On the Blacklist of Photographers from Flickr's Explore

View Comments

• Lame- somehow +Thomas Hawk was able have his sentence pardoned!

• No kidding +Julia Peterson.  I wonder whether +Marissa Mayer even knows about the Flickr Explore blacklist... hard to imagine it remaining in place after any sane, intelligent person found out about it (someone who didn't run Yahoo into the ground).  Hopefully she knows about it now.  I've re-submitted a Flickr Customer Support case with a link to this post, so they'll know that their answer is being eagerly awaited....

• Pardon my ignorance as I don't use Flickr. How do you know you have been blacklisted from this explore? How does one get on explore? From reading some of your words of one of the links it sounds like the images picked for explore usually have a lot of social activity related to them.

• This is strange as one could expect that with the demise of Yahoo360,MyBlogLog,etc that Flickr could have formed the basis for relaunching Yahoos social networking efforts.

• I agree - people who have behaved themselves should not be blacklisted.  Simply asking questions that are uncomfortable to flickr is not a good enough reason.

  What I also find interesting is that I was a paying member of flickr for 3 years.  I was never on explore.  When I let my paid membership lapse about 6 months ago, I was suddenly on explore about 4 times right away.  This seems like a bit too much of a coincidence.

• It's an attempt by Flickr to identify the "most interesting" photos on Flickr.  The algorithm isn't known, but speculation could be that it's some combination of overall activity and intensity of that activity (activity per unit time), perhaps compared to some baseline of usual activity for that user.

  If a photo is interesting to many people, it may be interesting if recommended to others, in the 500 per day presented in Explore.  Explore can result in a major jump in views, and that photo's ranking in response to searches later (including searches by photo buyers).  The operation and effectiveness of the site can be severely affected for people who are blacklisted, yet there's no indication that we're held back, other than a sudden and unusual exclusion from Explore, and there's no discount for the reduced functionality, we're charged the same annual fees as people not restricted.  

  It strikes me as an extremely underhanded way for Yahoo to do business with its loyal, paying customers.  Will that continue to be the standard for Yahoo going forward?

• I guess my confusion still comes down to how you know you've been blacklisted. Especially with the knowledge that this algorithm is unknown to Flickr users. Without knowledge of that there is little one could do to purposefully improve or decrease their chances on getting on explore, and there is no way to prove this blacklist (which theoretically would be a handful of Flickr staff cherry picking people out of the pool of possible explore photographs) even exists.

• Flickr Explore works in a way for the casual user to find new photos and photographers to follow.  For the uninitiated, who is in explore matters less than just that an explore exists.

  It hurts flickr's reputation amongst photographers though when they use a blacklist with it.  Flickr probably should be courting a lot of people on their blacklist not trying to drive them away.  Related to the blacklist problem would be the problem of people assuming or thinking that they are on a blacklist even when they are not.  

  It also feels unethical to me for flickr to allow an ex-employee to deny that a blacklist exists when one in fact does.  This goes against the idea of user trust which is more important than the potential secrecy of an algorithm.

  If flickr can't manage an algorithm without blacklisting then I think that's a problem that they need to figure out and fix.  It really should not be that difficult.  If they can't, I think I'd err on the side of over-including people that they would just as soon blacklist than in blacklisting.  

  It doesn't really matter to those "exploring" who is in explore, but it does seem to matter to many who are blacklisted or think that they are blacklisted.

  Tangentially related, Explore does a horrible job with personal relevancy which is why I stopped using it a long time ago.  I felt that the quality of the photography there was not what I wanted to see and I get annoyed by an oversaturation of signatures/watermarks which is pretty much a good definition of what explore has become today.

   Explore is not really for me though, it's for the casual and uninitiated flickr user.  It could do a good job for both them and me though if flickr more carefully considered personal relevancy.  

  I give Flickr so much data about what sort of photos I want to see.  I give them data by telling them who my friends are, by telling them who my contacts are, by what photos I fave, by whose photos I fave, by what tags are on the photos that I fave, by where the photos that I fave are located in the world.  Explore uses not a bit of this data and it could and it could become so much more powerful as a discovery/recommendation engine than what it is today.

  Algorithmic photographic curation based on solid machine learning is going to be a very big thing in the coming years.  Photo sharing is in its infancy and as it develops we will need a better filter to manage the growing number of photographs that are published daily to the web.  There will be a huge opportunity for the person who can create the perfect algorithm.  There will also be enormous power and responsibility for the person who manages this algorithm.

• It would be nice to give a fresh look to the explore just like Flickr did for displaying photos from contacts or in group pools. A bigger size would be welcome. It would be nice if the Magic Donkey could learn how to select more interesting photos than the average ones selected at the moment.

  I'm probably not blacklisted but I only got 1 or 2 photos in the Explore in the last couple of years versus 1 a month before.

• Explore photos tend to be pretty whitebread, so I (still) can't get real excited about this issue.  Rather than tweaking a poor system, flickr should throw it out and implement real collaborative filtering.