One picture is supposed to be worth a thousand words, but if you don’t know what the picture is about, it may not be worth much. Information about data is, not surprisingly, called metadata. It can be stored inside a file or along with it. It can come in many different forms, confusingly many.

One of the simplest forms of metadata is the file name. If you have an image file called IMG_0673.JPG, that tells you only that it’s a JPEG file and it probably came after IMG_0672.JPG. If it’s a picture you took in the morning from a castle (as my file by that name happened to be) and you name it MorningViewFromCastle.jpg, that tells you a lot more. The date on the file is another useful piece of information, except that it may tell you when the file was stored to your computer or last edited, not when you took the picture. You can also name the directory to provide information about a group of files, and you can add a text file to the directory with information about them. That file is external metadata, and you can manage it without any unusual tools.

Storing the information in the file has the advantage that the two won’t get separated, but you need appropriate tools to edit and extract it. Lots of file formats have a provision for storing metadata in files, sometimes more than one way. Let’s look at JPEG files in a little more detail. There are two common forms of metadata found in them. One is EXIF data. This is defined by the Japanese standards organization JEITA which is focused (if I may use the word) on camera data. It provides information such as the camera model, photographic conditions, and the date of the picture. If your camera is GPS-enabled, it can include exactly where the picture was taken.

Another widely used standard for image data is Adobe’s Extensible Metadata Platform, or XMP. XMP is the Borg of metadata standards; it’s based on XML, and it has assimilated older standards. Standards such as DICOM and IPTC are still around, but they tend to be incorporated into XMP data. This makes it possible to export all the metadata at once and work with it in another application. (EXIF is a strange beast, based on the TIFF file structure, and it’s been around for a while, so it’s resisted assimilation.)

Adobe Bridge is a wonderful tool for working with image metadata. If you open an image file in Photoshop or any other image editor just to make changes to the metadata, when you save your changes it rewrites the image along with the data. JPEG is a lossy format, so each time you save the image you risk losing some quality. Bridge touches only the metadata. It lets you edit the metadata for several files at once, so you can set all the invariant fields (such as the creator and copyright notice) in one step, then have less work adding the specific information for each file.,

Some applications handle metadata better than others. GIMP is free and does a lot of things very well, but its metadata handling falls short of Photoshop’s. It does preserve metadata that’s already in the file. Whenever you use an unfamiliar piece of software to edit a file, you should check that it doesn’t lose your metadata.

In MP3 (and some MP4) sound files, the dominant metadata standard is ID3. It lets you specify information such as the title, composer, performer, and album. Modern software uses ID3 V2, which is much more flexible than V1. MP3, like JPEG, is a lossy format, but if you edit the metadata in a player such as iTunes, it won’t change the audio data, so you’re safe. But iTunes doesn’t let you edit all the ID3 fields. In fact, there are lots of interesting tags in ID3 V2.3 which iTunes doesn’t touch. Really, as far as iTunes is concerned, only popular music exists, and you get it only by downloading or copying. It doesn’t let you see the tags that give information about the language, conductor, orchestra, key, and on and on.

If you’re creating your own sound files, or if you want to annotate your CD transfers in detail, you might want to use a dedicated ID3 editor such as Pa-Software’s aptly named ID3 Editor.

I haven’t even gotten to video and PDF metadata, but there will be lots more in future posts.

Further reading: