On 13/12/12 23:50, Lodewijk wrote:
Hi all,
just like last year, I am working on applying for recognition of Wiki
Loves Monuments as the World Largest Photography Competition. We should
be easily able to meet this record, but the evidence is always a bit
tricky to generate. Just like last year, I'd like to be able to send
them a full list of all participating images.
However, for other purposes we might need different definitions. I'd
like to suggest a few different definitions here.
== Retention statistics ==
For running retention statistics etc, I'd like to use wide definitions
of participating people and images: If the uploader *thought* he was
validly submitting, that is good enough. After all, here it is the
intention that counts. So this list should ideally include every image
that ever contained the Wiki Loves Monuments template. No matter if the
image was deleted, disqualified or if the country didn't submit finalists.
* No matter which country
* No matter the license / deleted or not
* No matter the monument template
* But within the time frame that the contest was ongoing (to have
'clear' stats for a full month of contest)
I have a good number of these, but not the full picture.
(would miss early uploads which were already deleted by the time the
script was running properly)
== Guinness ==
This will be the strictest definition because we cannot afford a single
mistake (that would make the process much more painful and long) and we
have easily beaten the record anyway so we can afford this.
For this purpose, we should generate a clean list of images that
definitely fulfill the criteria. This means among others that a
submission is for this definition only considered valid if it really
made a chance to win the international prize (similar to last year).
That means for example unfortunately that images from Belgium &
Luxembourg & South Tyrol cannot participate for this definition.
(Belgium & Luxembourg didn't submit finalists, so strictly speaking
those images never made any chance to win the international competition.
For Italy (including South Tyrol) there was a bit more complicated rule
due to local legislation that there had to be a permission from the
soprintendenza (if you want details, please ask the local organizers!) etc.
@WLM-IT: How can be determined if there was such permission or not?
* Participating in one of the 33 countries that
submitted nominees for
the finale.
* License OK
* Not deleted
* Monument template with existing monument ID in the database has to be
present (should there be exception countries to this?)
As much as I would like to all countries to be using monument id
templates, I don't think this can be enforced. Some countries didn't
have/use monument ids at all. Others have them only partially (eg. de).
pl *removed* the templates holding the monument id, and put it on the
category page added to the image.
It would be much easier IMHO to ask the organisers for the list of files
they sent to their jury. We can further filter deleted-since images.
For countries where all members used the Category: page directly (thus
no historic record of what they saw*), we can make an approximation.
What to do if a country first processed all the images and only later
checked if the monument id was valid/missing? I would still count it,
since it «battled» with the other images. And had it passed to the next
round, organisers would have probably tried to fix it even though it was
originally missing.
* Other than crawling the history of Wikimedia Commons pages.
In the case of
Italy, the monument has to be on the 'cleared list' (can someone provide
a complete cleared list as it was valid for the competition 2012?)
* Uploaded within the time frame for that country. Exception is mass
uploads from mobile platform and Flickr. I believe Platonides has a
whitelist for these bot uploaders?
I could look at the images I have registered as valid, with an upload
time later than the end.
Am I missing any other criteria that could invalidate
according to such
a strict definition? I'm skipping the 'activated email address' because
in practice we have reminded people of activating that if relevant on
several occasions.
I'd like to get started with this soon, so any quick feedback would be
welcome.
Kind regards,
Lodewijk
Regards