Skip to main content

One Search to Rule Them All

Why you still have to use multiple search engines to find library materials

Making the Myth Come True

So, "One Search to Rule Them All" is an impossible pipe dream. Or is it?

That depends on what you mean. If you mean "one search engine that will replace all the others and allow us to never use anything else", then yes, it's impossible. As impossible as the ball that can be used for every sport.

But if your goal is to build a search engine that can do a simple keyword search on the most common content types... then the answer is a resounding maybe. And given the demand from the public (because who wouldn't want a magic djinni that would do all our academic research for us?), the quest began for the holy grail of library science: a simple search interface for all library content.

I don't want to be too negative. A lot of progress has been made, and we've learned a lot about how to improve search engines in general. And after twenty years of development we've finally got some options that don't suck!

The Closest Things We Have... So Far

UCLA Library users currently have access to three web-scale discovery services:

All three of these are decent search options. What they are not, and may never be, is replacements for all the hundreds of other search engines out there. They're great for finding known citations or getting a good overview of a topic. But they're still "lowest common denominator" searching. There's no way to take advantage of the controlled vocabularies, specialized metadata fields, and other useful features you can get from a search engine built to work with a single metadata scheme customized for a particular discipline. That metadata is in there, but it's all jumbled up with other stuff.

Here's a real life example. Many medical researchers and students need to find articles based on clinical trials. In PubMed you can click on a checkbox labeled "Clinical Trials" to do exactly that. In ArticlesPlus, that phrase "clinical trial" appears in the records from PubMed... but it's not in its own field. So if you add it to the search box you'll also get articles that may not actually be about clinical trials, including articles with phrases like "though no clinical trial was conducted...." It will also miss any articles originally indexed in CINAHL, the major nursing database, which uses the phrase "randomized controlled trials" instead. You're better off searching in one or both of those specialized databases