April 16, 2008 by pmelcher.

I did not pick this image. I actually have no idea what it will be before I publish this entry. Why ? because it is a sort of semi “intelligent” algorithm in the background that will do it for me. A bit like Google ads scans a whole web page for keywords and post the relevant ads, this system, delivered by Dailylife.com, does the same.

It will scan the page for keywords and post the most appropriate image. Like an automated photo editor. And because it is looking thought the feeds of Reuters, GettyImages and AP, I believe, it selects from a pool of already very tightly edited images. One could also foresee a Flickr API, a bit like I did with the yahoo pipes.

I am guaranteed a good and hopefully, relevant image . This is the future of news photo editing on the web. At least for sites that do not care so much about the image and use them as an illustration of  a written report. Why pay some guy to look at a stream of pre edited  images, download one, resize it and post when the whole thing can be automated. And better yet, computers don’t whine, do not take lunch breaks, or holidays and never, never ask for a raise. So why keep a web photo editor, if only to do some “best of the week” gallery ?

Think about it:  the biggest news source of the internet has no photo editor. It is called Google news and it selects images with a similar technology. Indeed, it relies on images previously edited by pro photo editors. For now.

The dailylife link is completely free, with no uncontrolled ads, like a Picapp or a GumGum would like you to swallow. Sure , it has a link for the site itself but the same technology could easily be applied by anyone on their own site.

Finally, Dailylife.com, still in Beta, looks like an interesting destination. It seems they want to be a new Google news but put a heavy emphasis on photography and has a much better and smoother interface. More like a magazine designed for the internet, and not the opposite. Finally.

As newspapers and magazine are suffering more layouts as ad spending is weakening, most of the photo related professional are turning to the internet. However, because of its built in automation, it just seems that some of the jobs will not be recycle but ultimately replaced by machines. We will still need great pictures, thus talented photographers. Not so sure about needing photo editors.

Posted in multimedia, Search, newspaper, digg, magazine, gumgum, technology, keyword, yahoo, editorial, news, slideshow, wire service, google, photojournalism, getty | Print | 2 Comments »

January 22, 2008 by pmelcher.

This is what you get when you mix the talent of Brian Storm and his team and the incredible dedication and eyesight of Marcus Bleasdale, new member of VII photo agency :

If only everyone had a camera…with a conscience. Like Marcus.

Posted in digg, magazine, focus, lens, mediastorm, multimedia, editorial, slideshow, wire service, photojournalism, news | Print | No Comments »

January 11, 2008 by pmelcher.

A funny thing happened recently. One of the top microstock agency, Dreamstime, has decided to lift part of the veil on sales data. As an additional tool given to their contributor, they are now showing what keyword generated a download, thus a sale, of their image.

It is a great tool for contributors, albeit a double edge one. One could be tempted to use it for “keyword spamming”, the art of putting the wrong keyword to an image, in the hopes it will be seen, and eventually bought. A plague for microstock platforms. But what is new and challenging is Dreamstime decision to share part of its sales data.

The history of photo agencies and most recently sales platform has always been the opposite. In order to keep contributors with them, no one shared valuable sales data. It was, and is, proprietary information. A bit like enslaving someone by depriving him of knowledge. Getty, Corbis or JupiterImages, along many, many others would never dare make such a move for fear that their contributors would use that information to feed someone else, such as the competitions’ image database. And that will happen to Dreamstime, since exclusive contributors are maybe around 10% of their pool.

In a slide and phone world, that would have been quasi impossible . Requests for images did not come in the form of one or a sequence of one words. Thus photo agencies would have had a very hard time sharing that information. With websites, all this is different, obviously. But it still remains a very well guarded secret.

The real question is how is this useful ? Well, for one, it will bring a succesful microshooter valuable data on what types of keywords are entered for search. And which ones are the most entered and referred to their images. It will not, however, give a full picture. And that can be highly deceiving. It is not because I sold 100 images thanks to the keyword “butterfly” that this keyword is the most used in searches. It just probably means that I have the most relevant images of butterflies. Since it will not give all keywords, it will not help putting keywords you forgot to put in. For example, if my specialty is photographing fruits and I systematically forgot to put the keyword “fresh”, it will not show me how many sales I missed, if any.

Contributors reaction to this tool has been overwhelming positive and I command Dreamstime for making this available. It is a gutsy move to give photographers access to their sales data. It might, or will, force other microstock companies to do the same. Eventually, it might even force traditional agencies to follow and share more of their information. Their is a value for photographers to be less blind to the markets demand, while there is danger in using past sales to predict or influence future sales. It is not because a keyword has done well in the past that it will do well in the future and the risk is having more of the same image ad nausea. But this is more in a trend that is seeing more power given to photographers. And that is a good thing.

ON a completely unrelated note : Digg.com labs has released a image visual tool which allows to see, real time, what images are being voted. It has no practical usage that I can see but yet is very interesting.

Posted in Search, IPTC, Jupiter, digg, technology, keyword, web 2.0, getty, Royalty free, corbis, filter, prosumer, Microstock | Print | 4 Comments »

December 4, 2007 by pmelcher.

Social bookmarking website Digg.com has just launched, last night, an addition to their service. From now on, registered users will be able to submit images. For those unfamiliar with Digg, let me explain. You see an article you like, you submitt to Digg. If others like it, they also “Digg” it. The more Diggs, the more the article reaches the top. On the other side, a lot of people will go to Digg.com and read the most popular article.

What is very interesting here, is that it is now possible with images, creating the first social bookmarking of images. Why is this relevant to our industry ? Digg is a highly popular website, at least in the USA. It has a huge amount of traffic. This will allow us to see what images are the most popular. But with a twist.

The images that will appear to the top will be the images that most people find interesting to share, not to be confused with the ones they like the most. It is more likely that we will see more strange, funny, exotic images than groundbreaking photojournalism . But nevertheless, for advertisers looking to find the next viral marketing campaign, or publishers, seeking to populate their pages with eye grabbing content, it will certainly be a source of inspiration. For photographers too, it could become a source of inspiration.

Furthermore, it will be much different than a Flickr or other photo file sharing platform as the majority of voters will be not interested in photography. Finally, it will also be fascinating to see how many photo agencies, including the microstocks, will add the “digg” buttons under every single of their image.

With the crest of the web interaction at its peak, it will be a very compelling destination to monitor regularly for anyone curious about the relationship of photography and its users.

On a parallel note, and the second newsworthy item, is that Digg selected Idee Inc technology to prevent having duplicates. Thanks to Idee, if people vote for the same image but seen on different sites, it will add up instead of duplicating. A confirmation that the company has a real dedication to its core competence and not venturing into doubtful activities.

you can see Digg Images here Be prepare to see a lot of images of Cats and Dogs…

Idee Inc new website can be seen here. Check out their lab. very cool stuff

Posted in digg, idee, Search, flickr, news, photojournalism, Microstock | Print | No Comments »