Month: November 2005

I found out today that another blogger out there is using a photo that I took:

http://blogantes.blogspot.com/2005/11/tic-tac-tic-tac-o-tempo-no-passa-as.html

This, in itself isn’t completely a big deal, but the way that the image is being used means that I’m paying for the bandwidth to show my image on her page. This is called inline linking or hotlinking and is considered bandwidth theft.

I’m not sure how I feel about it though. The good thing is that when a reader of her blog mouses over the image, they will see the hyperlink pointing back to my site. I’m not sure this was intentional or not. I’m not sure of the intentions of this blogger at all, because the site’s not in English.

Any thoughts about what I should do? I know there are ways to restrict the serving of links to only users coming from my main site. I’ve considered serving up a different image with a nasty message to readers of her blog. I mean, when she made the inline link, she gave me the power to substitute any image I want, right?

Google Image searching is partly to blame for this. I’ve been seeing a lot of visitors come to my site via a search for “clock”. So this might only get worse in the future. But for now, I’m just chillin’ about it.

Update: Following the discussion in the comments, I’ve marked the photo with a copyright notice and my name. So at least my name gets out there.

Update again: Yikes!!! I took a serious look at my webserver logs tonight and found out some pretty interesting things besides this one blogger using my clock image.

• In the past 8 days, I’ve served the image 1496 times for a total of 40 MB! (That’s 7.44% of the total hits on my site. In comparison, the header image of my blog has been served 1134 times, or 5.64% of the hits.)
• It’s being used in about 37 different blogs and message forums. And someone has even incorporated it as the background image of a weblog theme (!).
• It’s the last image on the first page of a Google Image search for “clock”.

This last item caused the previous two, especially since it’s the clearest-looking image of a whole analog clock presented on the results page. Check out this screenshot.

In other words, I got screwed big time. So, I’ve implemented the robots.txt filtering that Jim suggested. This won’t do anything until Google and others recrawl my site. So, to stop my leaking bandwidth, I’ve renamed the file so no one can find it. This will break all the webpages out there that inline-link to the image. This will also break my weblog entry about the clock. (I thought about fixing my code to point to the renamed file, but Google Images would show my page with the corrected image, so people could still grab it.) In about a month, after Google drops my image from its search results, I’ll fix my code and restore my page to normal. Hopefully, thereafter, the problem won’t return.

What surprised me most about all this is how prevalent this practice is: people on web forums and weblog hosting services use inline linking a lot to spice up their entries. This happens because they either have no hosting options, or it’s just too easy to copy and paste a link from Google Images. They probably have no idea that this behaviour makes them such bad neighbours on the Web.