If you have any notes on specifics of what your users like/need about 
browse search, they would be very useful to me.  We are currently 
engaging in that exersize of "determine what those things are, and then 
figure out if we can achieve them through methods other than browse search."

Jonathan

Tod Olson wrote:
> Bill, 
>
> Here are relative percentages for our Horizon catalog, based on our 2008-2009 annual report:
>
> Browse Searches	76.2%
> Keyword Searches	20.9%
> Mulit-index Searches	2.9%
>
> That interface presents a browse search box before a keyword search box, so browses are encouraged by the UI.
>
> That said, we did a study with our graduate students this year and they rely on browse searches for some of their academic work.  One is the use of subject and author browses, which lets  the student feel confident that they have been exhaustive in their searching in their area of research.  This can possibly be accommodated in other ways.  
>
> In addition to known-item searching, our grad students also use title browse to be confident that we do _not_ own something.  In our relevance-ranked interface, sometimes the scholar may blame relevance ranking for "hiding" a title from them which we don't actually own.  It's an understandable reaction.
>
> -Tod
>
> Tod Olson <[log in to unmask]>
> Systems Librarian
> University of Chicago Library
>
> On May 3, 2010, at 1:08 PM, Bill Dueber wrote:
>
>   
>> I got email from a person today saying, and I quote,
>>
>> "I must say that [the lack of a browse interface] come as a shock (*which
>> interface cannot browse??*)"
>>
>> [Emphasis mine]
>>
>> Here, a "browse interface" is one where you can get a giant list of all the
>> titles/authors/subjects whatever -- a view on the data devoid of any
>> searching.
>>
>> Will those of you out there with "browse interfaces" in your system take a
>> couple minutes to send along a guesstimate of what percentage of patron
>> sessions involve their use?
>>
>> [Note that for right now, I'm excluding "type-ahead" search boxes although
>> there's an obvious and, in my mind, strong argument to be made that they're
>> substantially similar for many types of data]
>>
>> We don't have a browse interface on our (VuFind) OPAC right now. But in the
>> interest of paying it forward, I can tell you that in Mirlyn, our OPAC, has
>> numbers like this:
>>
>> Pct of Mirlyn sessions, Feb/March/April 2010, which included at least one
>> basic
>> search and also:
>>
>>  Go to full record view      46% (we put a lot of info in search results)
>>  Select/"favorite" an item   15%
>>  Add a facet:                13%
>>  Export record(s)
>>   to email/refworks/RIS/etc. 3.4%
>>  Send to phone (sms)         0.21%
>>  Click on faq/help/AskUs
>>     in footer                0.17%  (324 total)
>>
>> Based on 187,784 sessions, 2010.02.01 to 2010.04.31
>>
>> So...anyone out there able to tell me anything about browse interfaces?
>>
>> -- 
>> Bill Dueber
>> Library Systems Programmer
>> University of Michigan Library
>>     
>
>