ALA   American Library Association Search ALA      Contact ALA      Login     
ACRL home contact us search ACRL sitemap home join acrl
50 East Huron Street, Chicago, IL 60611, T. 800-545-2433 ext. 2523, F. 312-280-2520
 
 
About ACRL Issues & Advocacy Events & Conferences Professional Tools Publications
Standards & Guidelines Awards Give to ACRL President's Page
 
 Publications
 ACRLog
 College & Research Libraries News
 College and Research Libraries
 CHOICE
 Academic Library Statistics
 Books/Monographs
 Downloadables
 RBM
 White Papers and Reports
                         



Ten ways to improve your Web site: Take time to dust off those Web sites

C&RL News, September 1999
Vol. 60, No. 8

by Jim Cunningham

If your library is like most, you established a Web site two or three years ago. The project probably captured a great deal of attention in its early days as folks tossed around ideas as to what should or should not be on the site. You (hopefully) reached a consensus on design and content, people hacked away at HTML for a while, and there was much rejoicing when the site went live.

Then everyone went back to work, and, though for the most part the site has been kept up-to-date, the additions and modifications to the site did not get the same degree of attention as the premiere did. Sections were added piecemeal and changes were occasionally made on the fly in the pressures of time. In short, the Web site ended up being an unforeseen mixture of parts that may not be congruous, and now it’s time to fall back, look at the big picture, and pull things together.

Hopefully that idea got your attention. “Yes!” you think to yourself, “But where do we start? What kinds of things should we be looking for?” While some aspects of improving and cleaning up a Web site are particular to the goals and needs of the individual site, there are general improvements that can be made on any Web site that will make it easier to use.

1. Be concise. As the Web grows and the library adds online services, the temptation to put a link for every new feature or service directly on the library homepage can be overwhelming. Before you know it, the links will start piling up and there will be many more of them on your homepage than any patron will want to sort through.

There are several possible solutions to this problem. The most obvious is to exercise some “collection development” in adding links to your homepage. For example, does that link to local weather conditions really need to be there? When the library Web site was new, the Web was smaller, and the homepage was begging for links, these types of links were wonderful for filling up space, but are they justified now that your homepage space is at a premium?

If there are more links on the homepage than you care to delete, consider reorganizing them. Intermediary pages of lists are one option, but only partially solve the problem in that a user must confront several pages of links. A better option is to use pull-down menus on your homepage, allowing you to keep all of your links on one page but in a less overwhelming format. Instead of being bombarded with dozens of links at once on a single page, users can view several broad categories, pull down a menu and review the specific sets of links, all without having to move from page to page.

2. Avoid “link rot.” This condition will afflict even the best Web sites without due vigilance. Few things can annoy a user more than “not found” errors and inspire them to leave your site. This condition also announces general institutional sloppiness to literally the entire world. Institute and maintain a strict link-checking policy. A quick calculation of the number of links on your site times the amount of time each takes to check will probably result in an alarming amount of work that will make a Webmaster cringe, but worry not—there’s help available.

Various software packages such as InContext will run through your entire Web site and check every link, be it internal or external, and provide a list of those that failed. The only problem is that if a connection is made at all, even to a page that says “This Web site is dead,” the link will show up as functional.

3. List updates to your site. Little in life changes so quickly as the online world. For this reason, currency is essential. How many times have you looked at a Web page wondering when it was last updated? Users looking at your library Web pages will wonder the same thing. Listing page updates is simple and should be unobtrusive; a small notation at the bottom of the page is sufficient, in most cases. Your users will know they are getting current information, and placing the dates on every page provides an incentive that will inspire Webmasters to ensure currency, especially if their name and e-mail addresses are also on the page. And no fair changing the dates without actually updating the Web page!

4. Coordinate, coordinate, coordinate. This point cannot be overemphasized. For a Web site to work properly, especially a large one involving the work of many individuals, all of the parties involved with its development, maintenance, and general functioning need to talk to one another frequently and in detail. Failure to do so will result in a Web site that has no consistency at best or with numerous failed internal links at worst.

One librarian I talked to told me that each department maintained its own segment of the library’s Web site more or less independently. I asked what happened when one department decided to revamp its components of the site and suddenly most of the links to their section stopped working.

To my horror, the librarian just shrugged and said, “It happens.” Folks, this should not happen, ever! Be certain that the individuals responsible for maintaining your site coordinate their efforts, and preferably put a single individual in a position to monitor and ensure the site’s operational effectiveness.

5. Get thy Web site a dedicated search engine. When your site debuted, you probably had a couple of hundred pages. You now have many more, potentially thousands more. This is a good thing, but it’s also akin to adding books to a print collection—there is more for users to wade through to find what they want. This can be a problem no matter how well your site is organized or how good your interface is. To solve it, get yourself a search engine.

There are two ways of doing this. The first is to set up a search interface page that points to your favorite Web search engine (Yahoo, Hotbot, Excite, etc.).

You can also set up a Web page that uses a commercial search engine, such as Alta Vista or HotBot, but limits all searches to your library’s Web site, rather than the entire Internet. The advantage of this arrangement is that you don’t have to do much—set up the Web page and accompanying scripting and you’re done. And the price is right.

The disadvantage is a lack of control: you do not have the option of updating the search engine, nor do you have much choice of how the searches are done in the engine’s inner workings. For example, page updates on your site may not show up for weeks on a commercial search engine. You also have no control over how the search engine will search in its database—word arrangement, upper or lower case usage, etc.

If you have your own Web server and a person with the time and technological skills needed, a better option is to install your own search engine exclusively for your site.

There are several of these available, with some such as GLIMPSE from the University of Arizona, being free. Using a local engine enables you to control update frequency, thus ensuring your users always get current page information. In addition, there are numerous interface design and search options available, enabling libraries to provide the exact configuration they want.

6. Be clear to your users. Review your site, especially your homepage, from the user’s perspective. A homepage that is perfect from a librarian’s perspective may only be marginally useful to a user. For example, labeling the link to your commercial online services “databases,” while technically more accurate than using “article indexes,” will be less understandable to the average user seeking articles in journals or popular periodicals. Remember: the library’s Web site is for library patrons, not librarians.

7. Do not build anything that cannot be maintained over the long-run. What will happen to the fancy, Java-based interface on your homepage if the one library staff member who programs in Java decides to move on?

Make your site only as sophisticated as can be maintained by the average level of Web skills available at your library. If your library is “Web-staff challenged,” consider an alternative such as outsourcing coding to a reputable vendor or perhaps developing an agreement between local libraries to look after one another’s sites in the event of difficulty. The latter arrangement is especially helpful for smaller libraries with few staff members.

8. Keep your Web site modular. Eventually your site will require a major redesign; some parts will be discarded while others will be retained with only minor modifications.

Life will be much easier if the sections or pages to be retained can be moved to their new directories without making any changes to their HTML links. To do this, design your site with directories that group related pages or sections, and use relative rather than real links. (Dust off your HTML book if you’ve forgotten the differences between those.)

Failure to do this may require your Webmaster to recode hundreds or thousands of links on your site during an overhaul. They will not be happy about this, to put it mildly.

9. Retain text-based links where possible or create text-based auxiliary pages. Talking Web browsers for the vision-impaired do not recognize graphics or the lettering placed on them. Without text-based navigational capability, users with such browsers will be stranded and your site will be useless to them. And, believe it or not, there are still a few users out there with their graphics turned off to reduce page loading times! And Lynx still lingers on.

10. Keep commonly used files in a single directory. Image files such as page backgrounds, logos, and the like should be kept in a single directory on your Web site to avoid repeated and unnecessary downloading.

If a commonly used background file is placed in every directory on the site, a Web browser will pull the image down over the Internet again and again, wasting bandwidth and the user’s time. By placing a highly used file in a single location, a user’s browser need only download it once per session—the file will remain in cache on the user’s machine.

Library Web sites used to be for net surfers or for use on a few Internet workstations in the library. In the last couple of years, however, the trend has been to make the library Web site the standard interface on all public workstations in the building as well as on the Internet for remote users. This makes the library Web site in general and the homepage in particular of paramount importance—they will literally be the gateway to the library. Make sure your design is the best and most current it can be.


About the author
Jim Cunningham is systems development librarian at Illinois State University, e-mail: jlcunni@ilstu.edu





ACRL is a division of the American Library Association
© 2008 American Library Association. Copyright Statement
Last Revised: May 21, 2007