Scholarsteve’s Blog

Just another WordPress.com weblog

Search Interfaces

I just finished the chapter in Information Technology on search interfaces. This is the longest chapter in the book and covered a lot. I think what I got out of it most prominently was the idea that a lot more though goes into creating a search interface (if one is even needed) then most would imagine. I think that the ISU homepage and especially the search options in Milner Library’s page represent a large number of the diverse search options available.

One of the methods the books recommended if a site is not large enough to justify a search system is the creation of an index. The index, which acts as a Web site table of contents, is useful to users who want a fast way to find key pages in a site. I realized that I use a type of index almost daily as a professional as a student and as a professional. ISU offers an A to Z search function that brings up an index that is organized alphabetically. Selecting E will allow access to English, which will bring the user to the English Department Web site. I almost always use this function rather than do a search of the ISU Web page due to the abundance of results retrieved if I were to do a search for it. The authors also site this as one of the issues with creating a search engine. Users can become overwhelmed by search returns that are too high.

I believe that in my case, this is what happens with the ISU search page. I would be interested to know if most other people prefer ISU A to Z rather than the search bar. Both are presented on the same page and given equal amounts of space. Analyzing the page further, I notice that if the user does a search, the only ways to organize the results are by relevance and date. Since there are such a large number of returns, I would like to see the options available to have results divided into documents and site pages. I think it is frustrating to users to click on a link that says “calendar” only to be directed to¬† pdf of an old newsletter calendar. It can be deduced if search result is a document if the user reads further down in the result or the actual Web site link. Dividing the results would make this unnecessary.

The site is readily equipped with a section for search tips right near the advanced search link. It is prominent, which is good. However, it takes the user off-site to a Google Web site. I am not against users leaving the site when necessary (afterall, they shouldn’t feel trapped), but wouldn’t it be a simple matter of creating a page with these same tips. When sent to a google page for search tips, it begs the question to the user, “Will all of these tips help on a non google search page?” I suppose if near the official search bar a phrase reading “powered by Google” would mitigate confusion, but this creates one more thing for the user to think about. By this point they are likely already frustrated that their search was unsuccessful on some level and that they need to revise. However, the site’s search does follow one of the book’s most stressed points, that when searches come back with no results, they should encourage the user with different ways to continue the search rather than shutting down communication.

The ISU search page returns with suggestions. I suppose this one point is something I think is not as rigid as the authors sugest. I believe that most relatively seasoned internet users are used to having searches occassionally coming back with limited or no results and will not even read the tips, but revise the search of their own accord. I suppose though that it doesn’t hurt to provide tips to those new to the Web.

Advertisements

January 20, 2009 Posted by | Information Architecture | 3 Comments