Safeguarding your images on the Internet isn't as easy as you think. Perhaps the answer lies in digital watermarking?
The Internet makes it simple to copy and distribute media, but, as any record company or movie conglomerate will tell you, that isn't always a good thing. The success of file sharing tools such as Kazaa provides clear evidence of the rampant nature of copyright infringement, but it isn't only media multinationals who suffer.
Anyone who creates digital imagery is at risk from data theft. And that applies equally to consumers, too. What if that stunning sunset you photographed and put up on your Web page for your family and friends to admire ends up advertising a travel Web site without your knowledge?
The sad truth
Stopping content thieves taking images from publicly accessible Web pages is, I'm afraid to say, practically impossible. All you can do is make theft more difficult.
To turn this off, include the following line inside your HTML's <Head> tags: <META HTTP-EQUIV= "imagetoolbar" CONTENT="no">
A sneakier but more effective way of protecting an image on a Web site is to create an HTML table and set the image to be protected as its background. If you place a transparent GIF with the same dimensions on a cell above it, visitors can see your picture but can't download it directly from the page.
That still won't deter the dedicated thief - a viewed image can be pulled from a browser's cache - so get devious by slicing your images. Slicing is a feature commonly used to ensure Web images load more quickly, but a side benefit is that it makes it impossible for a visitor to download a complete image. Instead, they'd have to re-assemble each section in an image editing application, which may be too inconvenient to make it worth the effort.
But even those tricks and others, such as embedding an image inside a Flash SWF file, can't stop someone swiping your pictures by simply taking a screen grab. Perhaps the best idea is to tag your photos before they even make it to the Web so that ownership is indisputable.
Most digital images already carry information about their origins in the form of EXIF metadata, which holds details of the camera that took the photo and the time it was taken. And in most image editing applications you can also manually add IPTC metadata, which shows author, keyword and copyright information.
Expect to see more of this - back in June, Adobe announced it would wrap IPTC and additional data inside a new open-standard Extensible metadata platform (XMP). But even metadata is too easily editable to counter thieves, so this month I've been looking at a more permanent alternative: digital watermarking.
Digital watermarks fall into two categories. Visible watermarks, which can be as simple as a text layer above a background image, make copyright information clear - a disincentive to copy in itself. But aesthetically they are distracting.
Where's the watermark?
Invisible watermarks, on the other hand, can only be detected algorithmically. This month I've been testing the best-known digital watermarking tool, DigiMarc's MyPictureMarc, which works with the big image editing applications from Adobe, Corel, Micrografx and Ulead.
In Photoshop's case, once you register for the service at www.digimarc.com (an annual personal subscription costs $US49), you just choose and apply a level of durability and visibility. The watermark is embedded as additional noise, which was imperceptible to me. This information acts as a unique index that can be read by any supporting application on any platform, with the user pointed towards the owner's details on DigiMarc's Web site.
Disappointingly, it isn't too difficult to circumvent MyPictureMarc's watermarks. They do survive rigorous editing, including compression, rotating and cropping, because the watermark is applied across the whole image. But I discovered that by applying Photoshop's noise-removing Despeckle filter, I could render the watermark unreadable. That doesn't make it a worthless tool. In practice, few users will check for an invisible watermark so there's a good chance it will remain intact.
But if MyPictureMarc provides a clever way of embedding copyright into your pictures, it's just a pity there's no way to track their usage on the Web. DigiMarc does offer a professional service that identifies where your images are being used online. But the drawback is its $US499 Professional subscription. If DigiMarc was to cut its prices to a more sensible level, I could see a lot more professional and consumer interest in this.
Hard as it is to stop copyright theft, it's even more difficult to seek legal redress for any breach. Copyright of artistic works such as photographs is automatic in Australia, but infringement is largely a civil offence. It's only in serious cases of commercial infringement that you can involve the police, so it could take your time and money to prove a case.
The waters are muddied further when you're dealing with a stateless medium such as the Internet. Although copyright material stored on a Web server is generally protected in the same way as material in other media, enforcing your right in other countries could prove difficult.
Australia is a signatory of the Berne Convention, which protects Australian copyright in other signatory countries. But it's still a good idea to play it safe and make sure that each Web page containing an image displays the international © mark alongside the name of the copyright owner and year of publication.
Don't Google it
Another way of keeping images away from casual thieves is to exclude them from image search engines, such as Google's.
To do this, create a plain text file called robots.txt at the root level of your Web site. Almost all Web-search robots look for this file when they index your site and they obey instructions contained within it. To prevent your site's images from appearing in a Google image search, add these two lines to your robots.txt file:
User-agent: Googlebot-Image Disallow: /