A simple, simple search
A cornerstone of the redesign will be a truly simple, simple search. One search box, one button, nothing else. A single search box is only feasible if the search "behaves" intuitively, and this one does: words within quote marks are searched as a phrase, words outside of quote marks are automatically boolean ANDed, and the search is executed as a keyword anywhere search, with relevance ranking [1]. If that sounds Google-like, it was meant to be.
Why haven't we seen similar simple search interfaces for WebVoyáge? The search logic is made possible via an enhancement to Voyager 6.1 and thus is available to all customers, but the stripped down interface requires the use of a wrapper script and that takes some programming.
To judge for yourself whether this is an improvement, compare it to our current "simple in name only" search.
[1] Per the GKEY^* definition in the Sysadmin module and WebVoyáge config files. Note that the relevance ranking algorithm is far from ideal.
Why haven't we seen similar simple search interfaces for WebVoyáge? The search logic is made possible via an enhancement to Voyager 6.1 and thus is available to all customers, but the stripped down interface requires the use of a wrapper script and that takes some programming.
To judge for yourself whether this is an improvement, compare it to our current "simple in name only" search.
[1] Per the GKEY^* definition in the Sysadmin module and WebVoyáge config files. Note that the relevance ranking algorithm is far from ideal.
posted by Michael Doran at 2:29 PM
5 Comments:
An article in a recent Library journal, http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=22977267&site=ehost-live , reports focus group and usability testing of a statewide catalog & borrowing service. While the system tested was not exclusively an OPAC, it did involve catalog features, some non-bibliographic such as copy number. The gist of the findings is that users are dissatisfied with thousand-hit retrievals but they're unwilling to delve into advanced search features. Yet this is not a strong conclusion--most people were flexible and understood the problems of a simple search interface for complex information objects.
The article is not directly on point about a Google-like OPAC. It cites Judy Jeng, who did/does usability-related study of digital libraries ( http://web.njcu.edu/sites/faculty/jjeng/Content/my_cv.asp#publications ).
Oops, sorry, forgot to hyperlink the URLs. Let's try them again:
The LJ article:
http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=22977267&site=ehost-live
Judy Jeng's publications:
http://web.njcu.edu/sites/faculty/jjeng/Content/my_cv.asp#publications
The simple-simple OPAC search is wonderful in its own rank.
But apparently, there are OPAC search and browse functionalities inadvertently lost in the process of bring out the simple-simple OPAC. Before the switch-over and get-go, the simple-simple OPAC search is explained and advertised; however, the potential loss of some existing functionalities were not warned. Thus the lack of awareness of such loss did not set users up to voice concerns and disagreements. Many users were taken by surprised by the loss.
While the simple-simple search interface is geared for the “unsophisticated” OPAC users, the loss of existing OPAC search and browse functionalities does not suit for advanced users. If the OPAC only serves the “unsophisticated” users, the loss of functionalities will be negligible. But when 89% undergraduate students do not start with their search in OPAC, the loss of OPAC search and browse functionalities affect teaching and research of faculty, staff, and graduate students. These users may be in the minority in terms of numbers, but they are the ones take the most advantages of the OPAC. A graduate student is four times of undergraduate student when it comes to funding. Undisputedly the research and teaching needs of faculty are critical to the university’s mission as well as the library’s. It is noble to tailor to the needs of the “unsophisticated” OPAC users. But it is contrary of the library and university mission statements to ignore faculty and researchers’.
The simple-simple OPAC search is wonderful in its own rank.
But apparently, there are OPAC search and browse functionalities inadvertently lost in the process of bring out the simple-simple OPAC. Before the switch-over and get-go, the simple-simple OPAC search is explained and advertised; however, the potential loss of some existing functionalities were not warned. Thus the lack of awareness of such loss did not set users up to voice concerns and disagreements. Many users were taken by surprised by the loss.
While the simple-simple search interface is geared for the “unsophisticated” OPAC users, the loss of existing OPAC search and browse functionalities does not suit for advanced users. If the OPAC only serves the “unsophisticated” users, the loss of functionalities will be negligible. But when 89% undergraduate students do not start with their search in OPAC, the loss of OPAC search and browse functionalities affect teaching and research of faculty, staff, and graduate students. These users may be in the minority in terms of numbers, but they are the ones take the most advantages of the OPAC. A graduate student is four times of undergraduate student when it comes to funding. Undisputedly the research and teaching needs of faculty are critical to the university’s mission as well as the library’s. It is noble to tailor to the needs of the “unsophisticated” OPAC users. But it is contrary of the library and university mission statements to ignore faculty and researchers’.
The old "simple search" was not really simple: users had to master many arcane search conventions. For example, an author browse search had to be entered last name first. For a left-anchored title search, the user had to omit any initial articles. To do a subject browse search, the user had to be familiar with Library of Congress subject headings. In addition, some of these searches resulted in the user getting an index browse screen rather than a results list of items. For library staff and for many faculty who have mastered these search conventions, it's not that much of an impediment. However, for most of our users, that search interface was anything but simple.
When the WebVoyáge Advisory Committee did a presentation to Library staff back in January, we explained what the new simple search would look like and how it would function. We also explained that it would mean losing some of the functionality of the old simple search. We also tried to illustrate this fact with the "current and proposed" screen shots [1]. We stated that we would retain the advanced guided boolean search and would add as many search options to that page as Voyager would allow.
During the presentation to staff on January 10 and again at the session on June 29, we made the point that the quotation marks around the word unsophisticated meant that while, users may not be expert OPAC searchers, even new college freshmen have been searching on the internet for many years and their search expectations are shaped by that experience. The Library made a choice to harness the search conventions that were already familiar to most of our users rather than trying to make them conform to our systems.
While attempting to make the default search screen simple, we knew that it was important to retain an advanced search for the more expert OPAC searchers. The previous "Advanced Search" which is a guided boolean keyword search is still available. In addition, we have added some additional search types (e.g. ISBN and ISSN) to the Advanced Search. We also added, although somewhat clumsily, a left-anchored journal title search to that page.
The WebVoyáge Advisory Committee made a fully functional version of the redesigned OPAC available to library staff for four full months prior to the switch-over date of May 14. With the presentations, the blog, and the functional prototype, I'm not sure what else we could have done as far as awareness for staff. Faculty and students had considerably less notice about the change, but we did make them aware of it via a new item on the Library web site, and they also had an opportunity to try a prototype of the redesign for a few weeks prior to the switch-over.
I don't think anybody would disagree that the "research and teaching needs of faculty are critical to the university’s mission as well as the library’s". Neither the WebVoyáge Advisory Committee, nor any Library staff that I am aware of are attempting to ignore the needs of faculty and researchers. According to the Library's Mission Statement, on of our goals is to "make information access easy and convenient" [2].
It's typical that users that are happy with a changed search interface do not provide feedback and that the users who do provide feedback tend to be unhappy with the changes. Considering the extent of the changes made, the committee has only gotten a small handful of negative feedback from faculty. We are making note of the comments and suggestions, and we are looking at a different approach to the Advanced Search page that would retain the left-anchored searches/browses from the original Simple Search tab [3]. One of the issues with that, however, is that there are some users who would prefer that the guided boolean keyword search be the advanced search, so it's a matter of coming to a consensus as to what approach best serves our users.
I hope that has answered your question. If not, please respond and I will give it another shot.
Thanks for your input.
-- Michael Doran
[1] UTA Library Catalog Design > Initial presentation to staff
http://opacarama.blogspot.com/2007/01/initial-presentation-to-staff.html
[2] The Library Mission Statement
http://library.uta.edu/planning/missionStatement.jsp
[3] Voyager Test Server > WebVoyáge
http://vts.uta.edu/cgi-bin/Pwebrecon.cgi?DB=local&PAGE=First
(This site is our development platform and is constantly changing. The Advanced Search page configuration is available to view now, but may not be available in the future.)
Post a Comment
<< Home