Commons talk:Multimedia Features/Vision 2016

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search
Multimedia Vision 2016.pdf

Please share your feedback about the proposed Multimedia Vision 2016, as presented in this video -- or annotated slides.

This hypothetical scenario was prepared with guidance from community members and is intended for discussion purposes, to help visualize and evaluate possible improvements to our user experience.

This particular exploration proposes ways to integrate Wikipedia more closely with Commons, so that Wikipedians could contribute or curate more easily to our free media repository, wherever they are.

Please add your comments in the most relevant sections below. Thanks for your insights!

What do you like about this Multimedia Vision?[edit]

  • More obvious way to see a larger image without "leaving" the article page. - Jmabel ! talk 00:50, 10 January 2014 (UTC)Reply[reply]
  • Very nice! I liked the join-a-contest idea, easy flagging for problems, wikimaps, and notifications back and forth in particular. -- Phoebe (talk) 16:40, 10 January 2014 (UTC)Reply[reply]
  • Wikimaps are fantastic! I liked also the notifications back, that encourages newly registered users. --AlessioMela (talk) 10:11, 13 January 2014 (UTC)Reply[reply]
  • I like it all :-) --Aubrey (talk) 08:34, 20 January 2014 (UTC)Reply[reply]

What would you change in this Multimedia Vision?[edit]

  • As an uploader of tens of thousands of images, I don't want to be sucked into a bunch of "social" (non-task-oriented) discussion of my images, so I'm a bit concerned about a tool that encourages that. We have always discouraged non-task oriented discussion on Wikipedia and associated projects. If a social aspect is added, at the very least we need to do something to set expectations that the uploader/author will not necessarily be acknowledging every comment about his/her work. And what is to stop this from turning into an unmoderated space for flaming? - Jmabel ! talk 00:49, 10 January 2014 (UTC)Reply[reply]
    • I agree, I feel that it's a mix of Facebook and Flickr, which is something that would turn me off contributing and is open to abuse. Bidgee (talk) 06:53, 10 January 2014 (UTC)Reply[reply]
      • Thanks for the feedback. Not obligating the uploader to interaction is an important thing for consideration and is very duly noted. Keegan (WMF) (talk) 03:49, 11 January 2014 (UTC)Reply[reply]
        • Well, there is also another aspect: some users left the Project/s, while some others passed away, so in some cases it would be useless that other users direct their comments towards the uploader, in some other cases it could be undesirable. I don't know whether there should be an opt-in option for comments or not, but the point deserves attention, imho --g (talk) 02:09, 13 January 2014 (UTC)Reply[reply]
          • Very, sadly, true, Gianfranco. I appreciate you mentioning it because it's a point to consider, even more so as the projects age and life takes its turns. Passed away or just vanished, it leads to a problem with noise-to-signal. Keegan (WMF) (talk) 07:17, 14 January 2014 (UTC)Reply[reply]
  • Any possibility of a "magnifying glass" tool that allows you to see a portion of an image at full resolution? (Should be entirely feasible, I've built things like this, though it would take some tricky tuning to make it work acceptably in such an open-ended environment. Still, bandwidths are going up...) - Jmabel ! talk 00:53, 10 January 2014 (UTC)Reply[reply]
  • I'm not sure the short license description (eg. CC-BY-SA-3.0) is meaningless though those who don't know what it means, I think the full license title should be used. Also how will it handle mutli licensed files and those that have custom licenses? Bidgee (talk) 06:53, 10 January 2014 (UTC)Reply[reply]
  • In the concept video I saw "share buttons" and "add to article" in the same place. Social buttons were even more emphasized; maybe I'd change this, giving more importance to the "edit button". Also I'd prefer optional buttons: if I don't have a Facebook or Twitter account I don't want to see them every time (or also just one of them). --AlessioMela (talk) 10:08, 13 January 2014 (UTC)Reply[reply]
  • Perhaps add a means to "advertise" that we need particular images that we don't have? - Jmabel ! talk 16:24, 15 January 2014 (UTC)Reply[reply]
  • instead of "flag" and waste space for a special case, i'd love to have a general simple rating system, maybe with 1-5 stars. -- 19:09, 18 January 2014 (UTC)Reply[reply]

What's missing from this Multimedia Vision?[edit]

  • Just watched the video for the first time and missed Wikidata, the global South, 3D, mass uploads and metadata curation (beyond spam and copyvios). Will watch again and read up a bit and then return with further comments. -- Daniel Mietchen (talk) 21:27, 9 January 2014 (UTC)Reply[reply]
Furthermore, it seems there is an integration with Commons (you can see that at 0.22, with the picture header gallery, if I understood it correctly), but not the other projects. Commons also has all the books scan that Wikisource uses, and the new vision should take care of that too. Wikisource, Wikiquotes and Wikidata are often linked to article pages, and I would like to know how the Multimedia team envisions interproject integration. --Aubrey (talk) 08:42, 20 January 2014 (UTC)Reply[reply]
  • Even in the large view of an image, we ought to spare a few "above the fold" lines for the caption as relevant to the article. Otherwise, it's just a picture, and we shouldn't have to rely on its title to express its relevance to the article. - Jmabel ! talk 00:46, 10 January 2014 (UTC)Reply[reply]
  • How do positioned annotations in the picture fit into this (in particular, how do they fit into the large view)? - Jmabel ! talk 00:46, 10 January 2014 (UTC)Reply[reply]
  • Sounds are often-forgotten, but they are useful and important (pronunciation, animal sounds, distinctive machinery sounds, etc). Also, the videos (and sounds) presupposes that it's technically easier to upload a video (on-the-fly encoding, or some such?) -- Phoebe (talk) 16:42, 10 January 2014 (UTC)Reply[reply]
  • If I well understood the filepage scheme (i.e. min 01:07 in the video), I believe that the license area should perhaps be emphatised by explicitely writing the word "license", first of all :-) Then, I suppose some sort of link should be added, let's say pointing to a pop-up which explains the license and the stuff in detail. Also, I can't see how to edit "traditionally" to add a template, something which can't be solved by the "report issues" pop-up (min 03:24). --g (talk) 02:35, 13 January 2014 (UTC)Reply[reply]
  • support to organize contents: rating, selecting, remember which ones are already rated. -- 19:09, 18 January 2014 (UTC)Reply[reply]

Which features do you think should be developed first?[edit]

My personal favourite would be notifications (as demoed in the video) about uses of files I uploaded or whose page I edited or watchlisted. More later. -- Daniel Mietchen (talk) 21:29, 9 January 2014 (UTC)Reply[reply]

  • rating, and image selection, to help organizing contests and rating. -- 19:09, 18 January 2014 (UTC)Reply[reply]
  • notifications. but of course i do not see this as multimedia feature, so it should go out here :) i hope the notifications will be not only for images, but generally "cross-wiki". ie if there is one account, there is one notification system. -- 19:09, 18 January 2014 (UTC)Reply[reply]
  • Notifications would be amazing (although it would be helpful to turn them off for particular images). Cross-wiki notification is the key one. If I post an image to Commons and suddenly it appears on Hungarian Wikipedia and also on Farsi Wikinews and German Wikiquote, that's pretty cool and something I'd like to know about. —Tom Morris (talk) 21:10, 10 July 2014 (UTC)Reply[reply]


After watching the video and looking through the slides I have a few observations:

  • there is very little emphasis on metadata and descriptions and very little room for displaying them. Many images come with very rich metadata (especially batch uploads from GLAM institutions) which is very important to many people, since each description page becomes a mini article on the photograph (translated to many languages). Even with the bird analogy it is important to identify the birds, the place and time of year when observed, otherwise they will not be likely to be useful.
  • I did not see in your vision much mention of the international aspect of the metadata. I think one of your 2 fictional collaborators (or an extra one) should not speak English but should be able to do most of the functions without it.
  • Going back to the metadata I think wikidata like system where image metadata is stored as properties of the image and categories or tags are generated based on those, would be a system that would get as to where you would like to be in 2016, but than I do not think that would be 2016 but 2020. Especially if we try to retro fit existing images.

Thanks for sharing your vision. --Jarekt (talk) 13:48, 10 January 2014 (UTC)Reply[reply]

  • Once Wikidata integration is complete I expect a healthy amount of use for file metadata. It will certainly play a key role, but it's not one that we can outline a basic vision with when we don't know how it will be used in practice. It will be a part of the full vision especially considering it's language-free data integration. Keegan (WMF) (talk) 07:13, 14 January 2014 (UTC)Reply[reply]

Comments by Ymblanter[edit]

Some things are good, some are not good, and some are harmful. For example it would be indeed great to have echo which sees all the projects, but I am afraid since it has already taken forever to create unified watchlists and they are nowhere close, there is no way we will have a unified echo by 2016. It would be great to have one click less for a full-scale image. It would be great to have a useful mobile version: For example, it is expected that Wikivoyage would be mainly accessed via cell phones, and the mobile version of Wikivoyage has at best 5% of the functionality of the full-scale version.

However, thinking that Kim identifies the bird correctly is a bit of wishful thinking. May be this Kim, if this is her hobby, does, but hundreds of other Kims will at best upload a photo of a bird they can not identify, and at worst will misidentify the bird.

If Jan marks an image as copyvio without any comments, for me as a Commons admin it is just not actionable. I can of course spend 15 minutes of my times, go to Tineye and see whether I get any results there, but I would only do it if I have seriuos reasons to believe this is a copyvio indeed. Jan must have a compulsory field to provide his reasons.

Adding social media repost buttons on WMF project make one more step towards finally turning them into MMORPG. It would be a perfect, convincing, final answer to the question "Is the English Wikipedie ever going to reach its 5Mth article".--Ymblanter (talk) 13:27, 11 January 2014 (UTC)Reply[reply]

Hi. Just one point I'd like to make - we're nowhere close to having unified watchlists, in part because it hasn't been a priority (to the powers that be), so no one has been seriously working on it. (Yes, I am aware that there was some efforts by Victor, as well as Nathan in the past, but it hasn't seem to have lead anywhere. Due to possible scalability tricky-ness and unknowns, fixing that bug would probably require coordination with wmf. I suspect for a volunteer developer to get that bug fixed they would have to talk with platform team (and possibly ops) and convince them that whatever approach they take is scalable/won't bring the site down [Obvious, an employed-by-wmf developer would also have to satisfy similar concerns, but the dynamics are a bit different]). Anyhow that was a bit of a digression, the point I was trying to make is that is an unfair comparison, since presumably if unified echo becomes part of this vision then people will actually be working on it, which is not the case for watchlists. Bawolff (talk) 04:44, 13 January 2014 (UTC)Reply[reply]
Thanks for the update. Anyway, in the absence of unified watchlists I would certainly welcome unified echo. I think it is a very good idea.--Ymblanter (talk) 06:33, 13 January 2014 (UTC)Reply[reply]
Great feedback, Ymblanter. It's very useful to document issues that we have with opportunities like social media sharing. As I mentioned to concerns above, this ties in with the idea that we do not want to force social interaction or collaboration on anyone. Enhancing the options that we have available to do so and building new ones is what we want to work on. As for global watchlists, I think that's for another team to answer. Further thoughts are welcome. Keegan (WMF) (talk) 06:36, 14 January 2014 (UTC)Reply[reply]
Oh, and as for the other two main points: we are well aware of the cost in terms of volunteer time when it comes to maintaining such a repository as Commons, and the burden it can place on the community. We want to mitigate that with as much information as possible on how to do so. Thanks again. Keegan (WMF) (talk) 06:41, 14 January 2014 (UTC)Reply[reply]

Please act after consensus[edit]

In it.wikipedia a lot of users are against this tool and the way you proposed it: before to propose some drastic change like this it is supposed to discuss together with all the Wikipedia and other Wikimedia projects. In particular we don't like the proposal that are directed to transform Wikimedia projects into something similar to a social network. --Daniele Pugliesi (talk) 00:49, 15 January 2014 (UTC)Reply[reply]

"discuss together with all the Wikipedia and other Wikimedia projects": Daniele, this is why I linked this proposal at's VP while it is under evaluation. But, to be honest, 4 people (5 with you), are not "a lot of users" in our Project and do not represent a local consensus (which was never recorded, it's only a couple of days we have this link). Only a few commented, and we can't say anything about the opinion of those who didn't.
Personally, I'm drawing in my mind that this feature could force us to a lot of heavy work to safeguard a fair compatibility with the licenses' management, with the tags we need to add to the filepages for maintenance, and in general for the legal compliance; this wasn't really explained, or I didn't understand it. I hope that licenses and legal issues, and everyday's maintenance won't suffer from the change, or not too much. But I can see that Multimedia Team already made an impressive work, which I sincerely respect and thank for, and I wish that it can truly be successful in attracting new contributions, because this is going to be a critical point in the very next future.
I'm pretty sure that no action will be taken without consensus, this what we are used to in every Project. --g (talk) 01:48, 15 January 2014 (UTC)Reply[reply]
Quote: "I'm pretty sure that no action will be taken without consensus, this what we are used to in every Project." <<< one need only look at the botched implementation of Media Viewer to see that this is no longer the case, that WMF can and will proceed according to its own agenda, regardless of apparent community-consensus (or lack thereof) at the Project-level. JDanek007Talk 21:29, 4 July 2014 (UTC)Reply[reply]
I also don't see consensus for the typography refresh -- another example of what you mention. Last I checked on that poll on English Wikipedia, there were a great many more opposes than supports. Cathfolant (talk) 15:33, 13 July 2014 (UTC)Reply[reply]
I'm not convinced that this tool would improve the quality of the multimedia files stored in Commons, instead it will produce a considerable inndiscriminate accumulation of files without any real value for wikipedia pages plus a huge work about serious copyright control will be request. If storing multimedia on Commons will became a game as now on Facebook, Instagramm, and so on ..., according the people mood, we will get the same quality and some garbage that we see on these socials: nice for free time and joke with friends, useless for any other purposes.
If there is a group of developer willing to develop tool improving multimdial capability of Commons why do not produce features as a true graphic visualisator of the categories tree, an automatic traslator of images description because actually there are a lot of images with description written only in no-english language, so quite not undestandable to everybody for instance ? --Bramfab (talk) 10:06, 15 January 2014 (UTC)Reply[reply]

Thanks for your insights![edit]

Thank you all for your thoughtful comments about this proposed vision! Your feedback is very helpful to our multimedia team, and we are already acting on many of your recommendations.

For example, Jmabel's request for a "magnifying glass" feature is now being designed (see card #167 on our Mingle project management system); recommendations from Phoebe, Quiddity and others to give more support to audio files are also on our roadmap (see card #163); g's request for a pop-up explaining the license terms is now under development (see cards #118] and #197); and we're scheduling file notifications sooner, as suggested by Daniel Mietchen and others (see card#170).

These are just some examples of how your insights are being incorporated in our release plans. We're also reviewing your longer-term recommendations, which will take more time to implement, but are under consideration nonetheless.

Our focus this quarter is on completing the Media Viewer, to provide a better viewing experience for all users. In a few days, we will invite your comments on the new beta version we are about to release. Next quarter, we plan to provide File Notifications, improve the UploadWizard, as well as start implementing structured data on Commons, in collaboration with Wikidata.

To keep up with our work and participate in our conversations, we invite you to subscribe to our multimedia mailing list.

In the meantime, I would like to share some of the comments from our blog, so we can keep track of all community insights on a single page:

  • Stefan: I’m wondering how a multimedia vision(!) for 2016(!) can only contain images and video. Shouldn’t there be embedded geospatial data, interactive data visualisations or 3D models to explore and edit?
  • Hispalois: Very nice! As a frequent Wikipedia editor and occasional contributor to Commons, I would greatly appreciate these new tools; in particular the seamless integration between the two projects.
  • RandomDSdevel: I think that it would be so great if the cross-wiki integration that this blog post is talking about could allow wikis to share pages!
  • Stefan Seidner: Nice Vision and nice video! I had to laugh out loud about the wrong bird :)

Thanks again to everyone who participated in this conversation: keep your great insights coming! Keegan and I will do our best to answer your questions (for a faster response, you are welcome to mention our user name in your comments, so we are notified and can get back to you more quickly). Fabrice Florin (WMF) (talk) 20:03, 7 February 2014 (UTC)Reply[reply]

Comments on wikipedia about File:Multimedia Vision 2016.pdf[edit]

wikipedia:Wikipedia:Village pump (technical)#Future - thought I should drop that link here, since someone said this was the right place for it. Cathfolant (talk) 22:56, 4 April 2014 (UTC)Reply[reply]

This vision sucks[edit]

Here's what I have to say.

  • Wikipedia is meant to be read. That means that multimedia should be integrated with the article, appropriate to individual sections of an article. Creating pretty slideshows of all the figures in an article is pointless. Changing the format of Wikipedia to fit less and less text on a screen makes it harder and harder to read and keep track of where you are. Your release of your presentation as a video illustrates this to an extreme - I can only recollect what I saw, not copy a list of points and go down and refute them one by one.
  • Wikipedia should be free of surveillance. That means that I don't want the ubiquitous Facebook/Twitter tracking scripts on every single page ostensibly so people can "like" the content. I know, I know, that's blasphemy, that's outright theft of my personal data by withholding it from its rightful owners, no less egregious than downloading from The Pirate Bay, but that's where I stand.
  • Wikipedia should be free of ads. In fact, I don't even think it's our role to advertise Facebook and Twitter. Tear down these false gods! The only products we should link on any of our interfaces are open source public domain projects.
  • Wikipedia should work without Javascript. Javascript slows everything down and makes it less secure, and users should have the right to do without it. That means no relying on AJAX to show the large sizes of pictures.
  • Wikimedia Commons should work by community consensus. That means that I don't want users clicking on censorship links about "harmful" images to have them evaluated in secret by whoever receives the notifications. Discussions should be public.
  • Wikimedia Commons is not a toy. You present an example of someone putting rounded little images of birds crudely on a map and linking them to the Gulf with blue arrows. Are you proposing to write a full featured image editor to rival at least GIMP in HTML? If not, then whatever you're showing there is just some toy, like the free novelty graphic editors that come with a cheap Windows box. It's a waste of time, not a real way to make high quality content.
  • Wikimedia Commons is not broken. Our images are used all over the world. I could not pick among the featured content from the last content; it was all great stuff. Users are going to want to keep the old ways of doing things - anything new needs to be introduced on a slow and voluntary basis, a gadget of gradually increasing popularity.
  • Wikipedia is not an app. Because an "app" is not short for "application". It is short for censorship, for corporate control by an oligopsony of internet providers, for material dumbed down to the lowest standard, for form over content, for attitude over essence. The Internet is a messy, dingy place crammed with all manner of incompatible things - we came here because that is exactly what we wanted.

I will add a note of confusion about "tags" not being implemented. How is that so? I can search by multiple terms now. If we wanted, we could put "tag_bird" on all the bird images in a curated way right now. It's not clear how writing new software would make that any easier. What we really need is a search that is no longer flummoxed by funny characters. If I search "<span>" it shouldn't read "span". Basics, that's what I want.

To be fair, I think that notifying editors when their images come in use is a great idea and I'd like to see it ASAP. Wnt (talk) 18:31, 5 April 2014 (UTC)Reply[reply]

Good thoughts here, thanks. Cathfolant (talk) 23:53, 6 April 2014 (UTC)Reply[reply]

Just to be clear on something. At no point in time will we ever include tracking javascript from facebook et al. That would be in violation of our privacy policy, and well just no. It is possible to have those "share on facebook" links without the tracking js/third party cookies. Whether or not we do that is probably a flame-war worthy question, but if that got resolved with a "yes", then we would do it without adding external js. Additionally, MediaWiki is comitted to working with javascript disabled. Some advanced features may require js, but the core of the site will always work without javascript. Bawolff (talk) 16:12, 23 June 2014 (UTC)Reply[reply]
Yes, it's perfectly possible to have Twitter share links without any Twitter JavaScript/tracking cookies. That said, I don't think Wikimedia should be encouraging Twitter or Facebook sharing particularly (nor any of the other corporate-owned social silos). —Tom Morris (talk) 20:53, 10 July 2014 (UTC)Reply[reply]
While I agree with some points of Wnt's comments, I'd suggest that he's perhaps being unduly harsh to the Multimedia team at the Foundation. The fact is MediaWiki's usability sucks. It's normal for us as contributors, but there are a lot of people out there who could contribute if we made the tools easier and more up-to-date. I know technical people who aren't the least bit stupid and they don't contribute to Wikipedia because they see a page full of horrendous wikitext and an unforgiving community. When they find an error or something to improve, they email me about it.
If we have an article that needs an image, we should have a really simple user flow so that new contributors can add those pictures while also not disturbing the existing contributors. If someone sees that an article lacks a picture while using their phone—and looking stuff up on Wikipedia is one of the many uses of smartphones—we should make it so that they can easily upload a picture. If we don't give them a great experience and make it slightly less tortuous to contribute, well, Facebook and Instagram and Twitter and so on are more than happy to continue doing so.
Similarly, good design and a high-quality encyclopaedia (and dictionary and quotes book and all the other stuff) are not in opposition. Digital cameras are getting better and better. Hell, the smartphones of today have some pretty damn good cameras. But good photography on Wikipedia gets short shrift: it gets stowed away in a little thumbnail, rather than allowed to shine. I don't like the MediaViewer and have turned it off, but I would rather the Multimedia team experiment and build new things and try shit out than live in fear of being told that everything they do "sucks". —Tom Morris (talk) 21:07, 10 July 2014 (UTC)Reply[reply]
As far as I'm concerned, the only way to make it 'really simple' to upload files is to eliminate the licensing requirements, which obviously can't be done. Take Uncyclopedia, for example. Their upload page says nothing whatsoever about this and that licence; only that you must not upload shock images, which is common sense. They also don't have a weird buggy 'upload wizard', or inexplicable bugs in Special:Upload. I suppose someone could try to fix the bugs, which would be great, but even that wouldn't accomplish what you seem to be aiming for. Cathfolant (talk) 15:44, 13 July 2014 (UTC)Reply[reply]

Introduction of "Tags"[edit]

It is mentioned in the video that tagging images with keywords (like it is done on Flickr etc.) is not possible yet. That's not correct, since we already have Categories that serve the same purpose. I'm not going to argue about pros and cons of Categories as they are now, tags as they are elsewhere, or more sophisticated methods that could be implemented with Wikidata for now, as there are many things to consider. But let me make this abundantly clear: The Category system surely is far away from perfect, but it is Commons' most important tool for organizing our content. If you are planning to change anything in this direction, make sure you have a deep understanding of how it works and prepare for prolonged discussions with the community before the first line of code is written. Don't make the same mistakes again. If you just develop something and throw it at the community, I promise you it will be media viewer all over again – and that's something we'd all like to avoid, right? --El Grafo (talk) 12:11, 21 August 2014 (UTC)Reply[reply]