Jay Henry from Ringgold led off this session with a discussion of web scale discovery and why it matters, the role of metadata, and an appeal for standards. An improved search interface has always been the motivation for innovation, and many terms have been used in the past to describe it. The emphasis is now on web-scale discovery (WSD) which Henry defined as “a pre-harvested central index coupled with a richly featured discovery layer providing a single search across a library’s local, open access, and subscription collections. WSD is not just another search method, but it is the latest step in a progression from card catalogs to online OPACs to consortia and now WSD. This latest generation of tools is something different; people can now find things they do not necessarily have immediate access to. Search systems are being linked with acquisition systems.
The emphasis in WSD is on publishers providing highly structure metadata for the content. Coupling with acquisition systems has made a huge difference. Metrics are being added to systems based on user behavior, which changes the supply chain in a fundamental way. Metadata drives use, not only purchases, so what content is used (circulates) can be determined as can getting value (usage) for what has been purchased.
Is content still king? Metadata has become the real ruler of the realm. Using descriptions of content to generate purchase and use is more important now than ever. So how do we create the best possible metadata? Discovery systems are proprietary, and the producers are not telling how they work. Publishers must provide as much metadata as possible. More is better. Standard unique identifiers are critical and result in higher sales; for example, titles that meet a basic standard average 98% higher sales than those that do not. Unique identifiers establish the longevity of the data. Titles without good metadata might as well be invisible.
Here are some strategy suggestions:
Some practical tactics:
Darcy Dapra, Partner Manager, Google Scholar, continued the plea for standardization in names and terms. She noted that we are moving beyond the limitations of print for article metadata. Systems need to understand relationships. Today we are sowing the seeds of tomorrow’s search services when the challenges for scholars will be even bigger. Doing search well requires understanding relationships, so we must make them explicit and preserve the ones we already have. If there is not a lot of metadata collected, we lose a lot of the context of an article. Concise author names are an example. This is how ambiguity gets into systems. Use as verbose a format for names as possible. The best approach would be to use full names.
In the past, many things in print limited data collection (space limitations, page charges, limited number of references, etc.). Publishers should leave more space and allow a separate space limit for references. Cryptic formatting of references is a problem; for example, many publishers only use initials for author names and don’t put article titles in references. Here are some recommendations:
Jill Emery from Portland State University (PSU) expanded on her talk in the previous session, saying that discovery is in the eye of the beholder and is facilitated by next generation integrated library systems. Consistency is critical in metadata, so accepted standards must be used, and proprietary DRM and systems must be avoided. Next generation systems will provide better linking capabilities, better open URL resolution and referral systems, and improved tagging.