What Google cofounder Sergey Brin calls “the ultimate personalized search engine: the librarian.”
Amazon.com: Search Patterns: Design for Discovery (9780596802271):… Amazon.com: Search Patterns: Design for Discovery (9780596802271): Peter Morville, Jeffery Callender:…
I recently attended a workshop entitled “NO Dewey – Rangeview Library’s Brave New World” at SCLS, presented by Rachel Fewell and Jessica Ransom. It was very interesting.
The basic premise was that, in order to effectively serve 21st century patrons, the library needed to discard DDC and adopt a more subject oriented method of classifiying books.
The Rangeview Library District recataloged the collections in each of their branches, one at a time. The Dewey numbers were replaced by word based classifications. This entailed changing each record at the item level, changing each label, and reorganizating the library into the new divisions. It is based on BISAC classification, which is used by bookstores.
I don’t believe that this is “a way to make it easy for people who don’t know anything about libraries and who don’t care about finding any particular book.” Looking at the grids, I can see a very structured classification system. A patron would be able to find a specific book using the new word based call number just as easily as using a Dewey number. It is just different (just like LC and Sears are different). It doesn’t drill down as far as Dewey does, but in some libraries, especially smaller ones, Dewey’s level of precision is not always required.
They removed the linear shelving and replaced it with more aesthetically pleasing shelving that allowed each subject to be displayed in a way that the patron could more easily see the collection. They improved the signage. They made an effort to display books in a manner that the patron could more easily see each individual book, instead of rows and rows of books. And they extensively weeded the collection.
I believe that unless a collection is relatively small, recataloging an entire collection would be cost prohibitive. And, while it is very intriguing what they accomplished, I am not convinced reclassifying is necessary to accomplish the same goals that the Rangeview Library District was working toward.
But what can we glean from the Rangeview Library District’s initiatives?
Working with the classification system we already have, we could improve how we display our collection, including shelving and furnishings.
Dewey is divided thus:
100 Philosophy & psychology
300 Social sciences
500 Natural sciences & mathematics
600 Technology (Applied sciences)
700 The arts
800 Literature & rhetoric
900 Geography & history
The terminology could certainly be adjusted – “generalities” is not helpful, but “Computers” (DDC 004-006) certainly is a helpful designation. Within the computer section, there are subsections on topics such as programming and using Windows. Displaying the Dewey century number (000 Generalities) by itself is not helpful to the patron who is interested in learning to use Excel. Of course he could ask the librarian, but isn’t it friendlier to allow those who choose to find it themselves a bit easier (or “save the time of the reader”)?
Our patrons would find browsing subjects easier, while working within the system into which our libraries (and our taxpayers) have already invested significant time and money.
In conclusion, while I do agree that many of our libraries are in need of an aesthetic face lift, there is nothing wrong with DDC, and the new word based classification system in and of itself is not necessary, nor does it improve on Dewey. But I do believe that this discussion can only improve how we approach serving the patron better.
Functional Requirements for Bibliographic Records—or FRBR—is not Marc, nor is FRBR changing Marc. RDA is the FRBR inspired code that adds the aspects of work, expression, and manifestation to the Marc record.
I found this Tweet recently, and thought it should be reiterated:
@dorevabelfiore Doreva Belfiore
#FRBR is a model, not a code. #RDA is the cataloging code that defines attributes to record.
I went to vote in the primary today, and tried out the new voting system.
Basically how it works is you get a paper ballot, fill in circles (think standardized testing in high school), carry it to a computer, and slide it into a slot, where it is read.
Friends, voting privacy is now history in Suffolk County, New York.
I was sent to a special desk to fill out the form. While there were side barriers, there was nothing on the top – and the public school where I vote has security cameras, as do most libraries and other public places where voting takes place. This was not a voting “booth” at all.
I then carried my ballot to the voting machine in the “privacy sleeve,” a piece of cardboard folded around the paper ballot. I slid it into the machine, and it “misfed.” So it popped back out, so that the election official and the random man standing behind me could see exactly who I voted for. The election official commented that this had occurred regularly throughout the day.
My undergrad degree is in computer science, and I am typically a huge fan of technology – where it makes sense. However, there were never any problems with the New York lever-style voting machines. The system wasn’t broken. Plus this new system is a huge waste of paper. The ballots were on large sheets of paper – legal size or more.
Apparently some government official felt that we weren’t spending enough taxpayer money.
This may be the “Facebook” generation, but I, for one, would prefer that the contents of my vote be protected.
As I stated in my article A Solution to Subject Access of the Library OPAC, it is the responsibility of those of us organizing information to allow the searcher to find his information accurately in as little time as possible. The OPAC should function as the librarian does at a reference interview, and deliver accurate results that answers the patron’s question. The issue is findability.
In his book Ambient Findability, Peter Morville states that the issue of findability is the primary issue in web development: “You can’t use what you can’t find” (Morville, 111). As stated in my previous article, the OPAC requires a custom designed “search engine;” quite different than the keyword based PageRank algorithm utilized by Google (Langville & Meyer). Many OPAC interfaces have, in fact, integrated a similar ranking system to the keyword search. Keyword searches are easy for the patron, but do not provide accurate results.
As an information professional, the organization of information is at the core of my philosophy. Just as the semantic web attempts to organize information available on the web, and make that information available to the information seeker, so does a redesign of the OPAC search engine makes information available to the library patron.
I discovered this very interesting article on the topic: Beyond the OPAC: The Semantic Library.
According to John Blyberg, “the possibilities of the Semantic Web invariably leads back to library service.” The reasons he list are:
- Libraries have the infrastructure in place already to take advantage of this technology;
- Libraries are positioned already as information gatekeepers;
- Libraries are not encumbered by the need to protect proprietary information–sharing the wealth of knowledge is in our bones..
Check out the article – it is a very good overview of the benefits and the issues related to developing this technology.
Roy Tennant stated (and has since retracted) that “MARC must die.” On my blog he commented “The basic point is that limiting ourselves to the MARC/AACR2 (or even RDA) standards is inadequate — we need a metadata infrastructure (and the skills to use it) that supports a wide variety of metadata standards. The days of MARC hegemony are over.” There are still those who believe the problem is MARC. All that MARC is is a standardized format. Nothing more. In this writer’s opinion, there is certainly a need for normalization. This is a software issue, not specifically a metadata issue. FRBR/RDA would be an important addition to the bibliographic record. The relational database model is still the best way to store and retrieve information. But the critical issue is the software interface as it is designed to interpret the user’s needs and retrieve the data. Once the database is normalized, the database queries themselves are trivial. But what we need is a better way of creating queries to produce accurate results. This is a software issue, not a metadata issue.
Findability is not an option. Findability fulfills Cutter’s collocating objective of a bibliographic system – to show what the library has:
- by a given author
- on a given subject
- in a given kind of literature
We live in the Google generation, and need to not only make search as easy as Google does it, but we need to do it better. Collaboration between information professionals and computer scientists can solve this problem.