Recently it became clear that in just a little more time than it takes for one picture to capture a thousand words, we can now create or find a thousand pictures about almost any one thing.
The speed at which pictures can be produced and consumed has caused people to almost assume that a picture should be used unless there isn’t going to be one available.
The journey from Google Image to Flikr to Amazon is interesting because of what actually happens along the way.
It goes from one way of looking, to one place for looking, to one reason for looking.
Meanwhile, all three services present collections of images. In this case, the progression goes from a supply to an inventory to a catalog.
As a practical matter, none of the three approaches has made the other two unnecessary, although in the world of the image user, there is often little that separates the usefulness of one from the other.
This is a twist on the old 80/20 rule.
80% of the time, the user has only a 20% chance of not being satisfied by the results of the approach they use first. Or said differently, any of the three approaches is likely to be as good as the other two, 80% of the time.
The important thing, however, is what happens when the 20% difference is needed.
The problematic situation is when users find either too many versions of what they are looking for, or not enough choices, without the means to determine which version is the one they ought to use and where it is. Ironically, the overabundance of similar images comes with an increasing need to say what their difference is.
The keys to the difference are to identify what is most appropriate, and what is most relevant.
We might think that this problem would be smaller when it comes to text. After all, text has the power to “explain itself” or tell the user what its intentions are.
But experience shows us that when everything is digitized and web-accessible, there is a very high likelihood that similarity between discoverable texts creates as much confusion as it solves. Many studies reveal that as much as 60% or more of the time, people searching for text items first get incorrect or ambiguous results. In general, this reflects that searching for things by similar appearance is not highly reliable for finding things that have similar meaning.
Instead, “searching by meaning” is a major difference between possible success and probable success.
Putting this into words makes us aware of something we already knew: things have different meanings in one context versus another. Without context, it is harder to understand what something means. When context is part of the selection process, searching gives more relevant results.
Let’s say that we are the hopeful content users. To get that improvement, we move from the incredible online supply of text and “multi-media”, to at least an inventory.
In an inventory, someone is making decisions about what to keep and track. If we know how they make decisions, and if we agree with that, then we accept the content in the inventory as being appropriate. This means that we agree with the collector.
What is most helpful, however, is when we move from an inventory to a catalog.
In a catalog, someone is prioritizing content according to reasons from the users about why the content is needed. This means that the collector is agreeing with us. Because the catalog shows us the reasons along with the content, we know when the content is relevant.
If you are a content collector: re-using the content is going to mean being able to work with the supply that you create. If you can already anticipate the reasons for using that content, then cataloging it right away is a pretty good move to make it as useful as possible.
Chances are that your filing system is currently serving as an inventory. Cataloging the inventory starts with providing a view that clearly represents the mindset of a probable user. Grouping content together, according to who cares and why, is a simple start on cataloging.