The index re-mixed

Once upon a time, there were three ways to find an article on the topic you were interested in.

1. Browse the journals until you found something relevant and then use that article's bibliography to find other articles
2. Ask someone well-read if they know of an article on your topic
3. Use a periodical index

Nowadays there is an additional option

1. use a search engine like Google

It is very difficult to determine how many academic library users are using Google instead of library-licensed periodical indexes in their researching. HighWire reported that Google is responsible for 56% of the referred traffic to their journal publications. (if any of you out there know of similar evidence, could you let me know via a comment or email?)

At the same time, libraries are now offering more indexes than ever before. But how many index choices are enough? And perhaps more importantly, are we offering too many alternatives? There is a growing recognition that we are scaring many of our users away with too much choice. To solve this problem, we have been investing in metasearch. It's a compromise that allows librarians to keep all their expensive, specialized indexes while offering only one interface to their users.

So why not allow users to customize and save their own metasearch engine not unlike Google's customizable search engines?

And what if recognize that we have a track record of not being able to match our users needs with periodical indexes and be done with them completely? Let's dive down another level of granularity and create a platform that allows users to create their own customized index that searches just the subjects and / or the journals that interest them. Why are playing tens of thousands of dollars for someone else do this for us?