imprezzeo

Yesterday Getty Images went live selling a specially selected slice of Flickr content from professionals and hobby snappers alike. The Flickr Collection as its called contains content that satisfies the keen editorial eye of Getty’s experts, with apparently around 6000 or so images making the grade so far. Initial thoughts and questions:
 

1. Getty innovates again (they developed a model which others tried to replicate, they grasped the micro stock nettle with the iStockphoto acquisition etc.)

2. What though will be the reaction from their premium content contributors like Stone etc. that Getty has opened itself up to ‘amateurs’ (that said, some of it looks quite good to my untrained eye)

 

3. Flickr has billions of images. Where did the folks at get the time to pick the best 6000? This is probably a little sampler and Getty will hoover in more if this flies

 

4. Am I the only one who would have thought this collection would be royalty-free only – nearly a third is rights-managed

 

5. Smart move by Flickr to find another and potentially very lucrative revenue stream, and one with decent margin to boot

 

6. A very experienced photo desk editor once said to me that his newspaper’s taste in photos had changed, that “edgy” content was now good. Sample of one I know but I bet it will be harder to shift premium rights-managed content, even after this economic slowdown – this is virtually a concession by Getty that that will be so

 

7. Want to see a perfect example of the failings of image search via metadata? Look at the first image in the collection. To me it shouts out one obvious thing and the title itself, for those of us with rudimentary French is a bit of a giveaway too – SNOW! So search for snow within the flickr collection and there’s no sign of it, unlike the image two along. That’s because for some reason, the latter has the word ‘snow’ in the keywording, the former doesn’t

 

8. Click on the “more images like this” option in the second image, Mr Spaghetti pants. Why does Getty only offer similar images from within the Flickr collection? Isn’t that a bit like going into a department store to buy a pair of jeans and being offered other clothes from that same brand other than trying on other brand of jeans?

 

9. Also, when you get those similar results back, how does the 5th picture (accordion player) resemble the first? Intrinsically it doesn’t but half of it’s keyword tags do

 

10. There is actually another find similar images feature that appears after you click the thumbnail which is a little confusing and also, keyword-reliant

 

11. Is my maths wrong? Getty shows 60 images per page and 67 pages of results which makes 4020. Where are the other 2000 or so in the Flickr Collection

 

 

A leaked memo from Satya Nadella, Microsoft’s SVP of R&D, Online Services re Project Kumo (c.f. my comments on technewsworld) yields some very interesting ‘facts’: “40 percent of queries go unanswered; half of queries are about searchers returning to previous tasks; and 46 percent of search sessions are longer than 20 minutes. These and many other learnings suggest that customers often don’t find what they need from search today”. Setting aside the question of how an unanswered search is defined / quantified and if we accept the validity of the stats, it says that too many searches are abandoned, too many re-started and too much time wasted.

How much more problematic then the search for images which relies on text tagging to describe content that given it’s semantic richness is fundamentally unsuited to ‘description’. The hollow boast in search results that the engine has retrieved x million image search results in y milliseconds is to me more depressive than impressive. Surely it’s time for a different approach.

The more I and the Imprezzeo team ‘freestyle search’ image collections by clicking on examples of what we want, refining and traversing content to find what we want; the more we build features that allow you for example to highlight part of an image as the example upon which to base the search (only possible with an underlying visual search platform); the more we hone our engine to provide great results across a diverse range of content - the more we are convinced that the days of text-based-only image search are numbered. Shameless plug perhaps but reality, definitely.