Commons talk:Project plan

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

As a driving force between Wikimagery, I'd like to get involved. Here are places where I either disagree with you or have something else to say.

  • The name. I simply don't like the name 'commons', because it does not give me any idea of what it contains. However, I understand your problems with using 'imagery' too, and don't see a good alternative at the moment - 'media' has an obvious clash of meaning. 'Wikimedia materials' might be a possibility.
  • 'It would hold...' - how about video? Does that fall under 'artistic works'? I would also like to extend 'music' to 'sound files'. The sound of (for example) a riding steam locomotive migth be interesting to have on the project.
  • On the other hand, I don't see why we would want to do what wikisource and wikiquote are already doing, so documents would be out, in my opinion.
  • Something that I would like to add is links to such material elsewhere on the web. Something like "On this and this URL there are a number of pictures about such-and-such under so-and-so a license. Plus descrioptions of the pictures, when necessary. When there is a lot of material somewhere, this seems better to me than just downloading, because that will necessarily mean a collection.

"All material in the commons would have to be licensed under one of several licenses, not necessarily the FDL, but all allowing at the very least free distribution and commercial use. For texts, modification rights would also be a requirement."

  • My idea would be to have the list of licenses in theory infinite, but have a kind of definition of an 'allowed' license. Also, to have in fact two definitions, one of wanted and one of allowed licenses. The wanted ones are the ones we prefer to have the material under, the allowed ones are the ones that we also allow. Preferred would roughly be GNU/FDL equivalent, allowed would be roughly be what you describe here. My idea would be:
    • Preferred: No significant restrictions beyond the following:
      • The author(s) must be mentioned
      • Derivative works must be under the same license
    • Allowed: may have the restrictions from preferred, plus:
      • Derivative works are not allowed (but resizing, putting in another image format, using the image in a webpage or book etc. should of course be allowed)
      • Sale of the image 'sec' is not allowed (but it is allowed to put it in a commercially sold other work)
      • Mild usage restrictions (e.g., not allowed to use it for spreading hate, not allowed to use it to defame the maker)
      • Notification of the maker is necessary before use (but permission is not necessary)
    • Not allowed: Have restrictions going beyond this, such as:
      • No commercial use
      • Not putting in databases
      • More serious usage restrictions (e.g., may only be used for articles on a certain subject, only educational use, only in news reporting...)

Criteria for inclusion

  • Looks good. I would just like to add a restriction for "YAP" (Yet Another Picture of)-cases. That is, if the file is similar to other files in the database, it counts as an argument against inclusion and for deletion. This also means that files with precise description have more inclusion potential than files with less precise description - "a nice flower" might be deleted whereas the same picture with the exact species indication would not even be considered for deletion.

Implementation - Upload process

  • The number of licenses would necessarily be a *lot* larger than this. I would therefore propose to have a pull-down list of 'other licenses' from which one can also choose. On the left side, I would replace CC-SA by CC-BY-SA, and add CC-BY as an option. Perhaps even a pull-down list here as well, but include only the "preferred" licenses here rather than all "allowed" ones.

The commons Wiki

  • "New uploads would show up both in the upload log of the wiki from which they were uploaded, and in the commons upload log." - I don't understand this. I would think that there is not a "wiki from which the pictures are uploaded". They are either uploaded locally (and then commons does not have to do with it) or to commons (and then it is not known which is the local wiki).
  • An important role of the 'local' community would also be to include pictures and picture collections for 'future use', that is, which are not connected to direct inclusion on some Wikimedia project.

I have not been going through the "desireable changes" and "single sign on" sections. These seem not to have to do with the creation of Commons per se. Furthermore, I would prefer to have the Commons (or whatever name) wiki as soon as possible, and not wait for the inter-wiki download and upload to be programmed first. Either create it as soon as possible now, or do so when the upload interface with license information has been finished, but certainly no longer. - Andre Engels 08:27, 4 May 2004 (UTC)


1) Name: I have added a rationale to the page.

2) I have added video.

3) Re: Wikisource, I have already provided a detailed rationale on why I think merging them is a good idea here. I believe that stating the status quo (it is currently an independent project) is not an argument.

4) I don't like collecting links. I think all materials of interest should be directly hosted on the commons. Harddisk space is cheap, and collections is what the commons is all about. There would be a "Gallery of dogs", an "Animal noises page", and so on.

5) Licenses: My key concern is compatibility with Wikimedia projects. That means that any license for files that the uploader himself has created must be FDL-compatible (i.e. CC-SA, FDL or PD, possibly others). For files from other sources, we may tolerate other licenses like the ones you suggest, but compatibility would always be preferred. The individual projects could set their own policies as to what they want to allow beyond that.

6) Criteria for inclusion: I'm against excluding redundant images except for egregious cases. I see no reason to do it -- users of our materials should have a wide range of choices of which picture they want to use. You want to make a logo from a flower picture? Then you may need one which is shown from the top. Do you care about the insides of a car, or about its looks? Is there a less compromising picture of this or that politician? In fact one of my motivations for creating the commons is to have a place to move redundant image galleries out of Wikipedia and into the Commons.

7) Upload process: The user will not have to log into the commons. All new uploads will go to the commons by default, unless a "local upload" is specifically requested. This is important for usability. Furthermore, all existing freely licensed (tagged with {{msg:GFDL}} etc.) images will be automatically moved to the commons, and for others there will be a "Move this to the commons" pushbutton; there is no need for bots (sorry, I know you like using your bot! ;-).

8) When to start: I think just setting up a wiki would be a bad idea. Some Wikimedia projects (*cough* Wiktionary, Simple English *cough*) have suffered from a lack of advance planning. I don't want a wiki that is just of interest to a few people and which will then later have to be ported over to a more complex infrastructure, I want to change the way in which media files are handled on all of Wikimedia. And if enough people help, I want to address one key problem thas has plagued Wikimedia for years - single sign-on. I estimate that we can get all of this done in 3 months, and I believe that the effort is very much worth it.--Eloquence 04:13, 5 May 2004 (UTC)


1) That meaning of the word "commons" I did not know, and actually my dictionary (Penguin Concise English Dictionary, Revised Edition, 1969) does not know it either, describing the word as "the common people below the nobles in rank; the commonalty; members of the House of Commons; allowance of food, rations"

3) I still find it a bad idea. To me, they are very different things. You don't put a text on wikisource with the idea "it can be taken out for a wikipedia page" (or whatever). Either you put it on the page immediately, or you are against putting it on the page generally. On the other hand, pictures and media would be put on Commons (or however you call it, still not agreeing) with exactly that idea in mind.

4) Hard disk space may be cheap, that does not mean that the time of our contributors is. In the time that one uploads one image, one might well be able to describe ten. Some sites have thousands of pictures that we would like to have. It's a big difference to force someone to upload them all or to write a short description of what is on them.

6) I think you misunderstood me here. I wasn't arguing for having only 5 pictures per subject. I was arguing about the 200th dog picture or the 1000th flower picture. And I also was not arguing for just removing them, but for removing them if they brought nothing new. I don't want 200 'dog' pictures, but if you look at the various races of dog and the various activities that dogs can be doing, it would be easy to find 100 categories of dog, each of which I would not mind at all having 5 pictures in. I don't want 50 times George Bush smiling for an official portrait with the American flag in the background. But I don't mind having 100 or 200 times George Bush, provided it's different categories - this one is a portrait, that one shows Bush playing golf, the third is Bush in military clothing in Iraq, the fourth is Bush speaking before the House of Representatives, etcetera.

7) I wasn't thinking of the bot here, just of the way things are going. You seem to have far-reaching ideas of how to change the Wikipedia software for it, and I wish you the best for it. I am trying to get it working with things as close as possible to what we have.

8) I guess we have a totally different idea of what this is for then. My idea was to have some deposit for pictures, where people can then find material for their own Wikimedia and other projects. I don't see what's bad about that. And I think it is of interest of a lot more than "a few people". I don't see the probloem with just doing that, and get your infrastructure later. We created different-language Wikipedias without interwiki links or single sign-on. I very much want this project. All those other changes you are trying for are nice,k I am vin favor of them, but I don't see why a project that many people have asked for many times should be delayed once more so that it can be implemented. Look at the projects you stated. Their initial failure was not because they were using just Wikipedia software, and their later success was not because of software changes. The proposals you are making are useful. But so is Wikicommons (not going to repeat my remark) on its own. There's no reason to wait. - Andre Engels 09:08, 5 May 2004 (UTC)

1) Try a modern dictionary. Even to people who don't know the concept, I think it makes for a beautiful story that is easy to understand for non-geeks. I find it very important to concentrate on the idea of the common good, because what Wikimedia is doing is truly revolutionary and needs to be emphasized. If you can think of a term that better encapsulates this sharing of ideas than commons, feel free to suggest it, but I doubt that there is one.

3) No, you may not want to put the complete text on Wikipedia, but you may want to transclude the first paragraph of the Declaration of Independence. Using hidden labels in the source text, these could be reliably referenced. I believe this application to be very close to what we want to do with images. Furthermore, my hope is that the larger commons community will enrich the currently small Wikisource project, which will in turn drive further refinement of policy, software etc. I don't see any negative sides to merging them. However, I believe this should be decided in a Wikimedia wide poll, with additional weight given to votes of Wikisource users.

4) I don't understand what you are talking about. Force someone? Who is talking about forcing anyone to do anything? People will do what they want to do, and if they want to upload 1000 images, they should be allowed to do so.

6) I can agree with that in principle, as long as the criteria are reasonably relaxed.

7) + 8) What you seem to want is a limited media repository that would be a second place to look for images besides the local wiki. So you want to set up a quick wiki, and people will start making policies and working around deficiencies in the software, they will tag and describe images in ways that will later have to be changed again, there will be duplicate files and descriptions, and what is more, we might miss our best chance yet to build a single login system. And all because you don't want to wait 3 months to do it properly? You will have to understand that I would strongly oppose such an approach.

You want certain functionality. I want certain functionality. The natural response to that is to cooperate to get both, not to say "I want my functionality now and I don't care about your functionality". There is nothing wrong with delays that are caused by people working on creating something useful. I need your help and other people's to make this work, otherwise the project will get delayed without people working on it, or perhaps never be implemented at all.--Eloquence 09:36, 5 May 2004 (UTC)

3) Then still I see nothing won to have it on the commons site. And I do see negative sides. Different cultures might clash. Search functionality gets less.

4) When I say 'force', I mean force in the sense of "if you want to make this 200-image repository available through wikicommons, you will have to upload 200 images". Of course we're not forcing anyone to do anything specific, but giving certain possibilities and not others restricts their options, and thus is 'forcing' them. Some things we are forcing already, others we are trying hard not to. We are forcing people to upload an image if they want to use it. We are not forcing people to check whether pages exist when they link to it. We are semi-forcing them to learn Wiki-syntax.


"people will start making policies" - I think that's a good thing
"working around deficiencies in the software" - Which deficiencies? We are using the software already on many projects. All deficiencies are already worked around dozens of times. What's wrong about doing so once more?
"they will tag and describe images in ways that will later have to be changed again," - Now there is an argument. If there will indeed be a method of tagging an image upon upload, you might convince me that I'd better wait for that. But the rest? No.
"there will be duplicate files and descriptions" - No more than there will be if we start later
"and what is more, we might miss our best chance yet to build a single login system" - I don't see why it would be hindering it in any way.
"And all because you don't want to wait 3 months to do it properly?" - I have already waited 3 months, and I don't see any reason why your method would be proper and mine not.

You want certain functionality. I want certain functionality. The natural response to that is to cooperate to get both, not to say "I want my functionality now and I don't care about your functionality".

You want certain functionality. I want certain functionality. The natural response to that is to work on both, not to say "I want my functionality and you will have to wait with yours until mine is ready".

There is nothing wrong with delays that are caused by people working on creating something useful.

I think there is something wrong with delaying something useful just because it is not perfect.

I need your help and other people's to make this work, otherwise the project will get delayed without people working on it, or perhaps never be implemented at all.

And I do want to help. But I think my help is better used by setting up the thing now, and giving me a chance to start up a database and a userbase for Wikipedia Commons, where then your sweeping changes can be fit in, than by sitting and waiting until your software changes are ready. It's not that I don't like your plans. It's not that I don't want to work on them. But I don't want to see my pet project delayed just because yours is not ready yet. All I am asking for is for MediaWiki number 301 to be set up and get a domain connected to it. You want certain things from it, you tell me and I try to take it into account. But you basically said that the thing is not worth starting up until your programming is ready. From that I have to conclude that you are not interested in my plans. Which of course you are entitled to. But don't expect me to be interested in yours either. - Andre Engels 16:41, 5 May 2004 (UTC)

If the commons were able to store templates, this should be very nice with the extension of template syntax:

Lets say we have the following population-romania template on the commons:

population=22 271 839 | median_age=35.4 | growth_rate=-0.21 | birth_rate=10.79 | death_rate=12.25 | migration=-0.6 | sex_ratio=0.95 | life_ expectancy=70.62

We could then use this template in all wikipedia languages : {{PopulationTable|{{population-romania}}}} This would save us the work to manually update the data in every wiki. Ske

  1. Different Wikipedias have a different approach upon the usage-rights of pictures. So does the English Wikipedia allow fair use media items whereas the German counterpart has a strict GPL-compatible licence policy. The question is if the Wikimedia Commons induce a one size fits all policy for media items or if the different Wikipedias still may maintain there different approaches.
  2. From reading this page I got he impression, that the uploader must choose a licence for the medium uploaded and that one could do so by selecting the appropriate licence from a list. In case there still are different usage-right policies, I would suggest that the list also contains information, where an item may be used (e.g. "GNU Free Documentation Licence - all Wikipedias" | "fair use - english Wikipedia"). Btw, I would order such a list such that the least restrictive licence is on top and the most restrictive one which we still accept is on the bottom, of course defaulting to GFDL.
  3. I have some problem understanding the Mild usage restrictions example. In the German Wikipedia, there is an ongoing struggle about coats of arms. In Germany, the municipality reserves the right on the coat of arms and often uses copyright to restrict its use heavily - I understand this as not being covered by "mild restrictions" anymore. In Austria, reproduction is allowed given that the symbols are not used to defame the municipality. However, usage within commercially available products may require the authorization of the owner. Is this still a "mild restriction"? Moreover, usage of official symbols are often also regulated by law, how to deal with them?

--nd 21:54, 13 May 2004 (UTC)

The commons would not include fair use images, but it might include images which are non-free under one specific jurisdiction (e.g. German) but public domain under others. The commons exists in parallel to the local image repository of each project, so that the German Wikipedia or the English ones can have their own policies of what type of licensing to allow in addition to the ones allowed by the commons. When you upload an image, the default mode is to upload to the commons. You can request a local upload if the commons does not allow a certain license, but the wiki you are working on does. In the screenshots, if you clicked the "English Wikipedia" tab at the top, you would go to the upload view for en:, where additional licenses could be available (specifically, en: allows fair use, while the commons does not).--Eloquence
At #3: The 'not use for defamation' is typically what I thought of when I talked about 'mild usage restrictions'. However, 'permission needed before use in commercial products' is in my opinion equivalent to 'commercial use forbidden', and would thus not be ok. - Andre Engels 10:38, 21 May 2004 (UTC)

Do remember that there is already a problem of lots of different copyleft, open source and other licenses. A commons (free for all to use) project has to try to solve this key problem. That makes a single license for the "I'm the original creator" choice a bad idea - it forces limitation to only one license type and locks out many other good and useful licenses. What's really needed to create a true commons is a large list of licenses, grouped by general properties and followed by a public domain dedication, so you can release under lots of licenses just before dedicating to the public domain. There are some questions about whether PD dedication is viable in some places so it's cleaner to use lots of licenses just before as a fallback in case PD fails. It's also easier to understand, yes, I'm GFDL so I can use this because it has a GFDL license and that also helps with searching for works with a specific license type. I don't have strong views on what should be the most restrictive license type but a not for profit or with credit requirement beats the very restrictive copyleft licenses when it comes to breadth of possible users. Ruling out the less restrictive attribution type effectively eliminates the possibility of the project really being a commons - it rules out far too many possible reuser types. Ideally I'd like to see something like this heirarchy:

  • Any license which doesn't impose legal liability on me, listed below or added later
    • subtype for with attribution, including licenses added later
      • sub-sub-type for copyleft (which require attribution inherently) including those added later
        • for those who don't want to accept generic and added later, a list of individual copyleft licenses for individual selection
      • list of lots of other attribution required licenses, in any groupings which exist
    • subtype for category of use restrictions (non-commercial, educational only etc)
      • sub-sub types as required.
  • dedication to public domain.

Ideally everyone would choose the top two levels, all licenses which don't add legal liability plus PD - then there would be a really strong commons which almost anyone could use. Jamesday 01:01, 28 May 2004 (UTC)


I think Wikimedia Commons (I certainly do admire the name - and hope it does become "Commons") is a good chance to implement a more sensible naming system for uploaded files. Thre are so many files which have some redundant name "bush031453.jpg" simply because this is the name it was downloaded by user in. Commons would be a good chance to have true, and logical, naming systems "George_Bush_playing_golf.jpg" - but this could cause problems considering this project transcends language... A further problem are languages which do not employ this European alphabet - Japanese, Mandarin, Hebrew etc. etc. Does anyone have any ideas on how files should be named in a common and logical manner (thus making it easy to find an image of something) without causing translanguage problems? Perhaps a single image could have descriptions in many languages - on which to be located - or categorised to a definable extent.

I would also like to have an easier mechanism for replacing an image for a better one. The example I shall use is a painting - if I see a poor quality image of the Mona Lisa - with faded colours (caused by bad transfer) and odd scanning lines and a resolution of 200x100 - I will naturally want to replace it. But without admin rights it is quite a fiendish task. Secondly - would we have duplicate versions of the Mona Lisa from an incrredibly large size (2MB, for example) - which would be used by users for examination of brush strokes etc - and also smaller sizes (a 100kb version for the casual user). Would there be any way to generate a thumbnail or (100kb) version from the 2MB file - hence only needing one high resolution version on the database from which smaller versions may be generated serverside and presumably kept in the server cache for 24 hours. Would there be problems with this such as high workloads on servers? --OldakQuill 11:26, 2 Jun 2004 (UTC)

Your remark about the naming system is a good one, at the least it would be good to allow people to choose their own name for the upload rather than automatically take the name under which it is on their computer. Even better would be a real naming scheme, but I also have no idea what it should look like.
Unless things have changed, uploading a new version is not too hard, but too easy - just upload the new one under the same name. The bad thing about this is that it runs the risk that people upload something else with the same name, not realizing that that name is already in use.
Your third proposal looks nice, but I have no idea whether it is technically possible, so I will not further comment on it. - Andre Engels 12:20, 2 Jun 2004 (UTC)
About the naming: I believe the user uploading the image to the commons the first time should have the "right" to name it whatever he'd like, as long as it's logical in that language. Another option is having a "naming" table in the database, so you can actually give several names to the same picture and let the database take care of the original linking to the real filesystem name. Let's say for instance you upload (accidentally) "Bush3232.jpg", but then you can add names as you'd like: "Bush playing golf" or "Bush jugando golf" or "Bush spiller golf" depending on your language. Then you can link it with File:Bush playing golf without even having to think about the suffix. What about that? --Vikingstad 01:15, 5 Jun 2004 (UTC)
Yes, very good suggestions. Viking - could an image have more than one name in a single language. A plant for example - the scientific latinate name, and the common name. (Homo sapien & person/human). What would I do if I posted a picture of Foxglove the plant and a picture of a 19th Century cartoon called Foxglove featuring a fox wearing gloves was already there? Perhaps a thorough, planned categorisation scheme would be necessary to find the pictures one wants?--OldakQuill 01:54, 5 Jun 2004 (UTC)
This could be handled by making sure all naming-entries are unique. That way a warning would just appear when you try to apply an already designated name. Categorisation would be very useful if you'd want to browse the commons library, and search within certain categories and sub-categories (as we now have on MediaWiki 1.3). --Vikingstad 18:45, 5 Jun 2004 (UTC)


this is a necessary feature. some implementation details i don't agree with. per-language image: and image_talk: needed. Badanedwa 00:30, 5 Jun 2004 (UTC)


I've been in favor of this idea from the beginning, and I couldn't agree more for media files, that a feature like this is vital. However, I'm vehemently opposed to what still seems like an attempt at a hostile takeover of Wikisource as a part of this. I see Wikisource and the Commons as projects that complement each other. I don't think it really makes much sense to put texts, which are more something that would be linked to (as has been done with the Wikipedia and Wikisource pages on resolutions of the UN Security Council), rather than inserting in the way that you would an image. And on top of that, if a sample of text does need to be used from a Wikisource document, why the heck can't it just be copied? It's hardly the same as for images, and it's the one thing that really concerns me about this project so far. - ambivalenthysteria (NSI)


This idea has been posted for several months now. The next step to take is to formulate a roadmap for the project and get more ideas on board. I cannot simply do this on my own. If I get several responses to this we should start.-- 13:16, 6 Jul 2004 (UTC)

I have been saying we should start for a few months now, but alas... It seems I do not have the power to actually get something done here. Which would not be so bad if others would do it instead, but this project seems to have been, as we say in the Netherlands "hugged to death". - Andre Engels 21:41, 9 Jul 2004 (UTC)

I ran across en:Category:Wikipedia_images which we could use as a starting point. -- 23:21, 22 Jul 2004 (UTC)

My opinion on various points Licencing - We should have a prespecifed set of licences all of which meet the following criteria,

  • Do not limit use
  • Do not prevent modification

As for Inclusion/YAP, if something is fairly common then I agree. However for more unusual things a number of photos are fine I think. --Imran 00:08, 23 Jul 2004 (UTC)

mockup screens[edit]

You should allow mutliple licences on the left (if I'm the creator)...

It should be also possible changing the name of the uploaded file (e.g. a file named "Picture 2.jpg" should be stored as "Picture.jpg" --Fedi 18:38, 8 Nov 2004 (UTC)

Why haven't these mockups been implemented? They are much better for the beginner than what we have now! --Komencanto 7 July 2005 11:27 (UTC)

SVG library[edit]

It occurred to me that the Wikimedia commons could be very useful as a general SVG snippet library. Say you want to draw a diagram with some humans, some computers and some interfaces -- you check the SVG directory and download the files, edit them together in your favorite vector graphics application. Such a free SVG clipart library does not seem to exist yet and would be really cool even outside Wikimedia. Unfortunately MediaWiki doesn't have SVG support (i.e. it can't render SVG to bitmaps for older browsers), but the Wikisophia people seem to be working on this. Just thought I'd toss this idea out so we don't forget about it later.--Eloquence

First steps[edit]

I think a good next step in this plan would be to allow inclusion of WikiCommons pictures in Wikipedia (and other Mediawiki projects). This seems easier to implement than cross-project uploading, and would greatly enhance the usability of WikiCommons. - Andre Engels 08:49, 10 Sep 2004 (UTC)

Prior Art?[edit]

Currently there is a lot of talk about patents, software or otherwise. Do you think there is a place in WikiCommons for people to locate and document prior art for various ideas, or is that too different? Perhaps we'd need a WikiIdeas project for it ;). --ShaunMacPherson 15:15, 21 Sep 2004 (UTC)

I would not see how this could be a task for, or helped by WikiCommons. I simply see no overlap or large similarity between this and the task set for WikiCommons, being to provide material for use in the other Wikimedia projects. - Andre Engels 18:01, 21 Sep 2004 (UTC)

NPOV on history, please: no political propaganda[edit]

In the Naming section, it is unnecessary to use inflammatory language or inject PDS viewpoints taught during the DDRzeit into a discussion on use of the term commons to denote an area owned or used by all in common.

Instead of

In Germany, after the Middle Ages, the local commons (Allmende) in most towns and cities were confiscated by the feudal rulers of the regions, which was one of the motivations behind the German Bauernkrieg (war of the farmers against their rulers). The farmers were brutally suppressed, and the commons practically ceased to exist. To this day, descendants of the aristocrats who profited from robbing and murdering the farmers are among the richest families in Germany.

The concept of dwindling Common Land (which is what you really want to get across) could be stated more concisely and with a Neutral Point Of View as

Throughout the ages in societies around the world, especially in newly-settled areas, there were always certain lands not under the control of any invidual, but shared by all for grazing livestock, firewood, fishing, etc. As time went by, those resources often were gradually encroached by futher settlement or appropriated by local strongmen, leaving few or none remaining in those communities for the free use of others today.

StanZegel 14:42, 29 Oct 2004 (UTC)

Promoting Commons[edit]

  • on your local [MediaWiki:Uploadtext] template put the text: STOP! If you think this file will be useful for other projects/Wikipedias as well, upload it to Commons instead!
  • make people aware that if they copy a file from their Wikipedia to Commons, then as long as the filename stays unchanged, the original file can be deleted an no changes in the existing articles are necessary.
  • see also Commons:Village_pump#Interwiki_links

Discussion on video policy, please comment.[edit]

In response to some new developments, I'd like comment on a possible update to the currently-outdated meta:Video policy. Please comment on the discussion here, as this involves the Commons. Grendelkhan 13:06, 3 Jan 2005 (UTC)

Proposal - Commons Image Identification Service[edit]

There is an open source program called imgseeknet written by Ricardo Niederberger Cabral (see here) it is a Web client version of imgseek (here). From the website Riccardo describes imgseek as: "imgSeek is a photo collection manager and viewer with content-based search and many other features. The query is expressed either as a rough sketch painted by the user or as another image you supply (or an image in your collection). The searching algorithm makes use of multiresolution wavelet decomposition of the query and database images."

What's Interesting About This Program[edit]

  • You can submit an image to the precompiled database of the library of your images, and it will return similar images.
  • In fact you can even draw a rough picture of the image and it will return matches from the library.
  • There is a Web version which you can instruct to develop the database of your images. You can then query from a browser.

The Relevance Here[edit]

  1. The database creation element of imgseeknet could be run against the entire library of WikiMedia Commons
  2. The relatively small database can then be hosted at WikiMedia (or by me for now) and kept up to date.
  3. An alteration of the imgseeknet Web Interface could be set up with the image database on the back end.
  4. Users could then upload images that they want identified.
  5. Imgseeknet could return matches.
  6. The best match once selected could open the WikiMedia Commons entry for that image.
  7. If they want, users can select the Wikipedia URL for the image, from the Wikimedia Commons description for the entry.

Some Additional Points with Requests[edit]

  • Riccardo the programmer is encouraging the idea.
  • I'm willing to host the database and Web interface on my own space and bandwidth for testing purposes.
  • I need help on how to get the WikiMedia Commons entries to be displayed when an image match is chosen among the several possibilities offered by imgseeknet from the database of the Commons images. This is a programming issue. I don't currently program.
  • Also I or someone need access to the whole Commons database of images.
  • I need advice on how to approach the WikiMedia Foundation for them to grant me access to develop and test this new Search Interface.
  • Finally a key feature of this Search Ability should be that it is doable from a Mobile Phone. That is best left for later development but should be a goal as providing an interesting use for camera cell phones.
  • Any guidance is humbly accepted...--Socrtwo 23:18, 10 August 2005 (UTC)

Fair use commons[edit]

I can't find anyplace else to post this,

There is need for a fair use commons. While the fair use of photos may be dependant on country of use and its use therein, the use of fair use of symbols, logos, albumcovers, bookcovers and promophotos is of a different nature because they are created with the intention of being spread. In the current situation, most Wikipedias allow uploads of logos and symbols. It must be more efficient to have a commons storage for that kind of images.

Another advantage of having a common place for albumcovers is that more people could get involved in making sure the albumcovers are of a size and resolution which makes them eligible for fair use.

There are two possible solutions

  1. Allow fair use on mentions items on Wikimedia Commons.
  2. Create another Wikimedia Project with the purpose of dealing with mentioned material.

Thanks -- Fred Chess 18:21, 15 February 2006 (UTC)

Fair use is context-dependent. As such, the very idea of a fair use repository is self-contradictory (let's not even speak of the oxymoron in the words "fair use commons"). Individual Wikimedia projects and language editions develop their fair use policies depending on the nature of their content and the predominant jurisdictions applying to readers and editors. In Wikinews, we have even taken the drastic step of starting new language editions without an upload capability, and only enabling that capability when the respective community has developed a fair use policy. Fair use in Wikinews is different from fair use in Wikipedia (photos from competing news organizations are taboo, for example), and fair use in the English Wikinews is different from fair use in the Polish Wikinews, while the German Wikinews does not recognize it at all.
That is not to say that it wouldn't be possible to technically improve access. If you could logically model and match the different fair use policies, you could provide an automatic interwiki image transclusion mechanism for loading pictures from another wiki if they are allowed in the one you want to load them into. This would, however, be a major technical challenge. Given Wikimedia's mission of creating free content works, one could also argue that making it a little obscure to upload fair use content is a good thing.--Eloquence 23:56, 21 February 2006 (UTC)

Secure server?[edit]

Man, it really stinks that there is no way people using Hughesnet or any other ISP implementing rolling addresses can upload images here or on Wikipedia. I have images that would go perfectly with many articles, but have no way to upload them. What gives? ( 09:23, 10 August 2007 (UTC))

This is possible the least useful place to ask that question. If you ask at the Commons:Village pump, I think you will get better response. / Fred J 10:24, 10 August 2007 (UTC)