As promised at June's reference department meeting, here is a place for a discussion of federated searching. We have received funding for and are planning to purchase Serials Solutions' 360 Search (formerly called Central Search), with the hope to implement, at the very least in a beta mode, this fall. In case you missed the June meeting, I asked three questions of the ref staff just to begin a very preliminary discussion:
1. What are your concerns about a federated searching implementation?
2. If you haven't already, have a look at this rough draft of a new library homepage. What should go under "Start My Research"?
3. How will federated searching change the way you teach and/or conduct reference interviews?
Some pros: In my experience, the biggest benefit to teaching with federated searching was that you only had to teach them one interface. There will obviously be times when you will not want to use it; for example, a very specific assignment for which one or two of our online resources are particularly relevant or required. Integrating federated searching into the reference interview was not difficult; a general search for a 101-level class is a great candidate for using it, while an advanced legal query is probably not.
Now the cons: This obviously isn't a silver bullet. It's *another* change to get used to. And while federated searching has come a long way, searching a database's "native interface" will still likely produce better results because each database product has its own indexing, its own way of managing Boolean searches, etc. However, our main goal in implementing federated searching is to make the wealth of our subscribed online content more visible and easily accessible to our patrons, while at the same time making our home page more usable and in line with their information seeking expectations.
Concerns expressed at the meeting yesterday included slow search speed; giving enough lead time to plan for changes to teaching for fall semester; how the faculty will respond; and how we should market it to the faculty. Please continue the discussion below! And please don't feel like you're resisting change or being negative by raising concerns. If I know what everyone's thinking about and concerned about, I can hopefully address these concerns and make a better end product for us and our patrons. Someone else requested that I post or link to some examples so that we can see what it looks like on other library home pages. That will come in a follow-up post.
In the mean time, here is a generic trial of 360 search. Please take the time to conduct some searches using it to see what it's like. We will have a very short time to implement and train, even if we only roll out a beta version!
http://demo.cs.serialssolutions.com/demo
User name: sersol
Password: search
For more information about how and why 360 Search was chosen, see: http://carbon.cudenver.edu/~nmchale/federatedsearching/demos.htm
(The page is password protected, and the user ID=federated, and the password=searching.)
7 comments:
My concern about Federated Searching is that you're going to roll out this product in a short space of time and have no metrics or expectations about how it will be used.
Redesigning your Instruction curriculum to accomodate the product is an even longer process. Just as you may have project milestones for the rollout of the Federated Searching software product, you'll need to match those up with some Instructional milestones. The day that you roll it out, i.e. in two months is the same day that it's probably not ready for prime time instruction.
If you're not willing to take that into account, then Federated searching is no better than any other database product turned out and it should be run parallel with all the others until such time as we can figure out its relative value, if at all, least of all redsign of the library home page to accomodate it.
I don't see that the imperative or the case to redesign the library homepage to accomodate or support federated searching has been made at this time, much less to support instructional methods.
Initially, I'm concerned that Federated searching removes many kinds of academic research drills & routines one or more steps from reality or even the start of basic research which is either "X" number of suitable hits and their evaluation or enough "full text" journal articles from which to choose or which can be used as a literature review to support potential and as yet undefined research topics.
Going into the product, one still has to be aware of which databases might be appropriate as opposed to "select all."
An example is in order: Student asks for help in researching a topic on Birth Order and Criminal Behavior using Adler's theories
The first lesson of this kind of topical approach to a complex subject is that the ultimate or penultimate search strategy is not well-formulated, at this point and may not be in the correct domain without an awareness as to whether the selections offered under Federated Searching are even appropriate. In this example, the student knew enough to start with PsycINFO but not enough to understand why PsycARTICLES might be a better place to start, but only a place to start, assuming that they need to leverage full text resources first rather than lists of citations. However, the issue of an initial search strategy that worked was still up for grabs, even where the Student had already in hand 5 copies of works by Adler but had not yet even looked at them for sources or concepts. So everything was quite premature, even to the formulation of a research strategy or thoughtful approach or the statement of what was to be the actual topic.
This kind of "topical" google driven scenario is exactly what Federated Searching will not support without some critical thinking about what it means to teach students how to search psychology databases or to determine that their topic is cross disciplinary from the outset, or to even devise a search strategy that is usable with a Federated Search product.
Thus, beginning to teach research methodologies, outside of any knowledge about specific disciplines, i.e. psychology, communications, marketing etc. is a different proposition than teaching the craft of "go first to the psychology databases," or go first to the "aggregators" such as ASP, OneFile or LN and scope out your topic. It's not clear at this time how Federated Searching will support "Instructional Goals and Objectives," which have not yet been defined either, least of all ready for rollout in 6 months or less.
My 2$ Worth
W.A. Tietjen
Bill wrote:
"I don't see that the imperative or the case to redesign the library homepage to accomodate or support federated searching has been made at this time, much less to support instructional methods."
Please refer to last fall's usability testing report for the case having been made:
http://carbon.cudenver.edu/~nmchale/usability/finalreportfall2006.doc
I'm not asking for a redesign of the intruction curriculum. I'm asking that you all, as instructors and reference librarians, brainstorm ways to use this new product.
After doing a search on bacterial transformation I am concerned about the length of time it took the search to come up. Also, I feel the students will be very, very confused when trying to search our resources. I tried all sources which I feel the average student would do and it took a long time and was confusing as to what was a book and where it was located. I have seen ways I could teach it to try to clear up confusion but a majority of our students never have an instruction class and even many in the classes don't pay attention. It sure will get students to the reference desk if we try to bring this out this fall. I would really like to have a beta to work on all fall semester prior to launch because teaching this will be a challenge for me at least. Diane
I have a few questions, Nina:
Can we set a limit to full text only, as is possible in EBSCO &
INFOTRAC?
How about limiting to refereed
articles?
I presume we can also search the catalog, such as the Univ. of MN,
correct? I like their web page the best.
I hope you can join us on Monday
for this important instruction meeting.
Answers to Louise's questions:
Q: Can we set a limit to full text only, as is possible in EBSCO &
INFOTRAC?
A:Yes.
Q: How about limiting to refereed
articles?
A: Not sure; will check and have answer by Monday's meeting.
Q: I presume we can also search the catalog, such as the Univ. of MN, correct? I like their web page the best.
A: Yes, Skyline will be included.
Q: I hope you can join us on Monday for this important instruction meeting.
A: Will be there with bells on! :)
A:
I appreciate it that you reminded all of us that you did conduct a research study aimed at retooling the library homepage. Thank you.
I will freely admit that the devil is in the details. My angst is more with the project methodology than the need. When we talk about doing both, i.e. rolling out a new library web page with federated searching could we please have a timeline showing milestones & deliverables. Then we can all argue about whether we need 60 or 90 days.
Right now the vendor version of the Federated search tool does not emulate how it would be integrated into the library homepage. Your sample library page comes closer. But I would also like to know how long we will run parallel or when cutover would occur or what other usability testing is required either before or after the initial product is delivered.
So the advantage of looking at other library web pages is to see how it might be integrated. I do not care for the product itself as a standalone. But I like it when it is integrated in intelligent ways. This means that you will always have a customized solution and never the raw native product.
Features of other sites I liked or disliked with federating searching:
UMN:
1) Quick Start
2) What Am I searching? and what's behind it.
3) Assignment Calculator
4) Full Text Finder
UW:
1) Ask Us [however, we're not ready for this; we have a terrible customer service external look and feel that starts with our historic refusal to do much phone service]
2) Not thrilled about the use of the WorldCat beta
UIUC: Did not like this page at all
HSC:
1) Favorite Tools
Last Words:
My previous comments were a vain notion that Instruction would get active and drive some decisions.
No matter how good a job we do of delivering a new library web page and federating searching, our work only begins the day we roll it out. It we ever catch ourselves sitting on our thumbs for as long as we have with the currrent web page product, we should all be fired. By this I mean that installing a process of continuous improvement and redesign is what is also wanted. The world is not waiting for the Auraria Library as much as it is moving so much faster than our design and implemention decisions allow. We need a new process for product rollout that is incremental and probably uses a 90-day cycle from concept to delivery for product enhancements, aka RAD.
--W.A.Tietjen
Just some thoughts before the bibliographers have a demo this week.
-I hope that we'll continue to offer multiple access points to accomodate different learning and teaching styles, i.e. having the ability to quickly cross-search combinations of records representing online and print works as well as being able to efficiently search by one known title; one known format; one subject, or to even have a default option for those who think more linearally (I know - alot to ask for!)
-I would like to see Fall be a beta time rather than a roll out time. When we make small and large changes on the web page I'm generally contacted by a few faculy folks who ask "where is..?" since they're used to the old way and can misunderstand the new way even if the new way is improved. Sometimes they assume we no longer have something available (like when the dissertation database changed names!). Also, it's common for faculty members to integrate the first few steps of searching the Library's web page into their assignments. Since school starts August 20th and many faculty members won't even return for awhile, I don't think that gives the faculty enough time to learn the new interface and update their assignments. I'd like to see us heavily promote a beta version with the instructions that faculty (and students) should play with it because it will definitely be rolled out in the spring.
--I'd like to see it be simple to search for a book. Since our heading that said something like 'find books' went away, I've had more patrons come to the desk and just ask 'how do you look up books.'
Our current pages have been filled with some wonderful innovations so I look forward to holding on to our best stuff!
Bloggily yours, Ellen M.
Post a Comment