Ebooks: The Library Catalog and Federated Searching Part 1

After participating and watching the MLA ebooks webinar two things became very apparent to me. 

  1. Patrons do not use the catalog
  2. We need a federated ebook search system

If I tried to address both of these issues it would be a very long post, so today I will discuss the catalog and tomorrow I will discuss federated searching.

Patrons do not use the catalog:

We aren’t the only library to notice this problem.  When most of your library’s information content is in the catalog and when patrons aren’t using the catalog, they aren’t finding the information.  I blame librarians and ILS companies. 

Why do I blame librarians?  We are on the front lines, we should be seeing how our patrons are searching (or aren’t searching) and adjust accordingly.  Yet we really don’t completely do that.  If we did then we wouldn’t be cataloging in MeSH!  I like MeSH, I really do, I think it is the best way for me to search for literature in database like Medline.  But really only librarians are the ones who speak MeSH.  The general population does not.  MeSH is the Esperanto of the medical library where only a select few of learned individuals know and use the language yet the vast majority of the population doesn’t. 

Honestly, I only really use MeSH when I search literature databases which contain millions of articles on various subjects.  When it comes to searching the catalog I usually search using keywords, like most of the library patrons.  So why are we even bothering adding MeSH terms to the catalog itself?  Most of my keywords (and I am a librarian) and certainly most of the patron keywords aren’t MeSH, they are at best general subject terms. 

Earlier this week Julie Stielstra posted on Medlib-l a question about alternative cataloging systems.  She described how a public library began to catalog their nonfiction differently by using “plain language” subject headings with author lables.  For example: SPORTS BASEBALL Bouton or COOKING FRENCH Child.  She wondered if her patrons wouldn’t be better served if she cataloged items like this as well.  Her example was NURSING PEDIATRIC Wong  2010 and I kind of agree with her that it is much more intuitive than WY 159 W559e 2010.

Perhaps we need to really investigate why we insist on using MeSH when clearly our patrons don’t want to use it.  Teaching them to use MeSH for Medline searches is at best a challenge, getting them to use MeSH to search a library catalog is sisyphean. 

For those who are ready to strip me of my librarian stripes, you can still have your MeSH cake and eat it too.  Go ahead keep the MeSH in the record but start adding some general terms that make sense to patrons.  I would love to say, let the patrons add the terms, but that won’t fix the problem.  Patrons don’t use our catalog, and by doing that we would be relying on the few that do search it to take it upon themselves to do the tagging of the collection.  Librarians should start tagging the collection themselves so that there is at least a skeleton set of terms for people to work with and build upon.  Giving them a blank canvas and telling them to paint a master piece is not fair to them.  We have to get them started with paint by numbers first.

Why do I blame ILS companies? 

Because librarians can only do so much.  Most of medical librarians are not programmers nor have the time to create a robost ILS that is required these days.  Therefore we need ILS companies to do that.  However, ILS companies are still designing systems with librarians as their primary users not the patrons.  The librarians are not the primary users.  We are the primary users of the back end but not the system. It seems ILS companies don’t know how to design a system that marries the back end necessities to a patron centered front end.

Patrons want an Amazon.com or Barnes and Noble like system, and quite franklyI have not seen an ILS out there that provides that experience.  Some systems are trying to do better, for example Innovative Interfaces just released a news statment about their AirPAC product for smartphones and its use in libraries.  Those kind of enhancements are helpful but the over all experience of ILS products is still pretty dismal.

Here are examples of different libraries or library system’s catalog records for Hurst’s the Heart.  (Names of libraries have been removed.)

  1. Example 1 from a group of small hospital libraries.
  2. Example 2 from an academic medical library.
  3. Example 3 from an large library system.

Which one is better for the patron? 

Example 1 is just a mess of words with no break for the eye and a bunch of gobblty gook that the patron doesn’t care about.  The call number is in the upper left hand corner like a card from a card catalog.  In fact the whole record is pretty much organized like a card from a card catalog.  Get rid of this design/organizational and display method.  Most patrons these days have never used a card catalog so they don’t “get it.”  Hell we have librarians now who never used a card catalog. It is just more of a mess for them to look at and they have to hunt for pertinent information.

Example 2 is better visually but is still kind of a jumble of words (especially in the TOC). Other things that are odd to a patron, do you really need that many words to describe format and does that all make sense to a patron?  Notes does not mean the same thing to patrons as it does librarians, do we need to show that?  I don’t know, I was always told in library school that people like to know if it has an index, bibliographic references, or illustrations but I have rarely had patrons ask me this when I am looking for a book for them.  They want to know if we have it and if so where can they find it.

Example 3 is the best of the bunch, but it too could use some improvement.  I love the picture of the book in the right, that is helpful to see.  (I realize the other examples were to the online book and may not have had images, but why can’t they if they are the online version of a printed book?)  The two biggest things that the patron cares about, does  my library have this and how do I get it are up top just below the title information.  I am not a big fan of adding links to Google Books if the book isn’t free or available through there.  I think “Limited Preview at Google Books” is not helpful to the patron (How limited? One time only? Can I print? Just the first chapter or TOC? etc.)  This is a large consortia of libraries so the call number which is unique to each library is not listed at the top, but patrons can click on the link to the libraries that have it to see the call number.  (I’m not sure that this is intuitive but I am also not sure how else you would do that within a large group catalog.) Finally the TOCs are arranged in a readable manner with links to the authors of the chapters.  That is very helpful.  Only at the bottom of the screen is the librarian cataloging information, patrons are rarely interested in it and it should be that far down.

I realize that some of the examples not only reflect on the ILS but also the library or libraries that set up their catalogs, but do you see any that are as easy as Barnes and Nobel or Amazon.com?  If so I would love to take screen shots and list them here as good examples.  I would also like to know how their usage is and what those librarians report about patrons using the catalog.

7 thoughts on “Ebooks: The Library Catalog and Federated Searching Part 1”

  1. I’m curious about your proof showing patrons not using the catalog. I work for a consortium of five libraries and “everyone” kept saying the OPAC wasn’t used. I implemented Google Analytics and found that actually, the OPAC gets a LOT of use. I’m still trying to analyze what KIND of use at this point.

  2. I recognize the first catalog entry; it’s the same catalog that we have and I hate it. So does my counterpart at our sister library (we share our catalog). It was already in place when we started working in our libraries, and we haven’t really thought about changing it, mostly because no one uses the catalog. Even when we search, we don’t search by subject heading. One, we don’t have enough books to really need it (we have about 1,000 books combined), and two, the keyword is the default search. I sometimes change it to author or title if I know what it is I’m looking for but can’t remember the call number, but mostly I just use key words and I can always find what I need. That’s the benefit of a small collection I guess.

    Most of my patrons depend on shelf browsing and serendipity to find what they need. They use A-Z to find our online journals and our HTML list to find ebooks.

    Sometimes I wonder why I even have a catalog.

  3. Hi Lesli,
    We looked at our Google Anaylitics and some of the most popular pages/links were to the ebooks by subject and ebooks by title. I can’t remember off the top of my head where exactly they fell but I know it was in the top five of used links. It was surprising to us because as our ebook collection was growing we had hoped to get rid of these pages due to them being too long and difficult to maintain. But according to our stats people preferred using that page for searching our ebooks rather than catalog.

  4. I know of some other librarians who are not happy either. I have begged the company to invest in their front end and they keep saying it is coming, but they have been saying that for 2 years. It is frustrating they design more for the back office end of the product and don’t the front end is just as important if not more so because that is what your users AND ADMINISTRATORS who are paying the bill see. This perfect example of waiting for something to exactly perfect before launch, they could use a little does of perpetual beta.

  5. We use Endeca as a front end to our ILS. It’s a discovery tool that sits on top of the catalogue. It really acts more like amazon, but it still doesn’t get used much. It even has cover flow for books. It’s like virtual shelf browsing. I’m not sure what will work for our clients, nothing seems to get their attention. It is very frustrating.

    Ebooks are still difficult for our users to discover in the catalogue. We have an alpha list which I had to fight our web designers to include, I had to cite you to convince them. But as you noted HTML lists aren’t really scalable. Really perplexing.

    Looking forward to your next installment.

  6. Yesterday I received a nice email from Nadine commenting on cataloging, ebooks, and RDA. I asked her if I could copy and paste it in the comments section of this post and she gave me permission.

    Below is a comment emailed to me from Nadine.

    I am basically a cataloger and we are, in the cataloging community, up in arms regarding RDA, the new cataloging code that has been in development for a good 5 years. I won’t bore you with all the detail and controversy, but I will say that we as librarian catalogers are either way far behind or RDA is way too premature or a bit off the mark. Like, why did our catalogs not start with a gradual transition, that is, to at least have our records wrapped in XML so that they can be more easily and fully exposed on the web ? I digress from your point, I know. However I do want to point out regarding MeSH, that what we need for the future is tons and tons of linked data. There is no reason why we cannot link (granted not perfectly, but perfection is all illusion any way) MeSH to common terms and common terms to MeSH. In fact we have a very nice mapping with the UMLS and it needs to be used and even embellished.

    The semantic web, which I do believe is in the not too distant future will need more standards and so we want to have flexibility as you say with the vocabulary but there have to underlying standards not visible to the commoners naked eye. They don’t need to see it. However, to make this happen we need to create more official registries, build more frameworks (RDF), and get this stuff linked pronto! I am so frustrated you have no idea. We catalogers want to apply our intellectual skills, but we so lack the tools and systems.

  7. We use Vivisimo sitting on the top of Voyager with an OPAC named “PITTtcat”. We wrap the ILS in our homegrown search box:

    http://media.hsls.pitt.edu/hslshomepage.jpg

    We boost electronic online content on the results page. Realistically, this is what the majority of the patrons want to find- :

    http://media.hsls.pitt.edu/searchresults.jpg

    We also have the e-book search and e-book A-Z by subject pulled from Voyager:
    http://www.hsls.pitt.edu/resources/books/ebooks

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>