One of the biggest problems faced by academic library subject selectors today is building collections that include materials from a wide variety of sources, and particularly, how to integrate content that is available for free on the Web into existing collections. This is a complex problem with many facets: identifying relevant content, determining what is available for free online and where it is available, constructing and maintaining persistent links to the content in a remote location and/or downloading it and archiving it locally if allowed, and creating or obtaining and enhancing records to represent this content that can be incorporated into local discovery services (such as a library’s online catalog).

I would like to build an online community, perhaps similar to Wikipedia, where subject selectors could find information and tools that would help with the various aspects associated with building collections in the Web environment. Components could include:

• Identification of “core” resources in various subjects (contributed directly by experts and derived through/supported by statistical analysis)
• Utilize various APIs (such as Google Books, WorldCat, Internet Archive) to obtain bibliographic information and links for resources.
• Support customization of metadata and export of records in various formats to local catalogs or other databases.
• Support for tagging and generating customized feeds from the central server that could be incorporated into local web pages.

• Make it much easier to create linking entries between records, e.g. support point and click creation of 776 and other linking entry fields between records for the same bibliographic content in print or electronic format
• Provide access to full-text of cataloging rules and LC cataloging manuals within the OCLC cataloging interface (via an API that accesses the data sources directly?)
• Provide color coding in the record editing display similar to that provided in a software programming integrated development environment (IDE)
• Provide much more robust tools for analyzing record contents and updating records in batches, perhaps utilizing data visualization display techniques (as per Hillmann & Dushay)
• Improved record validation to check for internal inconsistencies in records (e.g. a in ILLS fixed field, but no ill. in 300|b)

These days, I often find myself wishing that WorldCat was built on a Wiki platform with robust web services built in, so that I could easily correct or add information to the master record when I find errors or omissions, and I could link in to WorldCat to pull out basic bibliographic information that I could incorporate into resource discovery tools built and maintained locally. The WorldCat API is certainly a step in the right direction, but we have a long way to go. Is anybody at OCLC thinking along these lines and/or working on this?