David, Thanks for your thoughtful reply. Yes I think if I sat down I could probably type up about 20 things to go into the Idea Lab, but it's quite a small group of people that ever reads anything in there, and most of the readers are just quickly checking other ideas to make sure they have filled in their own idea correctly. One problem is that there is very little human interaction at all on Commons. I was personally very sad to see this user [1] leave the project, but I can't help but agree with him/her. Sometimes I wonder what I am doing there myself! I just checked, and I have only left messages on 21 other Commons user pages since 2006. Of those 21 users, I have personally met only 4 of them. I can safely say I do not use Commons as a social media website. On the subject of dirty metadata from GLAMS, I am afraid that we concentrate too much on metadata from GLAMs and I would like to see some project work on our own metadata. Lots of it is dirty, but there are some featured items in there that can be used to clean lots of the dirty stuff to a higher level of quality and we should do that. I am worried though when I read things from experienced Wikipedians along the lines of getting rid of the creator and institution templates, or even categories. [2] The whole power of Wikipedia lies in the low threshold of technical knowledge necessary for participation. We need to keep the entrance gate as wide open as possible. I see it as a sign of failure on the part of Commons that I rarely use the new&improved default upload wizard because it almost never fits my needs. I find it annoying that I need to use my own link to use the old uploader. When searching for new solutions, let's not throw away what we have, but try to incorporate innovative approaches on top of the technical landscape we already created. I don't expect newbies to understand and use categories, but it should be easier to identify and fix uncategorized items, and we need to lower thresholds of participation. I think your suggestion of uploading a new version of a djvu file shows an example of a high threshold (the participant must have djvu software to be able to update the file). In the case of the Sagrada Familia, there is a lot more necessary to consolidate our knowledge about it than just media files and text. We need improved search methods and associated navigational tooling, as well as a way to aggregate the number of projects involved in a quick overview. It's more useful if you can see that more than 100 projects carry information about it, than if it was in just 1 or two projects. Even Google just references the article in your default language. I would also like to see redirects and disambiguation pages exported to a central place, so search will improve across language boundaries. I've noticed that it's often spelling differences that keeps Google search better than Wikipedia at locating Wikipedia stuff. As far as the English-centric vision goes, I don't feel one way or another about it. English is as good a "lingua franca" as any other, as long as it's one basic language. I have said before I am cool with this being Chinese, as long as it's consistent. My main problem with Wikidata is the spotty coverage of labels, and if German was the default language and was completely filled in for all data item labels, I would be totally happy to put German in my Google settings from now on. I get frustrated when people get into arguments about the "database-ification of Wikipedia", as if that is a bad thing. It is unrealistic to think we can make everything machine-readable, just as it is unrealistic to think we can keep the reading machines outside the gates. We need to satisfy both our power users and our power editors, and the only way to do that is to keep low-thresholds of participation for both groups. That low threshold of participation is first and foremost for our casual mobile readers who become Wiki(p/m)edians when they upload one thing and go away for ever. We need them too, and though I don't want to force them into learning about our category structures, I do want to serve them a few tips so they can leave the default uploader and take a look around the landscape. The main thing is, we need to share more about where we are going as we get there, so users don't leave because they get tired of waiting. [1] https://commons.wikimedia.org/w/index.php?title=User:Boo-Boo_Baroo&oldi… [2] https://commons.wikimedia.org/wiki/Commons_talk:Wikidata_for_media_info Jane 2014-05-16 14:20 GMT+02:00, David Cuenca <dacuetu(a)gmail.com>om>: