On the subject of digitizing, you might be interested by the project Wikimedia Deutschland just completed, that of digitizing a 16th century book for Wikisource. http://www.wikimedia.de/index.php?p=127 It might be an interesting thing to have the chapters and/or the Foundation vouch for the people proposing this kind of service. I believe not everyone is ready to give their pics/files whatever to people they've never seen and who just say "I have a scanner, I can do this for you". There are practical issues to take into consideration which go from conservations of the documents to returning them to their owner. Compiling a list of people interested by such a project though would be indeed a great idea, and commons probably is the place to start. Delphine -- ~notafish

On 20/06/06, Erik Moeller <eloquence(a)gmail.com> wrote: Er...you mean for the general public? I support seeking image donations for historical topics or specific holes in our coverage, but I don't know that we need more general amatuer my-dog-in-my-backyard,sunset-over-my-fence,my-street-my-school-my-car whatever photos. It seems to me the vetting system would have to discard 90% of what they were offered, so I rather question if it would be a worthwhile exercise. Also given how much trouble even Wikipedians have understanding that WM content can be used commercially, it might be very difficult to succinctly put across the extent to which a free license is free. Also, and this is a more general comment, there is still quite a disconnect between Commons and the projects, although it is definitely getting better as more projects switch to Commons-only and just time goes on. Sometimes there is a wealth of good pictures on Commons and no link on the WP article, or worse still no image at all. Sometimes (or often?) the beautiful FP nominations are not used in any Wikimedia project, anywhere!! Now I really wonder about that. We almost need another "project" of Commons people pushing the best content to all the WPs (they tend to be the most straightforward to illustrate for outsiders, plus maybe Wiktionary). So, I don't see that there is a good point to ask for image donations if we are not going to have people ensuring that the images will actually be used in some Wikimedia project. Same with the FlickrLickr content. Sitting around Commons is nice for us, but it could be so much better... and yes I know in some cases a Commons gallery is a natural extension of a W* page. not every image has to be a lead photo, obviously. Back to the topic of image donations. Not just image donations, actually - it would be really good if we could find out if there were some unis with, for example, recorded speech excerpts or phonemes or whatever, that they were willing to donate. Someone would have to convert them to OGG. This would be fantastic support for Wikibooks language-learning books as well as linguistics articles in /all/ language Wikipedias. I don't know how good our recordings are of clicks and trills, or uvuvular sounds, for example, are. It would be great to have native speakers' recordings. I just wonder how hard it is to convince unis to tighten their grip on the IP rights... cheers, Brianna

On 6/20/06, Brianna Laugher <brianna.laugher(a)gmail.com> wrote: Maybe. I think if we clearly define, in big friendly letters and with pretty icons, what we want and what we don't want, we could get a lot of good stuff. People aren't stupid. I suspect that there are quite a few amateur photos of celebrities, for instance. In any case, the campaign would be two-sided: an interface for instantly sharing media, and a contact form for negotiations. If we advertise this broadly, I think we could get contacts from places we don't expect. If we organize a "media donation" campaign well and make it part of the next fundraiser, I think it could lead to such people coming to us and talking to us. But I do think we should be open for contributions from the general public. Yes, a lot of images uploaded to Commons aren't used. But from my own checks of the FlickrLickr uploads, a lot of them _are_ used. FlickrLickr also automatically generates wikitext for galleries like this: http://commons.wikimedia.org/wiki/User:FlickrLickr/Slice_55 This is meant for easy copy and pasting of links into Wikipedia articles, and I encourage all "my" FlickrLickr reviewers to add images to articles wherever they can. If we use a similar, somewhat more refined process for a media donations campaign, I think it could work. Erik

On 21/06/06, Magnus Manske <magnus.manske(a)web.de> wrote: * Please forbid any image below say 800px on one side, and ALL mobile phone pictures. They are the worst! (Even if they still upload a small image, it might be fine, the important thing is to discourage it) * Please enforce the user to pick a better name than the default one. We have more than enough D23435646 and IMG345456456 and QZ23545657 image names already. * Please DON'T encourage sunsets. :( :( :( :( :( Maybe encourage pictures of their local area? at least they can ID that and we can figure out if we can use it or not. * Please ONLY allow this option: "I am the creator of this image, and hold its copyright." The other option will be such a pain and near impossible to verify, and trust me, there will be enough problems with their own work. * "The public domain (only recommended for images which are in the public domain already)." Why??? I would remove this disclaimer and list this first, from most-free to least-free (GFDL only). [OK not least free but biggest hassle.] Just say "You grant anyone the right to use this work for any purpose including commercial use, without any conditions, unless such conditions are required by law." That's scary (and true) enough. * CC-BY-SA: "...and to be released in turn under a free license" - change to "any derivatives of this work to be in turn released under this license" (to my knowledge you can't just choose something else!) * Please emphasise that licensing decisions cannot be undone in the future and if unsure, they should not contribute Instead of just having a big blank space, let's be smarter than MediaWiki ;) and have separate fields for the required info. "Description (Who or what is shown? Where is it located? Why is it interesting or important?)" Source should be self-made as I stated above "Date (when was the image taken?)" Author and Permission, no probs - but please don't put the {{license}} in the permission field, put it separately below and in the Permission field put a text statement like "X chose to release this image under the Y license as part of the Z image donation drive." I think we should limit uploads to something like 2 per person per day. If they're uploading dozens of images they should probably get an account (in fact, make that the alternative: do you have heaps of great images to contribute? Sign up to WM commons and upload as many as you want!) and also to discourage WM users from "anonymous" uploading (WM users are probably a significant majority of donors anyway, come fundraising time). * How are we going to communicate to them where if anywhere their image will be available? The email... eh. If we have to email them to find something out it's already too much hassle, dump it. (IMO) * Also categories... suggest five (somehow) based on their description text & CommonSense and get them to pick the most relevant, or just leave it up to the reviewers? BTW this doesn't mean I support this idea yet, I just like critiquing Magnus' UIs ;P Brianna

On 6/20/06, Brianna Laugher <brianna.laugher(a)gmail.com> wrote: If we use a FlickrLickr-like process, the renaming can be done as part of the review process. FlickrLickr already does a check on the filenames from Flickr, and if they match a simple regular expression like this, it forces the reviewer to rename the file. I tend to disagree here, but an e-mail address should be required if this option is selected. This depends really on whether images go directly to Commons, or through a review process. If they go through a review process first, someone uploading hundreds of images wouldn't be a problem. We would just have to explain to users that becoming part of the community is the best way to contribute. The point shouldn't be to clone the upload UI we have, but to provide a simplified process for people who don't know how to use Commons, something which we can advertise widely and broadly. Again, with a review process, the categories can be handled by the reviewers. Erik

Erik Moeller schrieb: Using FlickrLickr to filter donated images sounds like a great idea. I could upload them to the FlickrLickr database if there's an interface. After the safe-upload experience, I wouldn't want to store the image on the toolserver ;-) I haven't used Perl for some time now, and only very basic stuff back then. Two problems with this: * Each image in a zip file would basically get the same description. Otherwise, you'll have to include a text file with the descriptions, which is probably more hassle than uploading them one by one. * Zip files can be manipulated to grow into extreme amounts of nonsense data. Even a small zip fil could potentially result in hundreds of GB of uncompressed data. A potential attack vector. I have altered the text to show that option more prominently. I'd prefer to give the uploader some choice, though. Choice makes people think, whil a single checkbox just gets, well, checked ;-) Here's a thought: Let people mail their images, descriptions, and license to a central e-mail directly. Volunteers could then upload them after checking. And, we'd automatically get the senders e-mail address ;-) Magnus

Erik Moeller wrote: <snip> Agreed. MUAs like Apple Mail have this horrible habit of "embedding" attachments into the message so that the email ends up being (2n+1) MIME parts for each attachment: * Initial text * Attachment * Next bit of text * Attachment * Next bit of text etc. These sort of messages are *horrible* to process using something like OTRS. At least with a web interface, we get to decide what format things are arriving in... -- Alphax - http://en.wikipedia.org/wiki/User:Alphax Contributor to Wikipedia, the Free Encyclopedia "We make the internet not suck" - Jimbo Wales Public key: http://en.wikipedia.org/wiki/User:Alphax/OpenPGP