26 Feb 2008 PhotoKeywords

This is an update to a post originally published over a week ago. The original post was taken down due to actions of the creator, which I’ll explain.

The Original Review

Russian programmer Ivan Maslov has created a tool which makes keywording fast, easy and accurate. How it works is ingeniously simple.

Feed a few core keywords into PhotoKeywords.com and it will return a list of thumbnails from Shutterstock and LuckyOliver that match those keywords. Simply tick the boxes of the photos that are similar to yours and the system will return a list of the keywords used by those photos. The results are ordered by popularity of searches at microstock agencies, providing weighting information crucial to ensuring your keywords achieve maximum sales.

Simply un-check any that don’t apply and the resulting keywords are finally presented in a text box separated by commas, ready to copy and paste directly into your photos.

This tool is great for microstock contributors looking for an easier way to generate a list of keywords. It’s accuracy depends on selecting enough sample images from a range of contributors and un-checking the keywords that aren’t relevant. As the database is constantly updated, it will naturally follow trends in the market.

There has been some speculation as to the legal position of the service, but it seems to have survived most threats. Initially, the site used the database of Dreamstime and Shutterstock, but after Dreamstime expressed disapproval (by blocking the service IP) Ivan simply switched to use LuckyOliver instead.

I will be using this tool within my workflow from today forward, and I look forward to seeing how Ivan progresses his venture.

Deeper Issues

After first publishing this review I received feedback from the market raising some issues.

First and most significant is that keywords are copyright. PhotoKeywords is therefore a tool that helps contributors violate the copyright of other contributors. Copying keywords still occurs at all levels of the market anyway, but this tool facilitates the violation.

Second, contributors must be diligent every time keywords are generated to avoid using inaccurate keywords. The only manual part of the process is removing irrelevant keywords from the list. As there are always some such keywords, it’s very easy for contributors to overlook or neglect to remove them.

Third, the popularity index includes a column to indicate how many images use the same keyword. By factoring this in, contributors are using more common keywords, which gives them more competition.

Responses from microstock agencies themselves have expressed concern that contributors using the tool can become unintentional keyword spammers. No system can keyword an image perfectly every time, so the message is for contributors to use this tool in full knowledge of its limitations and risks.

So What Happened with the Original Review?

PhotoKeywords creator, Ivan Maslov, created this tool without consulting the agencies from where it gets its data. In my experience, agencies are very approachable and responsive regarding tools which use the data from their websites, so a quick email to discuss his plans would have ensured a smoother launch.

He also chose to promote the website by creating posts in the forums of popular microstock agencies. While this itself isn’t offensive, doing so at an agency that has already blocked your service will always cause upset. Not being a contributor or buyer at the agency doesn’t help, nor neglecting to communicate with the agency first.

I emailed Ivan, with whom I’d had an email conversation prior to publishing my review, requesting a response to these issues. I still haven’t received any response.

The Result

With the above caveats noted, PhotoKeywords can be a useful tool, particularly those for whom English is not their first language. Ivan has done a great job creating it, but just used some promotion techniques that have caused upset.

Update: PhotoKeywords.com has now been purchased by Yuri Arcurs and is available on his website.