Commons:Upload Wizard feedback/Archive/2009/12

From Wikimedia Commons, the free media repository
Jump to: navigation, search

Enable several types of categories[edit]

Currently we only have two types of categories: hidden and normal categories. The first are only visible to users that are logged in and enable the option. They generally include categories by camera type, copyright tag, and image source.

Rather than grouping categories into hidden and non-hidden categories, it should be possible to split the non-topical categories into groups and display these separately, e.g.

Picture of the day This image was selected as picture of the day on Wikimedia Commons for . It was captioned as follows:
English: "Passage" group, Hautecombe Nights, Savoie, France, 2008.

Sample File:Passage - Nuits de Hautecombe 2008 - 7.jpg (featured picture of today)

Topical categories
Nuits de Hautecombe | Males with microphones | Shadows | Arms
License/copyright categories
License migration redundant | GFDL | CC-BY-SA-3.0,2.5,2.0,1.0
User or image source categories
Photographs and images by Yann Forget | Quality images by Yann Forget
Date taken
Photos taken August 2008
Image type category
image/jpeg
Camera categories
Taken with Sony DSC-R1
Commons Assessment
Pictures of the day (2009) | Featured pictures on Wikimedia Commons | Quality images
Warnings
Personality rights warning
Work source type
Self-published work

It should also be possible to search for a combination of these types of categories. As many of the categories should be added manually rather than automatically (or the other way round).

-- User:Docu at 10:29, 6 December 2009 (UTC)
I like this idea. mahanga (talk) 18:40, 8 December 2009 (UTC)
Je trouve que cela apporte une grande clarification. Si on réussit à bien classer et hiérachiser les informations, on peut ensuite se poser la question : "quelle visibilité donner à quelle information ?". Les informations importantes vont être les plus visibles, les plus proche du haut de l'écran (ou du quart en haut à gauche de l'écran), et les autres vont être plus éloignées voire cachées. Dans les catégories il y en a qui sont les plus importantes qui doivent être les plus visibles, et d'autres qui le sont moins et qui peuvent rester cachées. Teofilo (talk) 17:45, 17 December 2009 (UTC)
See page 6 and following
This is actually very close to the mock-up I did (see document on the right). guillom 22:44, 5 January 2010 (UTC)


Looking at it closely, probably yes, even though I can't tell where the links lead. Excellent work.
Since I inverted the display of categories and the remaining description (see below), I tend to look at categories first, forgetting that the same information is also available in the information template and/or exif below. Most current tools at Commons don't make much use of these though. The main advantage of the categories is that they lead to images with similar attributes, e.g.:
  1. Taken with Sony DSC-R1 lists images while the same in exif only links to Wikipedia.
  2. The date in exif doesn't do anything, the one in {{information}} only localizes. Compare with Category:Switzerland by month that allows to browse new (and old) images by the month they were taken.
  3. The author's name links to the user page and eventually to the gallery tool, but this isn't as convenient as, e.g. Category:Photographs and images by Yann Forget.
  4. The license tags link to texts about the license, etc. ..
BTW your mock-up lists location. This isn't readily available as a field, except in the rare cases of descriptions using {{information2}}. It is generally included in the description field of {{information}}. Obviously it can be derived from {{location}} and categories. For categories, in the sample above, it would be Abbaye d'Hautecombe or Saint-Pierre-de-Curtille, parent categories of Nuits de Hautecombe.
In the mock up, "See other media files about" is fairly far down the screen. It doesn't necessarily need to be above the description, but preferably close to it.
Back to the initial point, in the mock-up, do the fields link to categories? (I noticed it mentions categories on page 17 and the license should link to an explanatory page (which could be the category page)). -- User:Docu at 14:42, 9 January 2010 (UTC)

Display categories below the image[edit]

Currently an option in Special:Preferences allows to display categories directly below the images. These seems to be reasonable default for Commons. -- User:Docu at 10:29, 6 December 2009 (UTC)

Agree. Categories have a more important role in Commons than Wikipedia which often have wikilinks or see also sections. Thus, they should be in a more prominent location, maybe even row-based on the side like Flickr has their tags. mahanga (talk) 15:23, 9 December 2009 (UTC)
I agree; we would like to make navigation easier, and that includes within categories. guillom 22:44, 5 January 2010 (UTC)

Changing of file formats[edit]

It would be great to allow a change of file formats of existing files. For example: If I wann change a jpg Image to be a png, I currently cannot update that file by uploading a new png-version of it.--Juxn (talk) 16:03, 8 December 2009 (UTC)

This is bugzilla:20971 - Upload new versions of files with different file type. guillom 22:44, 5 January 2010 (UTC)

Allow multiple uploading[edit]

For multiple uploads of files with same/ similar values (license, description, etc) it would be very nice to have a interaction that allows this as a massupload.--Juxn (talk) 16:03, 8 December 2009 (UTC)

This is bugzilla:488 - Upload more than one file at a time. guillom 22:44, 5 January 2010 (UTC)

Download link for Timed Text subtitles[edit]

There should be a download link to download Commons:Timed Text as a .srt file. This will make it easier for people who download videos and still use subtitles. Example: TimedText:051118-WSIS.2005-Richard.Stallman.ogg.en.srt. The current way to do save the text is by copying and pasting. mahanga (talk) 15:45, 9 December 2009 (UTC)

Added download link to subtitles to the mwEmbed player, will be in the mwEmbed interfaces in the next update. Mdale (talk) 00:23, 13 July 2010 (UTC)

Thumbnail picture for videos[edit]

Maybe there could be an option to select one of three thumbnails (beginning, middle, end) for a video. We're able to do this when linking to the file (ex. [[file:video.ogv|thumbtime=13]]) but not when going to the file page. mahanga (talk) 04:57, 15 December 2009 (UTC)

This is bugzilla:20647 - Allow way to choose thumbnail frame for video on its File: description page. guillom 22:44, 5 January 2010 (UTC)

Le nouveau répertoire des utilisations des fichiers dans les autres wikis[edit]

  • Traduction et vocabulaire

Traduire "usage" par "usage", c'est de la traduction de paresseux : Cf fr:calque (linguistique).

Si l'on tient vraiment à afficher cette section directement sur la page, le vocabulaire devrait être harmonisé avec la section qui précède, à savoir actuellement : "Utilisations du fichier". Je propose de renommer ce titre en "utilisations locales du fichier" ou "utilisations du fichier sur Commons".

"frwiki" "enwiki" "dewiki" "fiwiki" etc... ce sont des termes de jargon. Je préconise l'utilisation d'un langage non jargonisant lisible par les débutants. Donc "Wikipédia en français", "Wikipédia en finnois", etc...

  • Allourdissement de la page

Chacun rajoute son petit bout de logiciel et la page devient de plus en plus lourde. On a eu le gadget "ajouter une note" l'été dernier. Ensuite c'est l'automne, et on voit apparaître ce répertoire des utilisations qui est potentiellement très long (autant que de langues dans le monde : 100, 200...).

Et donc on accumule, on rajoute sans la moindre vision d'ensemble. On fait programmer des informaticiens sans leur donner de cahier des charges. C'est à dire qu'on ne sait pas ce qu'on veut, mais on fait quand même, et on se préoccupe aucunement des conséquences. Comme on ne sait pas quels sont les objectif, on ne sait pas si ils sont atteints.

  • Structuration de la page

Comme je l'avais déjà dit, il y a un problème de hiérarchisation des informations dans la page. Il me semble qu'il faudrait mettre le plus près de l'image les informations qui intéressent le plus d'utilisateurs, et éloigner les choses les moins intéressantes. Mon avis personnel (mais il y a une discussion à faire, peut-être que d'autres personnes pensent différemment) est que les catégories, c'est important. Le répertoire des utilisations éloigne encore plus les catégories de l'image.

  • Déplacement dans un onglet

Je préconise donc leur déplacement dans un onglet dédié, comme c'est déjà le cas pour l'outil "checkusage" qui fait exactement la même chose. Mais on pourrait utiliser un indicateur de couleur pour dire si cet onglet est vide ou non-vide (pour éviter de cliquer dessus s'il n'y a rien à lire).

  • Gel des développements de logiciels "fait accompli".

Je pense qu'il faut des cahiers des charges, des maquettes, et des discussions sur ces cahiers des charges et ces maquettes avant de mettre en oeuvre des bouleversements de l'espace de travail des bénévoles et l'espace de ressources des internautes en quête d'images libres.

J'ai l'impression qu'on a plus besoin d'une méthode de prise de décision que de logiciels en tant que tels. Et donc il vaudrait mieux tout geler tant qu'on a pas trouvé cette méthode. Tant qu'on ne sait pas où l'on va.

  • Multi-modal

Il faut prévoir différents modes d'accès en utilisant les "préférences" et les "peaux" (skins). Un accès épuré (philosophie de la page d'accueil de Google) pour les débutants. Un mode d'accès "pour les curieux". Un mode d'accès "pour le travail" et "à la carte" pour les bénévoles.

Teofilo (talk) 18:08, 17 December 2009 (UTC)

Not censoring Commons, but yet giving an option to censor this site's content[edit]

While I have no problem with the fact that Commons is not censored, there are other people who do. Often, this leads to talk page posts, image description page edits or (malformed) deletion requests where it's complained about "porn". It would be good if people (also unregistered users) could set in some kind of preferences how to censor Commons. I've thought about something similar to the Safe Search from Google (meaning that offensive images will be excluded from the search results, categories and galleries), though probably "No filtering" should be the default at Commons. --The Evil IP address (talk) 11:03, 2 January 2010 (UTC)

Who, then, would decide whether something ought to be subject to the filter? Powers (talk) 19:44, 2 January 2010 (UTC)
Users would decide for themselves. Something like the old spam filter lists: several lists were offered for various purposes and each user could select none, some or all filters. In fact users can (I do myself in my HOSTs file for example) add their own additional filters. I think this is a good idea if taken further than just an on or off preference. It might need some technical coordination with Commons (maybe html tags, classes, whatever). I would like one to suppress very large GIFs for example, to avoid wasting the bandwidth of the server and my own. I have linked to this idea from Commons:Village_pump#Simplified_proposal. -84user (talk) 13:47, 3 January 2010 (UTC)

Commons in the web 2.0 world[edit]

The usability of the Commons would be substantially improved by a genuine web 2.0 platform. Ideally, the project should aim to provide a seamless end-user experience that rivals sites like Flickr and YouTube. Here are a few observations:

Issue: Newcomers are not sufficiently enticed to join the community.
Solution: Expand access to the project by allowing users to log in via alternative (OpenID, Google, ...) authentication systems.
Issue: Our category system discourages inexperienced users and is overall very inefficient.
Solution: Rename categories to tags, and introduce a low-latency smart tagging system that engages drive-by editors.
Issue: The Commons interface is overtly pedantic.
Solution: Transform talk pages into vibrant discussions that are presented directly below the media, and create an annotation platform. Significantly improve aesthetics, speed, and accuracy of default search functionality; use an AJAX gallery by default.
Issue: The Commons does not accommodate rich media content.
Solution: Automatically convert uploaded videos to Theora, and raise the file size limit to 4.7 GB.
Issue: Current content guidelines antagonize casual contributors.
Solution: Liberalize policy to indiscriminately permit all free content regardless of quality or practicality, and instead develop a quality control scheme that prioiritizes the results of media searches based on user ratings.
Issue: Very few members of the general public are familiar with the Commons.
Solution: Attract, facilitate, and encourage the layperson to painlessly (and perhaps even inadvertently) contribute quality media. Take advantage of the viral effects of social media by incorporating an interface that enables users to share and embed media (e.g. email, Delicious, Digg, Facebook, LinkedIn, MySpace, Reddit, StumbleUpon, Twitter, etc) across the web.
Issue: The Commons is not "usable" from a corporate perspective, and virtually no media conglomerates or commercial ventures contribute any content whatsoever.
Solution: Create an easy to use system that encourages traditional copyright holders to make tax deductible donations of previously commercial movies, music, photos, scientific journals, and books. Once the Wikimedia Foundation legally owns the rights to said media, it can then be legally re-released under a permissive license. Form a grassroots volunteer group with the sole purpose of going down a list of every major media organization (e.g. MGM, CBS, Virgin, etc) in the world -- one-by-one -- to personally make requests for unprofitable media.   — C M B J   00:08, 27 October 2009 (UTC)
« Transform talk pages » : You might be interested in mw:Extension:LiquidThreads (currently tested on strategy).
« create an annotation platform » : ImageAnnotator has been active here for a while now.
« Automatically convert uploaded videos to Theora » : You might be interested in mw:Extension:Firefogg (currently tested here on Commons, see Commons:Firefogg).
--Jean-Fred (talk) 17:37, 19 November 2009 (UTC)
Based upon the vast number of non-technical users contributing photos to Flickr and videos to YouTube, it can be inferred that there is a very strong correlation between usability and content volume. Such an environment depends upon three two primary conditions: (1) it can be easily utilized by anyone aged eight to eighty, (2) viral activities are facilitated, promoted, and rewarded, and (3) lawful content is accepted without discrimination to quality. Obviously the latter is antithetical to current Commons scope and policy; in order to harbor such a nonchalant community whilst preserving the academic value of the Commons, it would be necessary to create an intuitive rating and tagging system.
Community driven extensions are a great way to improve any open source project; however, it is my opinion that the rigidity of our current interface (or any derivative thereof) is beyond the threshold of reasonably tangible rectification, and as such, will continue to dishearten and intimidate the layperson indefinitely. I hypothesize that it would be a wise and stewardly investment to allocate donor funds in pursuit of the aforementioned objectives.   — C M B J   01:36, 4 January 2010 (UTC)
If we cease discriminating based on quality and value, we risk losing the ability to discriminate based on legality, due to volume. Note how many blatant copyright violations remain on, say, Flickr and especially on YouTube. Powers (talk) 02:32, 4 January 2010 (UTC)
It has been my personal observation that users routinely contributing from the same camera(s) are typically not infringing copyrights, so it is not unreasonable to assume that a server-sided EXIF-analyzing script could approximate the infringement risk of any single file within three or four uploads by the same user. With that said, inclusion criteria is not an issue that I feel strongly about, and I do not want it to serve as a distraction from my other (more valid) points. Can users login (OpenID, ...) without creating an account? No. Are all users encouraged to tag images? No, not many people are willing to spend five seconds tagging a single file. Are visitors encouraged to virally share Commons files like most modern community-driven websites? No, not even close. Can a non-technical Internet Explorer user with a 1.6ghz Atom netbook and a 1080p camcorder from Wal-Mart come to our site and upload Theora video within 30 seconds? Not only no, but hell no. As someone who has actually uploaded Theora video in the past, I would estimate that the current process is beyond practical for all but the most patient, highly motivated, and technically inclined individuals equipped with state-of-the-art processing power. How many major commercial media outfits (ABC, CBS, CNN, NBC, Fox, Disney, NYT, TIME, ...) are making tax-deductible media contributions today? Even though many of these companies own rights to unprofitable (but valuable) media that will inevitably enter the public domain during the next ten to twenty years, many of them routinely make multi-million dollar tax deductible contributions annually, and even more of them are dying for new ways to cut costs (taxes) while better promoting themselves, the answer astonishingly zero.
If we want to remain competitive in the world of the future, these issues must be considered a top priority.   — C M B J   19:01, 5 January 2010 (UTC)

Archive added[edit]

I have just copied this revision of this page that was pipe-linked above as "Archive 1: 21 October - 4 December 2009" to Commons:Usability issues and ideas/Archive 1. I then added a {{search box}} at the top to hopefully help readers search both this page and its archive. The search index takes time to populate, and is sometimes temperamental and incomplete (another usability issue that could be improved). -84user (talk) 14:48, 3 January 2010 (UTC)

Searching is useless[edit]

I can't easily find any content that I may wish to re-use. There is no facility for me to say 'Give me a picture of a soccer player scoring a penalty', I can spend three hours and perhaps come up with two or three possibles, but why should I do that when I can go to flickr and enter 'soccer penalty goal' and get a couple of thousand results in less than a second? Commons for some reason seems to rely on descriptions for it's results, and not many are filled in with enough verbosity for the search results to be meaningful (how many great pics are missed in searches purely for that reason?). Today I came to Commons to look for an image of a young boy with dark/asian features wearing a red shirt, that was my brief - guess how many images I could find? It was less than one.. Nanonic (talk) 02:33, 20 January 2010 (UTC)

Question: What is the educational use you want it for? Paradoctor (talk) 03:42, 20 January 2010 (UTC)
Educational use? None. I required it for a mailshot for a local company that I'm knocking together and found a CC-BY-SA pic on flickr that I can use. Nanonic (talk) 04:55, 20 January 2010 (UTC)
Nothing wrong with using files for non-educational purposes, but why do you expect Commons to facilitate any search not related to educational uses? Paradoctor (talk) 05:05, 20 January 2010 (UTC)
Does it matter? The concerns are the same regardless of the use being sought. Powers (talk) 18:51, 20 January 2010 (UTC)
Not quite. The intended educational use should be our primary access path, everything else is merely a bonus. It would be nice if Commons could be used as a stock photo resource, but as Nanonic's experience show, Flickr is already there. As an aside, I'll bet you a dollar we don't have such an image of a boy. Oh, and BTW, there is a Category:Penalty kick, took me all of 20 seconds to find it. ;) Paradoctor (talk) 20:47, 20 January 2010 (UTC)
I think Nanonic point would be that, precisely, the search engine does not build on our category system, or at least with difficulty. His experience with searching for non educational purposes would probably be the same than others with educational purposes.
I can certain agree that the color of the shirt is indeed not an educational priority (though we do have Category:Red clothing, male ! Clin). But looking for penalty kicks sure has legitimate educational purposes (or can have), and the fact is that our friend here could not locate the relevant category. Jean-Fred (talk) 21:18, 20 January 2010 (UTC)
What I see is that Nanonic did not use either of three rather obvious approaches: If a search for three keywords yields no hits, try two, like "penalty soccer", which has the category on the first results page. Number two, ask Wikipedia: w:penalty links to w:Penalty shootout (association football), which has a link to Category:Penalty shootout (association football). Number three, Commons:Help desk. Seriously, I fail to see where Commons has a problem here. Paradoctor (talk) 00:23, 21 January 2010 (UTC)
Well, obviously, those approaches were not obvious to him. Of course there are ways to find stuff on Commons. The fact is that this user could not find his way around, and that is a problem for Commons.
IMO, the point of this page is to have users tell what problems they encounter with Commons, in order to have ideas to improve it ; not to explain users how they do not think kow they ought to and to defend Commons. Jean-Fred (talk) 00:48, 21 January 2010 (UTC)
"not obvious to him": Check his contributions, you'll see that Nanonic knows about the help desk. The claim that "There is no facility for me to say 'Give me a picture of a soccer player scoring a penalty'" is plain wrong, that is exactly what the help desk is for.
"the point of this page": Sure, but as I argued, this is not a problem of Commons, it's a problem of Nanonic, especially considering his/her claim of having wasted hours on the search. Paradoctor (talk) 01:55, 21 January 2010 (UTC)
I don't understand what Paradoctor means by an educational use being an access path. I also note, for Nanonic's edification, that entering "soccer penalty goal" (the specified search string) into our search engine does indeed come up with a result -- a featured picture, no less: File:Ryan Valentine scores.jpg. Powers (talk) 23:41, 20 January 2010 (UTC)
Hah, should've checked that one. ;) About "access path": Had Nanonic specified an educational use, we would have keywords we could talk about, and check where they lead. We would have something we can talk about, and be able to determine whether there is something we can and should do.
Motivated by Nanonic's complaint, I did spot something we can and should do: The main page does offer only the search box and a link to Commons:Picture requests. I think it would be a good idea to put a nice big friendly box at the top which points users to an introduction to finding stuff on Commons. This introduction should be adapted to the requirements of those users for which online existence is not second nature, which is probably the vast majority of our target audience. A specialized help desk might prove useful, too. Paradoctor (talk) 00:23, 21 January 2010 (UTC)
  • Even if the search term would be better (or one included in a category description), unless someone describes, categorizes or tags the images, the search isn't going to succeed.
    BTW flickr seems to have less results if one searches the correct set. -- User:Docu at 13:42, 21 January 2010 (UTC)
    • You won't find an image on Flickr either, if no one describes or tags it. Powers (talk) 14:48, 21 January 2010 (UTC)
      • That's for sure. Still, good to know that we already got more (or just as many) results ;) -- User:Docu at 10:22, 23 January 2010 (UTC)

Thumbnail size on panoramics category pages[edit]

Larger thumbnail (height is still just 76px)
Random panorama from Category:Panoramics with standard category thumbnail size

Panoramics categories (Category:Panoramics by country, Category:Panoramics, etc.) tend to be full of images that are much wider than tall. For these, standard category thumbnails aren't that useful (e.g. the random sample to the left).

It would be nice if these categories could rendered with wider thumbnails. It wouldn't matter much if there was just one per row.

Similar to __NOGALLERY__ a magic word could be used to switch display. -- User:Docu at 10:22, 23 January 2010 (UTC)


Related Bugzilla:22267 entry -- User:Docu at 19:53, 25 January 2010 (UTC)

It's closed already! Implemented ! (No, duplicated of Bugzilla:13802) -- User:Docu at 20:16, 25 January 2010 (UTC)

Hardly anyone gets what Commons actually is[edit]

There are hundreds or thousand people here who have no idea what they're doing. They don't understand copyright at all, don't know anything about the scope and also don't understand that licenses are irrevocable. They upload fair use files (sometimes even with fair use rationales), promote themselves or their company, use Commons as a private image hosting site or upload pictures of themselves or other things they care about and request deletion then, which is usually denied as licenses are irrevocable. This then often leads to frustration; furthermore, it's extremely time-consuming to clean up after them. I believe that something like raising the upload right to autoconfirmed or flagged uploading or such bullshit doesn't fix this problem. The problem is not that they're new, it's the problem that they don't get it. Maybe it would be good if there would be some pages that you have to read, maybe a step by step tutorial about Commons (of course in the localized language of the user), before you're technically possible to upload files. But there might be better solutions, as this might be annoying for people who already know this. --The Evil IP address (talk) 12:35, 24 January 2010 (UTC)

Maybe "donate your image(s)" instead of "upload file" would address some of it. -- User:Docu at 18:02, 26 January 2010 (UTC)
There are several problems with that wording: (1) We accept more types of media than just images. (2) We encourage uploading not only of the uploader's works, but of other free works as well. (3) The phrase implies that one physically gives up one's media, when, in fact, only some immaterial rights to the media are waived. We shouldn't add to the public confusion between physical property and immaterial rights. (4) The physical property connotations of the term obscures the concept that free licenses are issued to the world, not just to Wikimedia, which is another concept many uploaders apparently struggle with.
I apologise if that critique of a three-word phrase seems excessive. It's meant to be comprehensive rather than harsh.
I think The Evil IP address is on the right track regarding a step-by-step tutorial. Basically, a rewritten version of Commons:First steps should be required reading before the first upload, along with a quiz of three to five very basic multiple choice questions. I say "a rewritten version" primarily because of the first of the first steps. (The rest is quite good.)
The first link under getting started is a link to the Project scope policy, masked by the friendly link text "Understanding Commons." Policies, by nature, are verbose. They should be further reading, not part of a first steps guide. We don't need to tell new users about allowable reasons for PDF and DjVu formats as part of basic training. "Misunderstanding Commons" or "Getting confused by Commons" would be a more apt title. The first of the first steps should be a list of three to five things Commons is not and a list of three to five things Commons is. Condense COM:PS to its very essence and save the ifs and buts for the further reading links. LX (talk, contribs) 19:17, 26 January 2010 (UTC)
To illustrate, I mean something like this. LX (talk, contribs) 20:57, 26 January 2010 (UTC)

I don't disagree with your points (1), (2), (3), but the question is if this leads to people uploading just about anything and if an alternate wording would discourage people to do so without discouraging to upload files we'd want. A concise version of your proposal might achieve the same and could even fit on top of Commons:Upload.

To get a clearer view, some statistics on deletions might help us. I'm not sure if there are any, so let's compare from the logs of the last 48 hours:

  • 1290 files deleted, including 385 speedied as copyvios (817 by 3 admins, the reminder by 61 others)

During the same time, there were approx. 15600 files uploaded:

  • 6600 by 1 user
  • 800 from flickr
  • 1100 by 10 users
  • 4750 by 440 users
  • the remaining 2300 files:
    • 182 users uploaded 3 files (includes 36 users new in the last 48 hours)
    • 376 users uploaded 2 files (includes 72 users new in the last 48 hours)
    • 1000 users uploaded 1 file (includes 275 users new in the last 48 hours)

There were also 2200 new accounts created (383 of these uploaded 807 files).

Obviously these aren't directly related, don't provide much detail, nor take into account that the upload link in one Wikipedia version allows you to upload fair use stuff and in others leads you to Commons. Besides, a 7 day period would probably give more significant data, but if half of the last 2300 files are deleted, it's a problem. -- User:Docu at 08:59, 27 January 2010 (UTC)

That data is definitely interesting, and something to keep in mind when making changes. I think our goal should always be to eliminate the biggest problems first, so playing by numbers is a good approach. I'm guessing the 6600 files by a single user means the user is a bot account? Bots should probably be filtered out from statistics used for this purpose, since bots don't read instructions. LX (talk, contribs) 16:42, 27 January 2010 (UTC)
The 6600 are by Nevit. I wouldn't exclude any type of upload but group by own work/other sources and the number of the uploads a user did (deleted/not deleted).
To get a link between deletions and uploads, we might want to look at a specific week three or four months ago. Besides deleted files, we might want to look at the percentage that is uncategorized as well. -- User:Docu at 08:58, 30 January 2010 (UTC)

Suggestion of a new way for media donations[edit]

Suggestion (special wiki media donation project): a new highly efficient way for world public to donate images and video to wikipedia/wikimedia/commons etc.

  • There should a link on every wiki family project page saying "DONATE AN IMAGE OR VIDEO". Anyone should be able to click on it and make/upload a media donation easily, SPECIALLY SELF CREATED WORK with the profusion of relatively cheap personal devices available now in the NEW AGE.
  • In the donation form there should be all related disclosures and and Licenses that the donater can choose regarding donating a particular work.
  • There should be place where Donater has to fill all related information related to media being donated like location, subject, time, basic info about self, etc. and details if from other source.
  • People should be able to send copyrighted/unusable images to make suggestion of what we are missing and which we can put out on a list of required media, this list may be easily available to public view and/or ciculated.
  • There should be an email address to where images/video/media can be emailed, clearly stating that the donater has read the various disclosures and licenses (on our website or our news release requesting media donations) and he is donating under which license. Once a media donation email is received to a special mail account: 1) once an email media donation is received from the donater and automatic (from wikipedia/media/commons etc.) email should go out to their address from our no-reply email address reminding them to read, in mail, provided disclousers and provide appropriate license and info for use again. Donater should be asked to email back the reply keeping the same address line explaining preference will be given to media donations which have been replied to, hence making them more usable earlier. 2) It should be available in an online archive for mining by users looking for images for articles they are working on and the public at large.
  • Replied email donations with disclosures read and licences provided and direct media donations by clicking on link should be at all times available to our users and world public at large and journalists etc. There should be a warning to them to make sure BY THEMSELVES that they check (what could be our mostly unsupervised database) out if the donation has been made properly and if donater has read disclosures and provided consent and chosen the license properly. Users should be provided a basic guideline on how to make sure if the image/media is good.
  • Proper donations should ask for keywords that should activate various tags for easy mining of donated database.
  • The whole online media donation/uploading process should be VERY SIMPLIFIED with users asked to click/select choices with one click only from various choices after reading all. Short Disclosers should be page wise only, advancable by clicking NEXT so that all get read. Licenses should be chosen by a simple click from a choice. Media Info should be requested by filling blank by blank advancable software, including location, subject, time, DATE, donator info etc.
  • This Media donation project should be centralized in commons with centralized email for donations. Project should be accessible from all wiki family projects from all their pages at all times by clicking.
  • There should be a special option/Tags setting alerts for media donations regarding HAPPENING EVENTS and that should make news where world journalists/News companies can find Important or Immediate topics to pursue and other agencies like Aid agencies etc. to find places and subjects to assist. A media related to citizen reporting a historical national monument in bad shape should have the potential to trigger positive action to conserve. Potentially database mining should be able to facilitate new discoveries and affirmative action in right direction and build a tremendous world resource to record history/historical period datewise over the decades.
  • Anyone mining the database should be able to setup warnings with a simple click about offending/sexually explicit/illegal images and special users with experience and extra powers should be able to either remove the image or make it invisible where in doubt.
  • Should an option be provided where users of this donated media like journalists/new companies can provide citation like: Donated media from wikicommons server by ..(name of original donator).
  • Donators should be able to choose their nonconflicting wiki User name, and make it a Tag, so that by clicking on name tag all images donated by the users can lineup in a online gallery for public and for donator to promote himself in other/outside professional media fields, if he chooses to provide link to this online portfolio.
  • There should be a clear warning that there is no monetary compensation by wikicommons for media donation of any kind, it is a DONATION.
  • Wikicommons software should be able to mine technical info of media if possible and provide the same online for researchers, sometimes it may include type of camera used, aperture, date, time of day and in the near future models, the GPS position/coordinates of where the media was made. If needed donator should be asked to give consent to publish this info.
  • Public/Users should be requested to make media for donation with no recognizable faces/adertising/brand names as that may trigger having to take permission from people etc. shown in media made or fuzzing their faces/advertising/brand names etc.

Please forward to concerned persons/department for brainstorming and fine tuning.

I got the above idea while creating the article on Karvi shrub which only flowers once in eight years before dying.

Thanks

User:Atulsnischal (talk) 22:06, 25 January 2010 (UTC)

Pasting from wikipedia [1], [2].

Posted on wikimedia commons at [3]

Posted on - Wikipedia talk:Creation and usage of media files - at: [4]

Posted on - Talk:Proposals for new projects; From Meta, a Wikimedia project coordination wiki - at: [5]

atulsnischal (talk) 22:14, 25 January 2010 (UTC)

Building a “media donation culture” and a “world media archive” from donated media – help enlisted from volunteers to “visually document the world” for an ongoing visual world historical record.[edit]

This is in continuation of my last post, posted at the following locations: [6], [7], [8], [9], [10]

Building a “media donation culture” and a “world media archive” from donated media – help enlisted from volunteers to “visually document the world” for an ongoing visual world historical record.

  • We need to strive for creating a “media donation culture” where donating pictures and documentary videos, shot on people’s personal devices is thought of as a worthwhile activity / community service and a scholarly voluntary work which goes a long way in visually documenting the world we live in, in present times. An ongoing visual historical record available free for present researchers/authors to supplement their work with and for future generations to come, to look back upon. Gradually we should hope this will catch on as modern popular culture/fad, will also empower citizens and assist them in making citizen reports. Hopefully a time will come when if someone is taking a vacation then he will remember and friends will remind them not to forget making some documentary pictures/video for donating about the places they will visit during vacation. All this can be done easily now as there is a profusion of cheaper and cheaper personal devices available now incorporating embedded Still & Video camera technology.
  • This is about Campaigning and Requesting for “100 percent Media Donation” for “documenting the world” in a “copyright free world archive” to serve in the future as a world historical visual documentation. This could be an “uploading + emailing project” by itself and donated pictures/video/media can be freely usable in all wikis and similar projects or commercial publications/productions etc.. Commercial entities could be encouraged to develop an informal “best practice” where they “donate funds” to upkeep this world archive, especially if they use anything from here and if they can afford it now or in future.
  • There should be a link (“Donate an Image or Video”) to upload and donate media on every wiki family project page including on every page of various wikipedias in different languages and it should be centralized in one place like wiki commons in a world media archive. As and when possible over the decades viewers should be able to see/read this archive in their own language through translation software etc.
  • All media donations to be uploaded/stored in “high resolution” to be more usable to future generations. Most searches in the archive should only display a decent low resolution image where the user should have the option to call for higher/highest resolution version.
  • There should be a centralized Email address for media donation, and where ever it is publicized it should give all disclosures that this is in regards to 100 percent media donation.
  • The above media donation link and email address should be well publicized in newspapers, magazines, and press releases and media requests; with all related disclosures that media sent will be treated as 100 percent donation and the act of uploading/emailing means that donator has read all disclosures which are provided and well publicized here and elsewhere.
  • It should be preferred that media donation should be made by “original media creators” by themselves mainly as someone under normal circumstances can not donate someone else’s property.
  • It should be made very clear in simple straight forward language at the very beginning before uploading donated files that one is making a 100 percent media donation, “100 percent Media Donation” means there will be no copyright, no royalties, no monetary compensation or any compensation in any kind paid in return for the media donation by the wiki family or any secondary users and the world public to the original media donator/creator. One’s donation can be used, reused, modified, broken up and made into anything else by anybody in the world and that too without having to give any sort of credit to the original media donator. There will be no legal requirement to give credit; and all these rules to apply world over. Though some sort of voluntary “best practice” should be encouraged to emerge where secondary users of the donated media, i.e. the world public including journalists, newspapers, magazines, books, businesses, government institutions the world over, who use donated media to supplement their work, could be encouraged to develop a popular culture of crediting the media to wiki’s media donation campaign/world archive and have a reference/courtesy credit to the original donator. Voluntary citation could sound like: Wiki common’s media donation server, original media donated by…. Commercial businesses/institutions who use donated media could be educated by a regular campaign that if affordable monetary donations can be made to wiki for help in maintaining wiki’s media archives. This donation can be made in any country and wiki could set up a local body in major countries so that tax break forms can be given in regards to monetary donations.

Now the question remains what absolutely free but worthwhile perceived benefit could a media donator receive in regards to donated media. Can a $5 per piece ad hoc acknowledgement tax break form be sent automatically to the media donators in regards to media donations, at the time I guess not. A popular culture should be created where the biggest perceived benefit that the donator should receive is the satisfaction of having participated in historically documenting the world in a particular time. The only other major satisfaction that they could receive is that they are able to click on a tag by their name/username and their lifelong contributions are lined up in an online gallery of the donated media. This gallery they can show proudly to everyone and in many cases receive some other benefits from elsewhere or use it as a scholarly portfolio while pursuing arts and related professions. Some sort of awards be given out to media donators who have done exceptional work.

  • Professionals might want to earn money from their media but after money is made they can donate what they didn’t use or plan to use, that which they think is not as good etc.. Instead of deleting images/video footage they can donate.
  • The whole idea is to attract the help of amateurs around the world who can now create media with the extremely cheap good quality profusion of computerized personal devices incorporating cameras/video etc. Who already earn a living doing some other work and who are not dependent on income through media making. Amateurs are casually making media and enjoying the process and the results but most of the results are enjoyed for immediate gratification only. Anywhere upto 95 percent or even more, of the media casually created usually gets deleted as it is of no perceived use to the creator in the long run over the remaining decades of their life. Any media creation takes lot of work and expenses but Casual media creation is perceived as no work and the process is enjoyed by casual media creators, it is perceived as totally free in the mind of the creator hence unfortunately majority of the casually created media is deleted, “This is the targeted media that we must campaign to save, the one that is being deleted”. All over the world the common man/world public should be educated from now on that they can donate this media before deleting it from their personal devices and they could be educated how to create usable documentary media for donation. Additionally there should be links to special “Tutorials” on how to create “usable” media for donation which can be used by secondary users (researchers, authors etc.) in various ways including research on what was captured and illustrating articles etc.. Usable media should avoid brand names, advertising and recognizable faces, specially of friends and family etc.. as then it may trigger having to take permission from people shown in the media for it to be usable now. Educational courses and institutions teaching Media creating; and personal device manufactures who incorporate media making functionality in their devices could one day carry these tutorials and educate public the world over that media can be donated to remain forever in a world archive. Like we see Warnings!!! in Cigarette Packs: If you smoke cigarettes you could get cancer; similar principal may one day be used by personal device manufacturers educating the buying public in the device manual and brochures that the media you create with this device can be donated to assist in documenting the world in present times so that it remains in a world archive for present and the future generations as a worthwhile historical media artifact documenting a particular location at a particular date and time from a particular angle etc. or documenting any other subject or human or animal behavior etc. All major networking sites around the world in local languages should carry links to these tutorials and media donation links in time with the development of media donation world culture.
  • A lot of seemingly repetitive media will be created around famous subjects/locations but this should not be discouraged as no 2 media/pictures are exactly the same, they are created in different time, days, weeks, decades etc. and from different angles and magnifications, subject focus etc. this continuous record will be invaluable to researchers in time centuries later and could also be packed off with future unmanned space explorers to educate aliens who may receive these capsules in the chance that they exist.

It looks like in the near future GPS will be embedded in all personal devices. And if GPS data along with date & time is mined from media donations then in the future special software could be developed that would make it possible to play-out/ lay-out the donated media in various requested sequences. For example Taj Mahal is a famous tourist site; maybe pictures are created here every second of the day. Lets say over a century later a researcher having mined GPS data available to him from pictures donated of Taj-Mahal, could request the computer software to lay out a sequence where pictures are laid out in a movie type flow encircling the Taj from 150 meters (using pictures taken at every foot in the circle identifiable by the GPS coordinates imbedded in the donated media), starting from the year 2000 and the circle completes 100 years later in the year 2100. In this requested computer output sequence, mined information from donated pictures of the Taj including time and date along with GPS coordinates available in that (as it is evident that most personal devices will soon have inbuilt GPS) would assist the computer in arranging the sequence in such a manner that the camera would travel from the front of the Taj, all the way around clockwise and come back to the front from the other side. The future computer software would make minor adjustments in magnifications for the Taj to appear the same size in all the pictures. If over time the historical site deteriorates, researchers can see how it took place over time requesting daily or weekly pictures to play-out in a movie, from a particular angel and particular distance etc.. Many more applications like this may be available with future software. In the example above about the Taj mahal, it could be seen every year, month, week, day, time of day etc. and from many different angles/magnifications etc. with the assistance of GPS coordinates. All this will be possible if data is mined from the donated media and properly electronically catalogued along with the image. And if donators are requested to fill in various detailed tags about each image that they donate through the designated upload link where they will also see various disclosures that they are indeed donating the media 100 percent and foregoing all their rights. “Emailed” donated images/media could be sent an auto-email-reply with a form to fill in creating all the various Tags that could apply to the donated picture/media and the auto-reply should contain the disclosure that the donator is indeed making 100 percent media donation and foregoing all rights.

  • Should inappropriate media be censored and deleted completely by administrators? Well, I think not. The media donated with brand names visible, copyrighted material, too many friends and family visible, sexually explicit material should be temporally removed from public’s view and should not be available to search in the present times. This material could be sent/dumped in an unsearchable database where it could lie for a few centuries and for a few generations to pass and after that when no one remembers who these people were the material could again be provided to public as a historical record from a previous time assisting researchers in human/historical studies and studies in human behaviors etc.. It may be noted here that even uploading unusable media requires effort and this effort may prove useful to researchers centuries later. The copyrighted material will be usable again then as copyright would have expired long time ago and in most cases the original work may have also been destroyed without a trace as most originals like books, paintings, newspapers etc. are made up of biodegradable material which perishes if it is not stored in museum like conditions.
  • There should be various TAGs that should be chosen and created in respect to each donated media so that they may assist the public at a later time to pull images/media from the archive. Various appropriate Tags should be created by original media donator who should be first provided with list of short Tags he could choose from that were created by others and were eventually standardized, additionally when he starts typing, to reduce effort, other types of tags may be suggested from the ones that were created by other users elsewhere. Some standard tags could be name/user name of donator, Subject/Location, magnification/seen from what distance, angle, Date, Time, context, normal view or description of “special event captured” (Like rioter throwing stones on police, people fighting, people shopping, building on fire, reading, praying, neglect, human rights violation etc.), atmosphere tags like, sunlit, sunshine, sunset, sunrise, raining, overcast/cloudy etc.. If news-making event is captured then NEWS-making Tag should be chosen and donator should be requested for little extra notes/comments why he thinks the media has captured news worthy event and be asked to describe the event in greater detail. The news-making tags could be patrolled by actual news companies etc. and could provide them/journalists/authors potential leads as to what stories that they can pursue now or at a later date in the pipeline; additionally news making tag could provide government agencies the opportunity to take positive corrective actions and for aid agencies to find people/projects to assist. Original donators should create tags and then on a later stage when the archive is being viewed by researchers and secondary users they should be also in a position to quickly add/create some more appropriate additional tags to assist future searchers. For example someone casually shoots a picture of an unknown butterfly sitting on an unknown flower, a zoologist/botanist viewing the picture in future could add butterfly name and flower name and scientific names etc. and create tags or/and notes/comments to go with the image. Some sort of “voluntary acknowledgement tags” could be created for secondary users who actually use the image in a wiki or outside publication, they can leave a tag/info/comment if they used the particular archived image and where, could leave a citation like detail of their article/publication, where the donated media was used by them. There should be tags to rate the donated image or video so that researching public can rate donated media on quality scale and could also leave additional educational comments in case they know more about the subject captured in the media which future researchers could follow up.

Computer software should supplement the above created/chosen tags with mined embedded technical information that is available embedded in today’s electronic media files, like camera used, aperture used, lens used, date and time of day; and very soon most personal media creating devises will have embedded GPS coordinates about where the media was made. All the above will help in Citizen reporting & Citizen documenting of the world in a particular, soon to be, historical time i.e. Citizen documenting of history, especially visual world history as it happens. There could be a Tutorial on how to make and use tags effectively, also showing how to view translated tags/event-description in a particular language. As an article is written or improved on a wiki or elsewhere, tags can help authors search for appropriate media/images/video etc. that can be used to supplement their work.

A Tag-search could give an output of a list of appropriate media that could by it-self be used as an online gallery; or best chosen images could be lined up in a gallery. Donators should be able to line up their lifelong media donations by clicking on the tag of their name, there should be a link here (and elsewhere too) to a “tutorial” on how to make their donated media more useful as searchable historical documents where donators could be taught how to go back in and improve each media piece already donated by them and already listed in the world archive; basically most possible improvements should relate around creating extra and much more effective tags for fairing better in searches and writing researched notes with references in the comment space under their listed archived media about what was captured in a particular picture/video donated by them, where, when and in what context etc.. (Once a media piece is donated and listed in the world archive, it should be possible to go in there and create more tags to supplements those already created for the piece, rate the piece on a quality scale, write comments about the piece in spaces provided. There should be tutorials to show how to do all this better)

  • As this is about building what will undoubtedly be a world archive, the scope is world wide and immense and lot of funding will be required to build and maintain such a resource. Lot of worldwide large scale funding drives will have to be organized targeting large donors annually. United Nations could be a good platform to request help from as this project is about the world as a whole. If someone provides funding from another country then we should set up a local office in that country so that tax-break forms can be issued to the donator so that they can get an income tax-break/incentive against donated funds which is valid in their particular country.
  • Huge Archive hence Limited & More Accurate Searches to Save Energy: As overtime a huge archive will come into existence it will take lot of electronic energy to make a search in the entire archive so some of the material could be deemed to be almost duplicate by volunteers and boxed together; and time periods could be boxed together etc. and when some one wants to search the archives first these boxes should show up and only if researchers want to search a particular box then only that box can be searched. In this way searches can be made more particular and electronic energy saved by making smaller/limited searches. Additionally media should be tagged properly and accurately along with research notes/comments on what was captured so that it can be located easily when needed. If required lot of un-usable or copyrighted or duplicate media could be boxed off in unsearchable boxes for the time being and could be made to surface again years, decades, centuries later when it is deemed to be usable again as copyright would have expired, persons shown are not living etc..
  • Digital Archival Storage Economy: It is hoped that this huge digital archive of high resolution images and video (as donators will be requested to upload in high resolution) will need smaller and smaller digital storage space as technology advances with time and most storage when not being searched will need no or little electrical energy; and very soon most electricity will be produced with cleaner technology hence building and searching this archive will be a relatively smaller drain on energy/world resources and wont be as harmful to the environment.
  • Archival Strategy: Strategically it might become necessary to have 2 or 3 copies of this world archive with only one that is connected for searches. All copies of the world archive should be located in the Free-World which is free from dictatorships and meddling by medieval religious institutions and regimes. Physical locations of all the copies of the world archives should be located in secret underground tunnels/caves away from earthquake zones where they will remain safe from bombings during future wars or/and purposely targeted sabotage. Obviously in the future when man does colonize other planets then copies could be located there in addition to the archives they will build about themselves.

Please forward this to concerned persons/departments for brain storming and fine tuning.

atulsnischal (talk) 02:59, 1 February 2010 (UTC)

Also posted here in continuation of my previous post: [11], [12], [13], [14], [15]

atulsnischal (talk) 00:29, 2 February 2010 (UTC)

Please don't crosspost. Also, summarise your points more concisely. You will probably not receive much of a response if you continue to shotgun these novels all over the place without addressing the responses you already received. LX (talk, contribs) 17:26, 2 February 2010 (UTC)

Category input box implementation suggested[edit]

We have by default input boxes which can be used either for galleries/articles ("go" button) or to search keywords by choosing the "search" button. Most everyone knows that. I would like to suggest a third button which leads to a category functionality, so after inserting the category name and chosse "category" you would get the ... <guess what> ... category :-) I'd appreciate that as it will save time and effort (type the word "category", add a colon [and click the "go" button]).

Maybe this would be only available upon request through the profile (preferences). BFN,--Mattes (talk) 19:57, 10 March 2010 (UTC)


Vector skin on Commons[edit]

Please see Commons:Village_pump#CatScan_link_in_Vector_skin (permalink). -- User:Docu at 08:47, 11 March 2010 (UTC)


Add Commons Commander to the user interface[edit]

Commons Commander (currently a test version) has a few features that make categorizing images at Commons easier. With a few bug fixes and additional features, I think it would be a great enhancement to the current user interface. -- User:Docu at 08:52, 11 March 2010 (UTC)

I mentioned Commons Commander on mw:Summer of Code 2010#Multimedia usability. -- User:Docu at 06:28, 13 March 2010 (UTC)

Edit or delete hidden categories with JS Gadget HotCat[edit]

I would like to ask to extent the functionality of HotCat to hidden categories, too. --Mattes (talk) 22:07, 12 March 2010 (UTC)

I think that was de-activated on purpose (e.g. see here). You might want to suggest at MediaWiki talk:Gadget-HotCat.js to make it a user configurable option. -- User:Docu at 00:04, 13 March 2010 (UTC)
The problem is that most hidden categories are added via templates, which HotCat simply can't handle. --The Evil IP address (talk) 08:51, 13 March 2010 (UTC)
This is a pain in maintaining Category:Non-empty disambiguation categories. --Foroa (talk) 09:01, 13 March 2010 (UTC)
That'd be cool if we could add something to our .js to activate it on all categories. Rocket000 (talk) 12:39, 14 March 2010 (UTC)

Special:NewFiles layout[edit]

View (older 48 | newer 48) (Show new files starting from 08:11, 13 March 2010)

instead of

View (newer 48 | older 48) (Show new files starting from 08:11, 13 March 2010)

seems the more logical to me. -- User:Docu at 08:11, 13 March 2010 (UTC)

Hmm, file #1 on the page is the newest, file #48 the oldest (of the top 48). #49 would be "below" that; so "older" means "down". I'd associate that with "right" if the choice is "left or right"; also, within a file row, files get older, so there's a theme that supports the current order. I think. --Magnus Manske (talk) 09:37, 17 March 2010 (UTC)
I noticed that are few other screens that do it in a similar way without a "(Show new files starting" link though.
What probably confuses me, especially when looking at files a few days back, is that "(Show new files starting from 08:11, 13 March 2010)" follows "older". Maybe adding a new line before "(Show new files starting from 08:11, 13 March 2010)" improves it. -- User:Docu at 09:45, 17 March 2010 (UTC)


Special:NewFiles filtering[edit]

Similar to tools:~azatoth/latest_files, it would be convenient, if there was a way to filter uploads from a few days/weeks ago. This would allow to check only files that are not in a given category or already tagged for deletion. -- User:Docu at 09:58, 20 March 2010 (UTC)

Search results page[edit]

Currently search result pages give quick links to filter the result by one of the following choices:

  • Content pages
  • Multimedia
  • Help and Project pages
  • Everything
  • Advanced

For Commons, the following selection seem more consistent with our content:

  • Multimedia
  • Categories
  • Help and Project pages
  • Everything
  • Advanced

Categories is currently missing as a quick selection.

At a later stage, it would be interesting to add options for some of the category based criteria: graphics, SVG, b&w photographs, paintings, graphics, taken with this or that camera etc. -- User:Docu at 09:15, 25 March 2010 (UTC)

Symbol support vote.svg Support I came here with the intention of asking for a "categories" button in the search engine, and this is exactly what you wrote! So I have nothing more to say! Teofilo (talk) 11:46, 10 April 2010 (UTC)
As an intermediate solution, I made an edit request at MediaWiki_talk:Searchmenu-exists#Category_link. -- User:Docu at 15:06, 13 April 2010 (UTC)
  • One more idea : when a category shows up in the search result page, the displayed information is a quote of the first line of the text it contains, with the number of words and Kilobytes in that text. However it would be convenient if the number of items (number of files and number of subcategories) was displayed. Users should have an idea on whether it is a "big" or a "small" category before clicking on the link. If it is a big category, they would probably have the feeling that there is a greater probability that they can find what they want there. Teofilo (talk) 12:28, 13 April 2010 (UTC)

Categorization workflow[edit]

The main categorization workflow is outlined at Commons:Categories#Categorization workflow. One can either work on Special:NewFiles, specific categories, un-categorized files or bot-categorized files.

It might be worth adding options to split Special:NewFiles into several larger groups, e.g.

  1. check if an image is likely to be an exterior view that should be added to location based categories,
    • find geographic categories for files that don't have geocoding or location-based categories yet
    • check all other files if they should be added to this workflow
  2. check if an image is a graphic/flag/painting etc. (based on categories, file format etc)
    • screen other files to check if they should be added there too.

With a few additional filtering/sort options one could use Special:NewFiles for a few days or weeks back to do some initial categorization. -- User:Docu at 09:45, 8 April 2010 (UTC)

upload tool[edit]

Please see Commons_talk:Tools/Commonist#.22could_not_login_to_commons.22. -- User:Docu at 17:16, 8 April 2010 (UTC)

Suggest substrings in search[edit]

Recently, I noted that my new Firefox which saves search terms that I write within formulars no longer solely finds results when my string is at the beginning of one former search term, but also when it's a substring of a former search term. For example, "New" would no longer solely match "New York", but also for example "Fashion in New York" and the like. It would be great to have this for our search function here. --The Evil IP address (talk) 13:55, 14 May 2010 (UTC)

Centralize translation process[edit]

Currently, as it stands, there are thousand of things that are in need of translating. However, all those texts are scattered all over Commons. We have things that are translated via LangSwitch templates, autotranslated templates, some customized system messages, some own system messages, things translated via JavaScript, some translations for license texts on Translatewiki and project pages that are translated on separate pages (probably, there's yet more). As you already see, it's pretty much a chaos, and thus it would be good to bring them all in one place. My first (and only realistical) idea would be to move them to the translatewiki:, which has shown that it's able to handle this with the translation of the license texts. For this, the following technical things would need to be done:

  1. Making it technically possible to change the default of a MediaWiki system message
  2. Having the possibility to save a system message on Translatewiki under a different name than it's used here on Commons, so we don't get into conflicts there
  3. Giving certain users (probably admins) the option to easily change the English default messages and to create new messages. Currently, only developers can do this, which would take too much time to request this all the time. This works with license texts, because they hardly need any update.
  4. Of course, the language recognition for anonymous users should work, because system messages otherwise always fall back to English.
  5. Using localized system messages within JavaScript. I think this is possible, but I'm not sure.
  6. For single-page usage, like within galleries or user pages, {{LangSwitch}} should be kept.

Let me know if certain points are unclear to you. --The Evil IP address (talk) 20:32, 15 May 2010 (UTC)

Displaying location on file description pages[edit]

File description page mock-up

One thing I liked about the mock-up is that it displays an image's location.

Currently this can be somewhat hidden if there is, e.g., a specific category for a building and 3 to 5 other categories, but no coordinates. -- User:Docu at 10:18, 17 May 2010 (UTC)

Summary box expansion[edit]

I would like to propose to expand the summary box with checkboxes:

  • text added/edited/removed
  • category added/edited/removed
  • templates added/edited/removed
  • syntax fixed (incl. templates)
  • "you name it" (discuss)

It would be important to have multiple choices per edit. This would

  • save time typing
  • harmonizes summaries

BFN, --Mattes (talk) 21:59, 1 June 2010 (UTC)

Display image in the editing mode[edit]

When editing a file, it would be useful to have the image displayed in a small size (120px) beneath/above/below the editing window. This would prevent an extra window to look on for the editor. This is already used in the global usage functionality. --Mattes (talk) 22:01, 1 June 2010 (UTC)

Sounds good. I think it should be possible to implement this by editing one of the interface messages. -- User:Docu at 06:30, 2 June 2010 (UTC)
I made a suggestion at MediaWiki talk:Previewnote. We could try that and see if users like it. -- User:Docu at 10:39, 2 June 2010 (UTC)
Done there (only users using the English interface will see it, obviously.) --Mormegil (talk) 13:34, 2 June 2010 (UTC)
This could also be done in the form of a gadget. Esby (talk) 13:51, 2 June 2010 (UTC)
Thanks, that was fast! --06:26, 3 June 2010 (UTC)
I don't like it, how can it be disabled? Multichill (talk) 18:04, 11 June 2010 (UTC)

Record of uploads performed using a bot or a tool[edit]

This issue has probably been mentioned before by other people, but just in case, I write it down again:

Most of uploads performed using a bot show up in the bot's contributions rather than those of the human being who ordered the upload.

It would be good to have a second gallery tool. The usual gallery tool would be "Gallery 1" filled with a user's manually uploaded pictures/files. And next to it, you would find a link to "User Gallery 2" that would contain the pictures uploaded using a bot.

A nearly similar issue concerns the pictures which are rotated by "rotate bot". When I upload tilted pictures, they first show up in my user gallery, but after rotatebot has put them straight, they disappear from my user gallery. I wish they would show at least in my "User gallery 2" or "user gallery 3"...

Teofilo (talk) 19:13, 11 June 2010 (UTC)


Wizard for searching (and categorizing) for coats of arms (and flags)[edit]

To make it easier to identify a coat of arms, it would be interesting to have some sort of tool that helps search various categories for similar ones.

I'm mentioning coats of arms as:

  1. there are quite a few images (also of the same coat of arms)
  2. the field has a fairly extensive set of categories
  3. the terminology is stable
  4. the terminology isn't necessarily accessible to everyone

If the tool is kept flexible, the same approach can be used for other fields.

It should be possible to pick from a list of attributes (ideally with a short description and an image) to identify elements of a specific coat of arms and find similar ones (and the same one).

To categorize new or existing images, the same definitions could be used by a similar tool.

To some extent this can already be done by CatScan or by browsing categories and using search, but it isn't exactly easy, nor can the standard tools combine them.

This might already be addressed by some of the Usability:Multimedia:Specifications with red links (e.g. Multilingual search, Category suggestion, Structured data, Filtered search, Batch asset metadata editing, Tags / category intersect / flattening). -- User:Docu at 11:21, 15 June 2010 (UTC)


I am suprised ! A Usability:Multimedia:Specification is being/has been written and the Commons community is not kept informed about it ? Or was that mentioned on the Village Pump during some of the days I was away ? How did you learn about it ? So is it better to keep writing here or to go to the talk page there ?
I tried searching both birds and leopards in heraldry with Catscan and it seems to work pretty well. So I am not sure I understand what else you would like to do. Teofilo (talk) 13:52, 22 June 2010 (UTC)
Which one resembles your sample? Did you click on each link to check? Did you need to change "leopard" to "lion"? Was it "rampant guardant" or just "rampant"? (etc.)
Personally, I tried to do it with ship funnels, but for various reasons these are unlikely to work soon. -- User:Docu at 14:09, 22 June 2010 (UTC)

TheGoodspeeds writes:[edit]

Today I downloaded two .jpg images of "J" Valves used on "Diving Cylinders". The Wikipedia article for "Diving Cylinders" mentioned them as being used prior to the 1970s and, in fact, they were very commononly used by all SCUBA divers in the 1950s, 60's, and well into and beyond the 1970s. It is my feeling that these images would improve the article on "Diving Cylinders" but do not know how to add them to the text myself - so have at it if you please. —Preceding unsigned comment added by TheGoodspeeds (talk • contribs) (UTC)

You might want to mention that at en:Talk:Diving cylinder. This page is for suggesting improvements to Wikimedia Commons (although your confusion certainly illustrates a deficiency in our processes). Powers (talk) 00:24, 24 July 2010 (UTC)