Help Find My Assets
The whole reason for DAM is to build a system that provides a way to get at assets when you need them, but tuning the search component of a DAM is a fine art that must balance broad coverage versus meaningful search results. Porter says you need to be thinking about this when designing the system so that you can anticipate different search approaches. For example, some users may search for the word car, while others may enter auto or automobile. Things become more difficult when you want to fine tune the system to limit searches of autos to vehicles, not automatic weapons.
Porter says you need to "understand how the search engine in your system is going to work. That will help you formulate how you create the data, and if you are using a keyword-driven approach, then [the auto-car problem] can be accounted for based on how you set up your taxonomy and the methodology, the rules you have created on how you apply the metadata to the image." This means you could link auto and car in a synonyms list, so no matter which keyword users enter, they will get back results with car images. She says another method is to make sure that all the people using the system have a common set of keywording rules, although she admits that approach is more difficult to enforce and therefore less thorough.
For ClearStory, Titchener says it's about building multiple ways to locate the assets and configuring the system to help the user to get to the result as quickly as possible in the least number of clicks. "What you have to present are multiple ways that a user can find assets," he says. "So there is everything ranging from being able to browse to it through a familiar folder structure, and as part of that, you have simple search and advanced search and within this you get within the complexity of what the product is doing behind the scenes and presenting it to the user in a familiar form."
Another way that ClearStory helps end users get to the assets they need is through the provision of access privileges, where only certain users can see certain files based on their access level, which can be very helpful in reducing the number of results. Worthy offers an example of a customer, pet-products company RC Hagen. It is a regional company, so when a person from a particular region signs on, they only see assets relevant for the given region, not the entire company, and those results are narrowed further based on access privilege level. "Based on my region, I only see the assets relevant to me. I only see assets in my product line. When I am returned my hit list, I'm only returned those assets in the rendition I'm allowed to use. If I don't have the privilege to find high-res, I'll only get the low-res. So I only see file types I have to use," Worthy says.
Faced with a growing repository of video and image assets, it can be difficult to find the assets you need when you need them. Worse, if you end up creating assets, like a logo, over and over again because you can't locate the version you need, you are wasting time and money. A DAM goes a long way toward bringing your collection under control, but assets need to be more than managed; they need to be findable. The system you put in place will only be as good as the taxonomy, classification, keyword, and search components you develop and only as effective as your efforts to enforce the metadata and keywording system. The old chestnut "garbage in, garbage out" was never so apt as with DAM: if you don't build the infrastructure for findability, your valuable assets will be about as useful as a blank search results page.
University Corporation for Atmospheric Research (UCAR)
Sidebar: Case Study: University Corporation for Atmospheric Research
Carlye Calvin works for the University Corporation for Atmospheric Research (UCAR) and the Center for Atmospheric Research. She divides her time taking pictures of storms and building an image catalog for the UCAR Web site. UCAR recently purchased Portfolio from Extensis to manage its assets after outgrowing another solution.
Calvin says the old system was hosted on a Hyperion server with a custom search application, and every time they modified the system they would have to update the search component too. "Every time they updated the server software, it would mess up the search application scripting, so we would have to contact the programmer and have him fix it and we felt we didn't have a lot of control over the process," she says.
What's more, the custom search solution didn't allow for true Boolean search, and didn't even account for plurals, Calvin says. "So if you searched for cloud, it wouldn't find clouds," she says. Another issue was that the system required her to manually generate thumbnail images.
UCAR decided to go with Extensis Portfolio because it solved all these issues without requiring UCAR to pay dearly for a customized solution. "Portfolio fit all of our needs and it wasn't that expensive," Calvin says. What's more, it resolved the problems the customized search solution had. In fact, Calvin says, "It's great for locating images. Search works beautifully. So long as you do a good job of inputting metadata, it's easy to find what you need."