On Fri, Aug 22, 2014 at 4:05 AM, Henning Schlottmann h.schlottmann@gmx.net wrote:
- The criticism isn't just about that -- it's about a large number of
mostly individually small issues. Generally, the idea that we effectively "munge" some of the metadata by displaying a machine-readable subset below the fold is viewed very negatively, because 1) it doesn't reflect all the available information, 2) it makes it harder for users to discover the File: page, and potentially edit it.
If it does not reflect the license information it is broken. We do not want
First, I think it's worthwhile in these discussions - in a context of a project where consensus is important - to remember that there are actually many different perspectives on Media Viewer in the community. Even in German Wikipedia, 72 community members voted _against_ disabling Media Viewer (more in absolute terms, incidentally, than voted for disabling it on English Wikipedia's RFC).
German Wikipedians have blogged about the fair amount of FUD that specifically surrounds licensing and attribution issues, pointing out that already, Media Viewer makes it far easier for re-users to consistently attribute authorship and license information, and provides more prominent copyright notices right below the image.
http://lieselsartikel.wordpress.com/2014/08/25/die-unwahrheiten-der-wmf-medi... http://erinnerungshort.de/blog/eine-lanze-fuer-den-medienbetrachter/
So if we care about consensus, why pretend these voices _in the community_ are irrelevant and wrong, rather than listening to their reasoned arguments?
And that's the current state. I think with a little bit of effort, we can deliver improvements that will give re-users concise hints that are far more effective at conveying the information hidden in walls of text today.
To be clear: The CC license text makes allowances for "reasonableness" - and in automated re-use, whether it's in Media Viewer or elsewhere, referring to the File: page in the absence of machine-readable data is an absolutely defensible, reasonable decision (and one which was vetted by WMF's legal team).
We certainly must work together to clean up the mess that is metadata on Commons, but to block all re-use that does not include an exact replica of the beautiful insanity of File: page templates until such a time that every piece of information contained therein is exactly represented in well-structured property/value pairs that can be predictably extracted is .. unreasonable.
In any case, many of your favorite examples of de.wp local uploads correctly display the author name and license now, e.g.: https://de.wikipedia.org/wiki/Olympiastadion_Athen#mediaviewer/Datei:2014_-_...
This is thanks to community members who worked with us to update the respective templates to add machine-readable data :). As I mentioned earlier, it's a totally fair criticism that our outreach efforts on this front were insufficient. There are still some templates left to be updated, and Gergo is preparing a more comprehensive cross-wiki assessment of templates in other languages/projects.
We do not want a default setting, that does not show extensive descriptions, map legends, image annotations and the like. All that is content we created for the readers. You must not block our readers from this content.
Again, the idea that software is not complete until it parses this:
[[:en:William Thomas Green Morton|William Thomas Green Morton]] {{ImageNoteEnd|id=1}} {{ImageNote|id=2|x=173|y=363|w=65|h=110|dimx=1600|dimy=1153|style=2}} [[:en:James Bogardus|James Bogardus]] {{ImageNoteEnd|id=2}} {{ImageNote|id=3|x=235|y=365|w=78|h=108|dimx=1600|dimy=1153|style=2}} [[:en:Samuel Colt|Samuel Colt]] {{ImageNoteEnd|id=3}} {{ImageNote|id=4|x=310|y=360|w=90|h=110|dimx=1600|dimy=1153|style=2}} [[:en:Cyrus Hall McCormick|Cyrus Hall McCormick]] {{ImageNoteEnd|id=4}} {{ImageNote|id=5|x=370|y=365|w=68|h=95|dimx=1600|dimy=1153|style=2}} [[:en:Joseph Saxton|Joseph Saxton]] {{ImageNoteEnd|id=5}} {{ImageNote|id=6|x=573|y=510|w=90|h=98|dimx=1600|dimy=1153|style=2}} [[:en:Charles Goodyear|Charles Goodyear]] {{ImageNoteEnd|id=6}} {{ImageNote|id=7|x=588|y=430|w=85|h=93|dimx=1600|dimy=1153|style=2}} [[:en:Peter Cooper|Peter Cooper]] {{ImageNoteEnd|id=7}} {{ImageNote|id=8|x=650|y=510|w=80|h=80|dimx=1600|dimy=1153|style=2}} [[:en:Jordan Lawrence Mott|Jordan Lawrence Mott]]
is unreasonable. If that's your standard of "broken" and "abysmal track record", your rhetoric is a bit exaggerated, but perhaps the reference to "black uniforms and heavy boots" in an earlier message was a dead giveaway.
Image annotations loaded via the community-created gadget are a beautiful hack (indeed one of my favorite gadgets of all time), and if the current implementation can be rendered in MediaViewer without negative performance impact and at reasonable cost, we'll get there. If not, let's migrate these annotations to a more supportable system in the long run and treat it as a File: page feature for now.
The more reasonable thing to ask for (which we've already implemented in the prototype) is a more prominent link to the File: page which encourages people to go down the rabbit hole if they want to, while giving them a quick image preview to begin with.
Erik