Home Downloads Products References Support Talent Buy Now Contact

Keywording Theory

by James Cook
Whether you call them keywords, indexes or cross references, keywording is not so much for data entry as it is for searching later on. The words that you attach to an image play a crucial role in finding the right images when conducting searches. Omitted words may prevent the most appropriate images from being found.

A good keywording system takes time and thought and can not be constructed in haphazard fashion with expectations of great results. If you create and apply keywords well, the result is faster, more fruitful file searches.

Keywording systems fall into either one of two styles.

Most commonly used, although not always the best, is the open vocabulary. If you look at an image and use whatever words it inspires for the keywords, you're using an open vocabulary. One day you may keyword an image as Ship, Harbor. Entered on another day, you might have keyworded the very same image as Boat, Port. Another time you might use Watercraft and Harbour.

When searching for this sort of image, who knows what you’ll find or overlook because of inconsistencies in an open vocabulary. Of course, you can look for all entries that contain Harbor or Port or Anchorage or Boat or Ship or Vessel, but it’s hit or miss proposition that requires a vast patience or a good memory. If you find no entry for Boat, you must keep looking, trying other words in your keywords. When you discover Ship, you move on, never discovering all the others that got listed as Watercraft.

In a controlled vocabulary your keywords are limited to a specific set of words. Ideally this list of words is printed in an index or available on a computer for your reference as you apply keywords to images.

As you make entries in your controlled vocabulary system, you should have the option to add a new keyword. If you fail to find a satisfactory keyword for an image, add a new one to fill the hole. This way the keywords build slowly and thoughtfully as need arises. You are spared the monstrous task of building an entire keywording system before you know what you might need.

One long list of all of your keywords makes the task of locating and applying them impossible. Within your keyword index, you're going to want levels of keywords; categories and sub-categories. Use the primary categories to quickly make significant distinctions within the subject matter. Use sub-categories to make quick breakdowns within the category and add additional sub-categories as needed, but don't get carried away. Generally three or four levels are as far as you should go. A good breakdown makes it easier to add keywords efficiently and makes the searching easier too.

Suppose your files have lots of animal images. Start with a general category, Animals or Nature or Wildlife. Now, let’s say you have lots of bears among your animals. Try a sub-category, Bears. Use yet another level for details to get down to the fine points, Brown, Grizzly, Kodiak, Polar… and so on. These details are hardly suitable for any other category or sub-category.

A category, Birds, could have sub-categories of types of birds, maybe Eagle is one. details for Eagle might be Bald, Golden and haliaeetus leucocephalus. If you’re tempted to include details like Flying, Egg, Hunting and such, remember, these details apply to many birds. Flying, Egg, Hunting and such might serve better as sub-categories of Birds and not just Eagles.

Limiting the number of choices increases the accuracy of new entries and expedites searches but don’t be so extreme in avoiding new terms that you diminish coverage. A happy medium exists, but it's up to you to decide where it is.

When keywording an image, look at it and think about what it really shows. What is significant in the image? Look for both the literal and the conceptual significance. A single image is likely to fit into multiple categories and sub-categories, so don't think you need to apply one set of terms and that's it. Most images contain a literal subject such as a dog playing in snow. Such an image would obviously earn keywords such as Dogs, Collie, Pets, Winter and Snow. Conceptual keywords for the same image might be Cold, Joy and Freedom. Resist keywording every object that happens to be in the shot; Door, Tree, Mailbox and so on unless they play a significant enough role that someone searching for those objects would really want to see this one too.

For efficiency, use the most basic, generic terms. Don't use your keywords to show off your vocabulary skills. Your keywords should be kept to a level that is appropriate for your audience. If your images are for osteologists and paleantologists by all means use terms like zygomatic arch and optic foramen. Don't expect the rest of the world to know what these terms mean though.

Reduce synonyms to the single most suitable word for a category or sub-category. This narrows rather than expands your search vocabulary list. For instance, here is a list of synonyms: precipitation, condensation, downpour, deluge, cloudburst, torrent, rainfall, shower, drizzle, sprinkle, monsoon, plus all their plurals. Rather than create a category, or even a sub-category for each and every one, use a broader term, such as Rain.

And finally, be sure of your spelling. Typos and spelling errors cause images to be missed in searches. Get it right in your master list and make sure to stay with it.

Good captions and keywords enhance the value of your images. An excellent image that never turns up in appropriate situations is a lost cause. Keywording is an important part of successfully bringing your images to market.
Copyright 2012 HindSight Ltd All Rights Reserved