I have to say that I have a complete disconnect with this statement from Lodewijk:
"professional photographers ... can process actually thousands of images per hour - so if they are willing to spend a few hours and are helped by the right equipment (jury tool!) they can process a /lot/." for screening entries for the jury.
Viewing 1 image every 3 seconds would be 1200 images per hour, but with the inevitable break to keep your eyes from glazing over and your mind from wondering, it's probably 1,000 per hour on a practical basis. Somebody yesterday (Mexico?) gave some numbers that suggested more like 100 per hour, a factor of 10 difference.
Given that there is some computer/connection dependent time involved in displaying a photo and recording the result, I just don't see how viewing a photo for 3 seconds gives any real basis for screening.
There are quite obviously some photos that can be screened out very quickly. I'd say 10% or maybe even 20% of the photos I saw last year, I wouldn't consider putting in Wikipedia, e.g. highly over-exposed, or very dark, or my pet peeve - signs - "This way to the Statue of Liberty". These can be screened out quickly, but others will take some time. I'd also say that about 20% of the photos in the US last year were quite good, maybe even "approaching professional quality." Running by these in 3 seconds or less, it would be impossible to do a quality job of screening. Maybe you could identify a certain type of lighting, or an obvious wow factor on many, but I'd think that at least half of deserving photos would be left behind.
For those countries with more than 10,000 photos to screen the difference is huge. 10 hours of screening (should be pretty easy to handle) vs. 100 hours (a real organizing challenge and time commitment).
BTW, I don't think the jury tool has been designed to process over 10,000 photos.
Pete User:Smallbones
Hi Pete,
I know you disagree, and that is why I said this is my experience in the Netherlands. I remember a very similar discussion last year. I'll let everyone make up their own mind. But I was told by the people from our jury that 1 second per photo (average 1,5 - which would mean 2400/h), with the right setup (i.e. pre-loaded images, or very fast connection, and simple-one-click-dismissal), is actually doable for most photos - because many are simply too boring or qualitatively bad and can be dismissed as 'top-50 material' from 10.000+ photos. Others even just scroll through heaps of photos. Also, please note I was talking about people who go through huge amounts of photos on a professional basis. Obviously not everyone can do this so fast. Whether that is the kind of judging you like, is another question - but I think it is possible to go that quickly. But this also means you probably want multiple people looking at the photos. Because if you do it so quickly you're bound to make mistakes; having multiple people independently reviewing should reduce this quite a lot.
Again: every country will take their own approach and that is totally fine.
Best, Lodewijk
2013/8/27 Peter Ekman pdekman@gmail.com
I have to say that I have a complete disconnect with this statement from Lodewijk:
"professional photographers ... can process actually thousands of images per hour - so if they are willing to spend a few hours and are helped by the right equipment (jury tool!) they can process a /lot/." for screening entries for the jury.
Viewing 1 image every 3 seconds would be 1200 images per hour, but with the inevitable break to keep your eyes from glazing over and your mind from wondering, it's probably 1,000 per hour on a practical basis. Somebody yesterday (Mexico?) gave some numbers that suggested more like 100 per hour, a factor of 10 difference.
Given that there is some computer/connection dependent time involved in displaying a photo and recording the result, I just don't see how viewing a photo for 3 seconds gives any real basis for screening.
There are quite obviously some photos that can be screened out very quickly. I'd say 10% or maybe even 20% of the photos I saw last year, I wouldn't consider putting in Wikipedia, e.g. highly over-exposed, or very dark, or my pet peeve - signs - "This way to the Statue of Liberty". These can be screened out quickly, but others will take some time. I'd also say that about 20% of the photos in the US last year were quite good, maybe even "approaching professional quality." Running by these in 3 seconds or less, it would be impossible to do a quality job of screening. Maybe you could identify a certain type of lighting, or an obvious wow factor on many, but I'd think that at least half of deserving photos would be left behind.
For those countries with more than 10,000 photos to screen the difference is huge. 10 hours of screening (should be pretty easy to handle) vs. 100 hours (a real organizing challenge and time commitment).
BTW, I don't think the jury tool has been designed to process over 10,000 photos.
Pete User:Smallbones
Wiki Loves Monuments mailing list WikiLovesMonuments@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikilovesmonuments http://www.wikilovesmonuments.org
wikilovesmonuments@lists.wikimedia.org