This has been covered before on other sites (and to some extent on ours), but since I recently explained all of this to someone on Twitter I thought I’d take what I wrote then and massage it into some form of post for our blog.
To begin with we need to cover a few terms. One is that, as I mentioned in our Should I Get Lightroom or Photoshop or…? post, a digital image isn’t really an ‘image’ in the traditional sense. It starts as light, gets converted into electrical impulses by the camera’s sensor, and is then translated into binary code – 1s and 0s – as a digital file. Be that as it may, there are two aspects to each digital image. One is the image information itself – the code that is assembled to create the image on your computer screen or as a print, and the other is information about the image. This starts in the camera and can include the camera make/model/serial number, the exposure information, date and time of the image capture, the lens, focal length (for zoom lenses), GPS information and more. When the image is transferred to the computer one can add copyright information, keywords, owner contact information… All of this is collectively called ‘metadata’. This metadata is stored in one of two ways, depending on the type of digital image you’re working with. For .TIF, .PSD, .JPG and .DNG files, the metadata is stored within the image file itself. For raw images, a second file, often called a ‘sidecar’ file is generated. Sidecar files have an .XMP extension. Therefore, for a Canon raw file for example, one would have IMG0001.cr2 for the image data, and IMG0001.xmp for the metadata information.