of images. Raising the priority partly requires WMF actions. There also is d) key technical and design features being broken or insufficient. Prototyperspective (talk) 17:31, 9 June 2024 (UTC)

I don't think pageviews are that important, that's actually not a metric we should be too concerned with, IMHO. Gestumblindi (talk) 17:53, 9 June 2024 (UTC)

The site is also useful otherwise but obviously files and categories are of no use if nobody uses them and whether they use/find them can be inferred from pageviews. However, I named several ways in which files can be used where each use would only have one pageview/fileview so obviously it can't be the only way one evaluates the usefulness of the site. Prototyperspective (talk) 18:10, 9 June 2024 (UTC)

While I see what Kritzolina is describing... Selena is partly right. The Commons community is VERY inward looking and tends to focus a lot on collecting material, and is bad at focusing on the sharing part. Anyone external person who takes one look at a File page will tell you that they can't make heads or tails from it. 2/3's of a popular image is 'editor information' not information that is critical to outsiders. This all grew organically over years and years, but turning it around is very difficult. However the WMF should recognise that they also don't allow for much space to turn this around. No-one is seriously doing interface design prototypes, the (Google) SEO of file pages has notoriously been broken since forever, metadata collection can vastly be improved, Special:Upload cannot handle uploads larger than 100MB etc etc. Additionally, all the complex technology work resides in Commons and Wikipedia just gets the free benefits of that. If we were to shut down Commons, the Foundation would have to recognise they would have the EXACT same technology changes in Wikipedia for instance. So it doesn't matter which project you have to fix it in, both benefit and THAT is why I think it is good to focus on Commons. Diversification creates growth for all because we share the same technology. —TheDJ (talk • contribs) 12:07, 11 June 2024 (UTC)

After countless hours and immense dedication curating, categorizing, cleaning up, not only getting Commons content more easily available to our end users, but actually reusing it in Wikidata, Wikipedia and Wikisource, I lived to see all that described as nothing but hoarder activity by the WMF product chief. I wonder if this person is familiar with the existence and purpose of media libraries, to make such extravagant assumptions and remarks. Also, I'm quite sick of hearing WMF constant whining about "our limited resources", which apparently are not so limited when it comes to fulfill such lavish caprices as the branding and strategy debacles, and the countless unproductive (and worst) projects, organizations and initiatives WMF keeps funding with little to no scrutiny. And it's not like Wikipedia has been a scintillating example of investment, either. Many parts of the platform seem to still be in the 1990s, and for more than 20 years it has failed to deal with such an egregious security abd privacy breach as allowing people to edit revealing their IP addresses. There has been more than enough tongue service about all that, but year after year what I see is disinvestment on the platform and the people that work here every day. I still have not lost hope that a change in attitude will come, but from @SDeckelmann-WMF words, it's not on the near future. Darwin ^Ahoy! 03:22, 7 June 2024 (UTC)

I partly disagree with those last two sentences. As a volunteer developer, I actually see a LOT of focus at the moment at dealing with lots of fundamental problems of Wikipedia and MediaWiki. A lot of things are finally coming together. Especially major changes in the web, that simply took 10 years before we could make use of them, but also a lot of rework of the core platform is happening. The skin system has matured, css variables, ES6, Vue and Codex are making code development a lot easier.. It's just not that visible to most people as a lot is just replacing the complete inner guts of a 25 year old product. And it is SLOW, very slow. But when you look at dark mode right now... That is ONLY possible because some of those 10+ year projects are finally coming together and it is much easier to do than many of the previous changes we have seen. As stated below by Rhododendrites, I think a lot of the problems that Commons sees, are due to not wanting change (esp not to the File page), and then disagreeing about what big vision we want to achieve at the same time, all of which requires a vision and significant changes. —TheDJ (talk • contribs) 12:20, 11 June 2024 (UTC)

• Most if not nearly all of the most urgent or impactful changes do not depend on external changes.
• Vue etc is still not used widely, actually I don't know where it's used at all.
• The dark mode is horrible. While I use dark mode on many websites I would never use it on WP – it's black mode not a dark grey like most other websites have as darkmode and which makes it readable. Don't know if there is an issue about this. At least the dark mode of the app is reasonable.
• It is slow because there is no effort to make it a) simpler, b) get more volunteer devs involved c) get involved volunteer devs to solve more issues such as via issue-bounties once has solved the 20th issue.
• False claims about issues with contributor stances to changes to Commons – at the recent technical needs survey users widely agreed on several impactful overdue changes, and these are just some of the highest-priority ones and didn't even include e.g. fixing the broken video2commons or the showing of subtitles for videos that have them.

Prototyperspective (talk) 14:56, 12 June 2024 (UTC)

Two things:
First, The thread about categories vs. SDC above may be a good example of why this community is difficult to serve: we want investment, but can't agree on what we want. We want better infrastructure and user experience, but want to hang on to an ancient [in internet time] hierarchical tag-based category system filled with inconsistencies that's difficult for newbies and for computers. Rather than do everything possible to support a system that would just let us find the intersection of a location and format, we want the WMF to somehow build a system that works with our thousands of individually created "Panoramas of France"/"Panoramics of Japan"/"Panoramics in Boston"/"Panoramic photos of Sweden"/"Mexico panoramas" categories. We want improvements, but don't want changes to the way we've always done things. If all we want are minor improvements to existing systems, we can say that, but we also say we want big changes (and then disagree about which).
Second, I mainly just want to ping @SDeckelmann-WMF: for some additional information about their second point above regarding collection vs. dissemination. From your use of those words, it sounds like you may be using them in a loaded way, related to language in one or more planning/mission/strategy documents. They could use some spelling out, as well as the apparent valence behind them. For example, in what ways does Wikipedia not merely collect articles but also disseminate them whereas Commons merely collects but does not disseminate media? — Rhododendrites ^talk |  17:34, 7 June 2024 (UTC)

@Rhododendrites I never said a system like SDC would not be useful. In fact, I recall much of the support it got in the beginning, including mine, was based precisely in the false information that somehow it would "magically" produce categories without the need to constantly creating them, even more in an structured way. Then it was developed as a kind of an additional hashtag system, which was useful but definitely not a priority. Then even that was not finished and was left broken by WMF after they got tired of the new toy. Anyway, one just needs to look at the recent threads about the last botched release of the UW, apparently without following even basic testing patterns, to see the poor state of affairs we are now in thanks to all divestment and negligence of WMF regarding Commons. Darwin ^Ahoy! 19:55, 7 June 2024 (UTC)

@SDeckelmann-WMF Hi, As was said above, I don't understand your distinction between collection and dissemination. Publishing anything on Commons is dissemination by definition. It makes the content available either by using only free format, free of advertising, better referencing, not behind a paywall, etc. Even if Commons objective is not restricted to use by other Wikimedia projects, Wikipedia is still the main reason content is uploaded to Commons. I believe that multimedia content is essential to an encyclopedia, and should not be considered, as it seems the case, a second-rate content. As the adage says, “A picture is worth a thousand word”, multimedia content should be one the top priority of the WMF. Yann (talk) 21:18, 8 June 2024 (UTC)

What cannot be found and what people find too difficult to use, does not facilitate dissemination. You can start a museum and fill it with stuff, but if the entrance fee is high and you have to pass a math quiz to take a picture of something in the collection of the museum, you are not disseminating. A lot of what people do here is focused on collecting. But we are bad at telling the story of why, and getting lots of people in the door to access that collection. The Commons vision seems to be "collect, curate and they will come", but I think Commons could benefit from looking at Commons a bit more as a museum that needs to attract visitors by being an attractive destination, and a bit less as an archive that needs to preserve as much as possible. —TheDJ (talk • contribs) 12:34, 11 June 2024 (UTC)

Some Wikipedias explicitly refer to Commons for readers who seek more illustrations. So for them, it's assumed that Commons is there (which may not be easily accessible for the average reader nor easy to navigate, but that is another issue -- one for Wikipedias, one for Commons, both for ..).

Another issue is that in some Wikipedias they don't really like non-writers to illustrate articles (understandable, if that consists of replacing reasonable illustrations with less great, but own ones). Enhancing999 (talk) 12:44, 11 June 2024 (UTC)

@TheDJ: Sorry, I didn't see your answer earlier. Commons Google ranking is much poorer than Wikipedia, but I don't see what Commons volunteers can do about it. I usually upload content with the intent to link from or add it in at least another project. Many of the files I upload from elsewhere are not easily available for various reasons, so uploading them to Commons increases their dissemination. Yann (talk) 15:08, 20 June 2024 (UTC)

But we are bad at telling the story of why Disagree. File titles, descriptions, and especially categories are about the why. Once again, the issue there is mostly a) technical in the sense of a useful modern UI and b) proper search engine indexing. Both can readily be addressed by WMF to some extent now (and need to) such as investigating why so few Commons files show up in image search engines and why sufficiently-populated commons categories rarely show up for relevant search engine terms.

There is also is c) Wikipedia users should make it an established practice to integrate links to Commons categories more visibly than somewhere deep down in the external links that nearly nobody clicks at and is buried between the large References section and d) post educational/high-quality content or pages from here on more-frequented sites / social media (example example) including links to here and asking communities there to upload here. Prototyperspective (talk) 16:45, 21 June 2024 (UTC)

I think that the idea that "we can't agree on that" is false. We can't agree on what because there's no way to go forward. If the commitment is ZERO hours, then it's quite evident that we are not going to have any strategical view or agreement on what to do. Theklan (talk) 14:13, 10 June 2024 (UTC)

Replying here, but responding to a few comments at the same time:

Thanks again for your engagement and your replies to this discussion. I am grateful for the examples of reuse of material from Commons and also the research paper: understanding these use cases and thinking through with you all is important to the ongoing prioritization of product and engineering work to be done. For impact metrics, we have been collaborating on wiki with individuals, multiple affiliates and partners to produce Commons Impact Metrics.

Regarding the point of “collection vs dissemination” mentioned by User:Rhododendrites but also others: my observations were based on seeing that Commons seems to embrace an approach to free media that goes beyond the goal of serving only Wikimedia projects, and this has been confirmed by many comments in this discussion. However, Commons has little direct page traffic and is not well adapted to re-use on the Wikimedia project which does have significant worldwide page traffic (Wikipedia). Community engagement on topics related to integration with Wikipedia is relatively low relative to the opportunity from my perspective. Are there pages, user groups or forums of Commons community members who are focused on integration with Wikipedia?

Regarding the policy linked from the board in 2007, thanks User:Jmabel for drawing this connection, and I am taking some time to learn more about this policy. SDeckelmann-WMF (talk) 16:41, 28 June 2024 (UTC)

@SDeckelmann-WMF: Just speaking for myself, this didn't really make things clearer, and it's hard not to get the impression that you're saying the foundation doesn't care about volunteers or non-Wikipedia projects except insofar as those volunteers/projects can be metrics-optimized according to what the Foundation (and not the communities it serves) considers most important.
Commons has never been, since its very conception, solely to serve Wikipedia projects. It has always been intended as a vast repository of free media, and indeed basically any google image search for CC-licensed photos returns a bunch of Commons photos. Tons of photos I've uploaded which were never used in Wikipedia articles pop up all around the internet and in various publications because while Commons isn't Wikipedia, it seems to have quite a bit of reach. Since the beginning, Commons has also struggled with software. MediaWiki is a little awkward for a massive image hosting site, and absent allocated resources from the Foundation we've adapted with a large collection of user-created bots and scripts which regularly break as volunteer developers burn out. Over at POTY we have an undocumented set of scripts that's been used for many years, which only a handful of people know how to use, and all of those people burned out. We'll see if we have a POTY this year. SDC is an exiciting bit of investment, but it's been partially done for years, with insufficient community engagement (granted, the loudest members of the community on SDC tend to be the ones who don't really understand it, but still -- it would be easier to sell if it were more fleshed out). Final thing: the communities aren't actually separate. Nearly all of us in this discussion are active on Wikipedias, too, and active on Commons in ways that directly help Wikipedia. I became active on Commons when I decided I wanted to take photos for Wikipedia. I still do that, and I organize photos to make them more accessible -- for Wikipedians or for absolutely anyone. I also participate at FPC and in some other processes, trying to help other people improve their photos which, you know, improves Wikipedia. They're not actually very separate, but we do have software needs here that aren't served if you allocate resources to prioritize just the part of the English Wikipedia that's located at en.wikipedia.org. — Rhododendrites ^talk |  17:16, 28 June 2024 (UTC)

@Rhododendrites Thanks for responding. I appreciate you continuing to engage in what I perceive is a pretty frustrating set of exchanges. I really care about supporting the contributors that have made free knowledge globally accessible, and that comes from believing in open culture, the value of the internet to the world, and committing most of my life to supporting the people who work on open source and open culture projects. The questions I'm asking and issues I am raising are coming from a place of wanting to know more, and wanting to understand how best to serve the communities. I hear you when you say that the communities aren't actually separate. I'm excited to work on integrating all these conversations more, and happy to collaborate with anyone interested in doing so.

The scope of the software work we could start on Commons easily would consume the entire staff if it isn't well defined, scoped well, and the necessary workflows are documented. We face similar challenges in other projects. The reason I go back to the metrics -- and perhaps this might be better explained if I walked a few people through work that's being done, and the challenges we face in measuring impact -- is that metrics are a way for us to asynchronously, with data, look at the impact of the work we're all doing. That doesn't negate qualitative measures of success. It's just a tool, and a very powerful one if we're able to come together and agree on what we're trying to accomplish in quantitative ways. It's one of the best proven ways to get a software development teams to deliver more value to all users of Wikimedia projects, which I am here to make happen.

I'd love to know more about the POTY tools and anything about FPC that might need support. I hear you on SDC, and am open to improving this. SDeckelmann-WMF (talk) 17:56, 28 June 2024 (UTC)

@SDeckelmann-WMF: Two things. (1) I continue to ask (and I don't think you've answered, but please point me the right direction if you have): is something forthcoming to indicate what are the needs of Wikipedia that Commons is failing to meet? (2) I think one of the most important uses of Commons -- and one for which it would be very difficult to get a metric -- is that a lot of poeple who build web content (and even who publish books) use Commons heavily as a place to find PD and free-licensed images. However, because Commons itself holds no rights to this content, there is no reason for them to credit Commons when they do so, and so no way to track how much Commons is used this way. For example, I know that literally hundreds (quite likely thousands) of images I've uploaded to Commons (and only to Commons) have been reused with credit to me, but typically not to Commons, because there is no reason for the reuser to do so. I have no idea how we would be able to measure this impact, but from my own experience I am sure it is not small. - Jmabel ! talk 19:42, 28 June 2024 (UTC)

I find the distinction between "collection" and "dissemination" quite strange, as it is a WMF's decision to make impossible to share directly files in Commons externally (

Tracked in Phabricator
Task T364479

and having statistics about what we share. The lack of knowledge of what the dissemination looks like has been designed by the WMF, so it feels weird to use that lack of knowledge as an excuse not to invest on improving one of the most important sites we have. Theklan (talk) 13:56, 10 June 2024 (UTC)

@SDeckelmann-WMF In my personal opinion, this view is too much all or nothing. Commons and its core technology has lots of overlap with the other Wikis. Focusing on Commons specific (or rather things that are more VISIBLE in Commons) can bring value to Commons, Wikipedia and MediaWiki at the same time. A lot of the work that has gone into Commons over the last years has been with specific grants on specific big vision ideas, but it ignored all the other pre-existing problems. This is bad product ownership (because no one was owning it) and a recipe for disaster. A more evolutionary and small scale approach is warranted and additionally much less likely to fail. This is why I have been advocating for a Commons product owner. Someone to identify problems and see if fixing those can bring value to all (not just Commons), slowly mature and improve the state of play and only from that point built out on newer and more ambitious projects. Yet for every year we delay this, we fall further behind and burn more credentials among the community and even the developers who now have to wrestle to keep thing alive. There is also a difference between doing some exploration and vision building and full commitment to overhauling Commons. —TheDJ (talk • contribs) 12:46, 11 June 2024 (UTC)

Just because you put a text in supposedly nice words doesn't make it nice. Even from the WMF, I have not often read a statement that was so disrespectful and derogatory towards the community. "We don't care about your worries, you don't fit into what we're planning" can hardly be put more clearly. Marcus Cyron (talk) 14:56, 20 June 2024 (UTC)

1. "the urgent concerns and challenges Wikipedias face (e.g. stable but stagnant editor numbers, declining adminship..."

   i have a solution. how about making adminship non-permanent and requiring all admins to explicitly identify their names and faces? abusive admins are the most frustrating shit on wikimedia projects. they are abusive because they hide and have no consequences wielding their tools, unlike real world law enforcement or technicians.

2. "WMF has an obligation to invest resources in a way that furthers not just knowledge collection, but knowledge dissemination. To what extent is the Wikimedia Commons community invested in the dissemination of knowledge via images on Wikipedia articles? To date, my observation is that the primary focus of the Commons community is the collecting of free content, rather than its dissemination."

   bruh... sorry to say but this is really ignorant, out of touch, or... stupid. do you know that "uploading to commons", this action itself, is one of the most effective ways of "dissemination" of those files through the internet? commons dont need "wikpedia articles" to disseminate the files. even if you have no clue at all about the myriad of ways commons files can be found and reused, do you forget wikisources (which rely on pdfs/djvus commons host) or wiktionaries (which rely on pronuncation audio files commons host)?

RZuo (talk) 20:43, 25 June 2024 (UTC)

I am agreeing. And there is yet another thing. A decade back or almost two I thought "why we need another 100 of photos of Heidelberg Castle if we already have several hundreds". It took me some time to realize it. Commons also is a resource for science/scientists. It might be of interest to habe pictures of each part of the castle adding to a long time row of pictures showing the castle as time goes by in a environment chanchin world of pollution and climate change what has its effects on the building. There might be users who need this footage in 2070 who even are not born yet. We need regularily pictures of each and every building featured somewhere in the Wikipedia, at least four times a year from every accessable direction, better each month. We also need more photos of vegetation in national parks and the like. It is fine that through the White House governmental website we habe images of the U.S. president frequently but we need muh more, plenty much more photography depicting other people, actors, singers, politicians, sports people, on and on. It is not that Commons has too much images, it needs much much more then we have. Matthiasb (talk) 18:53, 2 September 2024 (UTC)

exactly. i think 1 photo per month (=12 per year) for a landmark building is not excessive at all. 1 day photo and 1 night photo per year for any building is not excessive at all either.

for places like national parks which can often be several hundred km2 big, 100 photos per month is probably not even enough.

(for documentation purposes.) RZuo (talk) 04:39, 3 September 2024 (UTC)

@SDeckelmann-WMF Like others above, I am puzzled by the question whether we do this for dissemination at all. Of course we do. As someone who started editing Wikipedia in 2003, I can say I would love to see - as implied in our statement - more and (especially) larger illustrations in Wikipedia, more illustrations per article, display of all related media per Wikipedia article. That would make Wikipedia so much more interesting and attractive and would make it fit today's expectation of knowledge-seeking audiences. But Wikipedia has deliberate (and huge) limitations. As explicitly said in the statement: Wikimedia Commons covers so much more human knowledge than Wikipedia's scope allows. I hope that as organizations supporting the Wikimedia movement, we will actively work so that non-Wikipedia but culturally significant media files are actively exposed to as many people as on Wikipedia.

Yes, we collect on Wikimedia Commons, because more coverage of topics matters for quality there (documenting all national monuments in a country, documenting all species, documenting immaterial heritage... and covering these over time has extra value). But we wouldn't dedicate all this effort just for the sake of "stamp collecting". It's the desire of knowledge dissemination to large audiences why many people from this community have been dedicating resources and time to platforms like Wikidocumentaries, Wikiflix, Open Art Browser, it's why so many of our communities invest in campaigns like Wiki Loves Monuments and Wiki Loves Africa. We create, collect, curate this material to have the exposure of Wikipedia. As Wikipedia's scope is narrow (encyclopedic content only; you can only create articles with reliable and mostly written sources) and its contributor communities are generally often not very favorable to include larger illustrations and include media-centric knowledge, we either need to change the Wikipedia scope (which, as a Wikipedian myself, am not hopeful about) or need to invest in the other - indeed vibrant - platforms we have that offer alternative ways of disseminating knowledge that matters to the world, in alignment with the Wikimedia 2030 strategic direction, which explicitly also states that Wikipedia has its limitations. Spinster (talk) 07:58, 26 June 2024 (UTC)

@Spinster: "display of all related media": certainly not all For example, no one reading the article on the Space Needle wants to see the hundreds of rather interchangeble shots of it that we have on Commons (probably over 100 just showing the vew of it from Kerry Park). Most of those are potentially useful, some of them very good, but no ordinary encyclopedia reader wants to see the entire lot.

Yes, I'd like to see more of our images readily available to Wikipedia readers. en-wiki is generally negative about even well-curated picture galleries on articles; that would easily be fixed on en-wiki, but there is nothing we can really do about it from the Commons side. - Jmabel ! talk 20:00, 26 June 2024 (UTC)

I firmly disagree. I always find it presumptuous if we as Wikimedians decide what our readers should be interested in and what not. As someone deeply interested in architecture (like myself), you may very well be interested in seeing all photos of a building. I find myself very often browsing through entire (large) categories on Wikimedia Commons. The key is to provide a pre-selection (for instance based on image quality and size, or recency), and then allow to sort and filter by date and other criteria. With structured data, we have all the ingredients to make that technically possible. Spinster (talk) 12:12, 27 June 2024 (UTC)

@Kritzolina, Gestumblindi, Jmabel, and SDeckelmann-WMF: and others: Maybe we can advance our conversation when we look into some scientific research about the different functions and use of Wikimedia Commons alone and in relation to Wikipedia: I found Unpacking Stitching between Wikipedia and Wikimedia Commons: Barriers to Cross-Platform Collaboration to be a very interesting read. —DerHexer (Talk) 16:21, 20 June 2024 (UTC) PS: It's astonishing that we don't know much about Commons and Wikipedia interaction at all. One of the other few scientific papers about that which come to my mind, is A large scale study of reader interactions with images on Wikipedia. PPS: Finding external uses of Commons photos is a mess; I once did that for a photo project that I was driving and it was extremely time-consuming.

The first study has many interesting parts. I found this useful: For example, there was no formal way for Wikipedia editors and Commons curators to discuss the imagery needs of Wikipedia articles […] Louie continued to say that Wikipedia editors did not have a way to inform Commons curators about what articles were missing illustrations […] The lack of mutual awareness of project needs inhibits the formation of broader networks. The projects would benefit a lot from a well-known efficient system for image requests, if anybody is interested in helping set it up or has some ideas please comment, I suggested "Better ways of requesting missing images and increasing the use / solving of such requests" and listed several major missing science-related illustrations here which themselves are such requests (more input on this subject there).

Another important aspect covered but this study but is multilingualism – getting back to scientific research and development, machine-translated file titles and category titles+descriptions (correctable by editors) would be very helpful and would require substantial technical effort to implement. Sadly the study didn't even mention it albeit I doubt it would make a difference.

In general, I think the study and your comment are again mainly about 'Wikimedia Commons in relation to Wikipedia' but WMC is (maybe rather could be when considering the generally low pageviews) also useful on its own separately, e.g. for reasons and in ways that I'm going to start listing here: Commons:Why Wikimedia Commons is useful (it could be moved later, will include barriers or ways to improve each, and it would be good if more editors contributed to it). Prototyperspective (talk) 17:23, 21 June 2024 (UTC)

Thanks! I will have a look. SDeckelmann-WMF (talk) 16:35, 28 June 2024 (UTC)

I'm sure WMF board has much complete information than me about why priorizing spending in Wikipedia over Commons in the short term, and there can be very good reasons for that. Probably, many Commons improvements aren't so urgent and can perfectly wait for 1 or 2 years. It isn't fair to say that nothing good has been recently done with Commons: 4 years ago, there weren't proper media backups; now, they are. But I think that it makes no sense to talk about Wikipedia and Commons communities as something different, much less opposing. Many of us contribute to multiple Wikimedia projects at the same time. There are no Wikipedia and Commons communities: there is Wikimedia community. We are building the same sum of human knowledge across different projects. If one user focuses only in Commons, it's like when one focuses in Wikipedia articles about one topic: he/she is contributing to Wikipedia in any case. I think that all Wikimedia projects should be viewed as if it were one final product: a Wikipedia article isn't complete without its Commons images; an article about a public domain book is supplemented by its full text in Wikisource. They are not different things: they're the same product. Back in 2000, we couldn't even dream about having what we have now. We have to take good care of it, and continue improving it, both as a community of editors and as a Foundation. MGeog2022 (talk) 17:43, 14 July 2024 (UTC)

I think User:MGeog2022 is right, we sometimes demand more and don't acknowledge what has been improved (especially in these heated, divided times. I would say, as in many cases, (frequent) communication is king, here between WMF and community. I assume, it would help a lot to know, why some priorities are like they are, why some requested features take a bit longer, and if such implementation is easy to execute :). I am eager to see the textured meshes feature for Commons to come, and I think it would help a lot to geht more exchange in this issue, also because there is more interest (and will become more relevant in the future) --PantheraLeo1359531 😺 (talk) 06:59, 15 July 2024 (UTC)

I'd have much less of or even no problem of WMF not naming Wikimedia Commons among its priorities or similar if they did more to get more volunteer developers onboard (for which there many options) who then could help address issues and e.g. help develop the Commons app (such as adding a Discover Feed / home page). They don't necessarily need to mention it explicitly (and that's not necessarily the same as "prioritizing") or develop things themselves.
But hiding categories on mobile seems like a simple common sense things that should be undone and could be done in fairly little time, the UI for these links could be improved later which may need more resources but at least they should show up on mobile as well, not just on desktop and in the Commons app (which is still used little, I think partly because key features like the above make it engaging are missing like Twitter would not be used without its Feed page). Moreover, considering the large amounts of WMF financial resources and how they spend (I'd say to a large extend waste) money, I don't think we should be this understanding and satisfied with the things WMF did, which in MGeog2022's comment are somewhat conflated with the broader Wikimedia community achievements. Commenting mainly to add this page which supports the case of the open letter and earlier remarks:

We will innovate in different content formats, develop new software functionalities for Wikimedia projects […] Build the necessary technology to make free knowledge content accessible in various formats. Support more diverse modes of consumption and contribution to our projects (e.g. text, audio, visual, video, geospatial, etc.). […] there is a need to facilitate the reuse of our content on platforms beyond Wikimedia. […] at https://meta.wikimedia.org/wiki/Movement_Strategy/Recommendations/Innovate_in_Free_Knowledge Prototyperspective (talk) 09:46, 15 July 2024 (UTC)

Yes, I don't know about WMF finance management, but I've read many complaints about it (and not only in Commons talk pages). Some external projects supported by WMF, such as Kiwix, are great and much needed; probably some others aren't. With my comment, I meant to say that, while there may be reasons for, with a limited technical budget (without going into whether or not this should be the case), priorizing spending in Wikipedia over Commons in the short term, this is no excuse to affirm that contributors to Commons are a different, less important community because they don't directly contribute to Wikipedia (which is false in many cases, but even when it's true, contributing to Commons, Wikisource or Wikibooks, also is contributing to Wikimedia, a kind of super-project that includes Wikipedia, so it's like contributing to Wikipedia: you are not a less important editor if you edit Wikipedia articles with low pageviews count, we don't work that way). That is, it seems that someone who belongs to WMF board wasn't able to provide good arguments for the decisions they took, and it shouldn't be so difficult, if she wants to defend her viewpoint (just saying that increasing the number of active Wikipedia editors is the best way to also increase the number of Commons contributors, for example, would do the job). MGeog2022 (talk) 12:17, 15 July 2024 (UTC)

I think that we can make still huge progress, as from the "stone age" of wikis until today. And as long we can make improvements here and there (cooperations with institutions or private people, but also technical ones), it's worth to do so. There are so many objects that are (almost) not covered, and we may need new infrastructure and functions to incorporate them (for example point clouds generated by LIDAR systems). It would be very unfortunate, if we lose chances of getting new material by people, only because one function is still missing --PantheraLeo1359531 😺 (talk) 15:44, 15 July 2024 (UTC)

I see the focus on the use on Wikipedia as a problem. As an example: Many years I was rather active writing articles for Wikinews. Of course it is possible to write a Wikinews article about a sport event then I am interested on images picturing the event. When I ask "can I have picture material on this or this" I get answers like perhaps, but upload is bad at the moment, maybe in three weeks. I won't need this pics in three weeks anymore because I cannot wait three weeks on images before I publish a Wikinews article. Worse: I might refrain from writing the article anyways.

Another project which has to deal with the problem might be Wikisource where we need in advance of transcribing PDFs or DJVUs depicturing each and every page of a book. I total for one book this are hundreds of files and in sum can be dozens oder hundreds of megabyte needed to upload. I know about users being on the brink of leaving Wikisource because of they do not manage to get those images uploaded in time oder in a reasonable time. Time in some times is the same as in reasonable effort. With Wikisource the Commons are not necessary for distributing but for creating the content later distributen. At least primarily. For some time Commons:Scope needs to be reviewed in the need of take seriousely the needs of other projects than WP. Though this needs are mentioned within Commons:Scope some stakeholders – including some admins and WMF persons – do act differently. Matthiasb (talk) 18:16, 2 September 2024 (UTC)

@DerHexer: That finding external uses of Commons is so hard is caused of incomplete/improper licensing, in many if not most cases. But what about hotlinked images? Matthiasb (talk) 19:34, 9 September 2024 (UTC)

@Matthiasb: are you saying that reusers give incomplete/improper licensing, or that Commons does? - Jmabel ! talk 05:57, 10 September 2024 (UTC)

Reusers do, all the time. Matthiasb (talk) 12:21, 10 September 2024 (UTC)

@Matthiasb: For pages like w:de:Wikipedia:Landtagsprojekt/Nutzung I used Google Reverse Image search before it was replaced with the malfunctioning Google Lense. I know that TinEye might do the same, but I haven't had time to use it. All of these should work even with bad or no attribution. I'm aware that Raymond also collects re-usages pretty thoroughly. Maybe he can share his way how to do it. Best, —DerHexer (Talk) 16:49, 16 September 2024 (UTC)

@Matthiasb, @DerHexer I am using Template:Published on the file talk pages to collect re-usages in non-WMF projects. It is easy for me to google for my (worldwide?) unique realname. Some examples:

File talk:Garzweiler Tagebau-1230.jpg,

File talk:Hart aber fair - 2019-09-23-9057.jpg,

File talk:United Airlines - N851UA -Airbus A319 - San Francisco International Airport-0383.jpg.

In most cases the attribution is OK(ish). What I am not activly do is searching for files with no attribution due to lack of time. I know that a low of (my) images are used with no attribution. For whatever reasons. Raymond (talk) 17:19, 16 September 2024 (UTC)

But that is also a question how to focus on your contribution to WP. I don't know you as a regular author/editor so caring on the own photo work possibly seemsfeasable to me. My commons work is/was mostly transferring EN WP files and uploading NASA EO satellite images which I don't care on, speaking in the relation to correct licensing. I even could not do it because of the many other activities I try to pursue. Matthiasb (talk) 21:19, 19 September 2024 (UTC)