Pricey gourmet store pasta salad–it’s prepared in a way that appeals to everyone. It looks great, has all the tastiest ingredients but, for some reason, it doesn’t taste as good as your own–-because you make your pasta salad exactly the way you like it. With just the right touch of “this and that” and none of what the rest of the crowd might think is yummy, but plenty of what you know is perfect. I’ve found the same holds true for search engines. We can use the Google, Bing, or any of the other off-the-rack search engines for most of our searches, but when we are willing to spend a little extra time to customize our search experience, we are gaining a foothold in creating a part of the Web’s infrastructure.
When we think of user-generated content, we think of photo sharing, blogs, and videos, but we should also think of phenomena that put us in control of our Web experience. While quirky utilities such as Yahoo! Pipes (and readers who have firsthand experience with Pipes will, I think, agree that while Yahoo! has afforded us the power to design our own information experience, the first word that comes to mind when we think of Pipes is “buggy”) fall into this category, I’m sure most of us who have experimented with CSEs have found them to be a useful resource in our toolboxes. CSEs permit us to direct our queries to the Websites we have determined are most useful.
A CSE is a search engine we can program. It allows the user to query the Web sites the user has determined are the most relevant. Only a handful of developers offer us a way to do this. The most basic is Rollyo. A more complex option is Gigablast. The most popular build-your-own search engine is probably Google’s Custom Search Engine.
Why build a custom search engine? At times we may want to zero in on Websites that will help our patrons locate information that the big search engines will bury in 100s of pages of retrieval. For example, Terry Ballard at New York Law School’s Mendik Library configured a Google CSE called DRAGNET that searches recommended legal Web sites. Sometimes CSEs are built to replace utilities that have been abandoned by their original agencies. Tara Calishain recently configured a cool CSE that searches official United States state Web sites. She found that, because Google Uncle Sam search had closed over a year ago, her engine could fill that void.
My book UContent covers not only the different ways librarians are using social media, but also details my firsthand experiences with each type of user-generated content I discuss. In Chapter 11 my coverage of CSEs led to building several of them– take a look at the pages and give them a test drive. I used Rollyo, Google, Gigablast, and a couple of the less well-known CSEs.
Anyone with a particular information focus may want to try developing a CSE. To get an idea of the broad range of CSEs that individuals have worked on, take a look at the Directory of Google Custom Search Engines or the Custom Search Engine Links Directory. That’s another advantage of CSEs – once someone builds one, you can use it too. They’re easy to configure. You can start building yours right here.