On Thu, May 15, 2014 at 10:58 PM, Erik Moeller erik@wikimedia.org wrote:
I would caution against a simplistic characterization of technology as a solution for what's inherently a complex socio-technical problem.
Please forgive a sentimental moment -- I am so happy to hear this clearly acknowledged by somebody in a position like yours. It is amazing to me how little attention this simple and obvious point has gotten in the years of discussion we have had on these topics! It seems that somehow, our discussions often seem to ignore that free speech, varying cultural values and editorial objectives, and individual rights have ALWAYS led to drama and problem, starting many centuries before Wikimedia ever came on the scene. It is really astonishing to me how much currency the "XYZ group is ethically challenged" frame gains in our discussion -- whether that group is "those guys at Commons," or "those guys at WMF," or whatever.
These problems are fundamentally difficult. Yes. That is the appropriate starting point for these discussions, and we should all keep our shared value to "assume good faith" in mind as we dig into the next layer.
All that said, I have long argued that there is a very simple solution to 90+% of the problem, and I'm curious whether this has ever been pursued.
A quick visit to stats.grok.se indicates that the "Search" feature of English Wikipedia was used 63 million times last month, while the "Search" feature of Commons was used 18,000 times in the same month. Of course, most of the Wikipedia searches were for text, but at least *some* of them would have involved the "multimedia" tab, which returns almost exactly the same results whether you are on Wikipedia or Wikimedia Commons. (If there's a way to quantify this, I don't know it.)
So my question is this: does a reader searching for images on Wikipedia automatically want the same kind of results as a reader searching for images on Wikimedia Commons? Does the Wikipedia reader *expect* to find images from a *related* site? I think the clear answer is "no." (Of course this could be tested, but to do so is beyond my resources.)
I think it is much more likely that a Wikipedia reader would expect to find those images *used in Wikipedia articles* than a massive collection of stuff that is somehow tangentially related to Wikipedia in a way that they don't fully understand.
So why on earth does the main "multimedia" search link on Wikipedia automatically return unused results from Commons to begin with? Is that really the right way to go?
I think a better way for the Wikipedia multimedia "Search" feature to work would be if it returned for "electric toothbrush" ONLY the images (from Wikipedia and from Commons) that are *actually* used in Wikipedia articles. There could also be a link there, "Would you like to search the media repository Wikimedia Commons, a sister site to Wikipedia?" But that could be a secondary step, thereby preventing lots of people (who found the electric toothbrush image they wanted already) from ever seeing the horrific masturbation photo.
Is there a good reason not to have the "Search" feature work this way? I'm not the most technical person, but this doesn't seem like a hugely challenging project. Could it be accomplished for (say) $10,000? And if so, is there a good reason not to do so, and eliminate a large portion of the astonished users who get surprised by something like a toothbrush being put to an unexpected use?
Pete [[User:Peteforsyth]]