On OPACs and SOPACs
February 2nd, 2007 by Christopher HarrisLike many, I am quite excited about John Blyberg’s release of the SOPAC (and other enhancements). The social tools – tagging, rating, recommending, etc. – are an incredibly powerful additions to the library finding experience. This is in no way an attack on what has been accomplished; believe me, John Blyberg is a hero around our offices! We are quite excited about looking over his code (thank you for releasing it!) to compare his methods with what we have been working on for tagging, rating, etc. in our program. I worry, however, that this is the proverbial addition of beauty products to a porcine companion.
The AADL SOPAC is, at its heart, an OPAC. And therein lies the problem. The purpose of an OPAC is to connect users with MARC records. Even if we connect them in a socially empowered environment, we are still connecting users with something that has no value to them. They are not looking for MARC records, they are trying to find resources. The back end OPAC that is powering the search for the SOPAC does not provide a positive finding experience.
Wanting to check out the SOPAC improvements in action, I headed over to the AADL Catalog to search for one of my favorite books, Elizabeth Moon’s The Deed of Paksenarrion (the first part is available as a free e-book from Baen). Being a good little OPAC user, I dropped down in the search field to pick “Title” and then entered my search term, “paksenarrion.” I know AADL has the book, because I have annotated the catalog card for it. So, imagine my surprise when I was told that the book was not found!
The OPAC did suggest that I might want to “Search as Words” to try and find the book, and so I clicked on the offered link. Instantly, the record for the book pops up! So what happened here. The “Title” search seems to be doing some sort of “begins with” browse, and not a search of words in the title fields of the records. But then it gets confusing. On the main catalog screen, it defaults to “Keyword” search which works like a charm to find the book. But if you are inside the catalog, there is no “Keyword” option. Instead you have to use the “Other” search type to get keyword searching. Then the OPAC introduced a third name for this search type with the link to “Search as Words.” In other words, too much searching and not enough finding. If Keyword (the default) is the best option then why not just limit the main screen search to a single field that only performs keyword searches. Users expect that from their other search engine experiences. They know to look for an “advanced search” link if they want more options. Offering the options up front ended up causing more problems for me when I approached this interface as an advanced searcher.
Then we get to the results. I love the inclusion of tags (none for Paksenarrion yet) and the recommendations, but am less satisfied with the rest of the display. Like many others, these results reinforce the impression that the purpose of the OPAC is to connect users with MARC records. The publication info that could easily be parsed into a more display friendly format remains in all its AACR2 glory. A link to display the full MARC record is also displayed prominently in the top navigation bar.
Again, let me stress that this is not an attack on AADL’s catalog. Their OPAC is much better than many other OPACs we could analyze; and yet it is still an OPAC. My contribution to the group think on “OPACs Suck” is that maybe we need to totally move away from OPACs. That is what we are working towards with FISH. As an Integrated Search Handler, our ISH is designed from the ground up to be a middle layer between users and the MARC records so ne’er the twain shall meet. Our hope is that this will lead to a more positive finding experience.
February 2nd, 2007 at 6:36 pm
Chris, I’m in agreement with you on most points and am eagerly and gratefully following both SOPAC and FISH. But I have to quibble with something you say above. Maybe I’m dense — this is not really my field — but it seems to me the purpose of an OPAC is not to put users in touch with MARC records, it’s to put people in touch with useful information. An OPAC is both a tool (Ranganathan’s #4: Save the time of the reader) and an environment. MARC records are a subsidiary tool, an abstraction used to represent information in manageable chunks, a hedge against information’s natural tendency. As tools OPACs should be made more powerful and as environments they should be made more congenial, but I guess I’m a little befuddled by the hostility toward OPACs as such. Would you explain a bit more?
February 2nd, 2007 at 8:01 pm
Perhaps I should have been more clear. In my opinion, MARC records are NOT “useful information.” Or at least not for our users. Remember, this is a specially designed format that was made to be machine readable. To really fulfill the spirit of Law #4, shouldn’t we interpret the MARC record and present it in a human readable format?
Yet most OPACs seem locked in to presenting information directly from the MARC record. Most of them include links to view the MARC record. The point, from this users view, seems to be to connect me with the MARC record. But I didn’t come looking for a MARC record, I wanted to find a book. So why shouldn’t the finding tool focus on connecting me directly to the book and remove (or at least hide from user view) the intermediary MARC record?
The point I am working towards is that the reason why OPACs suck is very similar to the reason why typewriters suck. We have new technologies that make typewriters obsolete. Sure, there are still situations when specialized users may need a typewriter, but the general public is much better served with a word processor on a computer. We can waste time talking about how the typewriter sucks, or we can stipulate that its time has come and work towards making the best darn word processor possible (including by freeing ourselves from the restrictions of “what has always been done on a typewritter”).
Why, for example, do so many OPACs present information straight from the MARC record using AACR2 formatting? Why do ISBNs appear as ” 0671721046 (pbk.) : $12.00″ and notes as “‘A Baen Books original’–T.p. verso.” Neither of those is in user friendly language. Even worse is the standard OPAC description. Users get nothing from “vii, 1024 p. : map ; 21 cm.”
Seriously…how many users are using size as a primary considersation for book selection?
February 2nd, 2007 at 10:31 pm
Thanks for the inspiration Chris :) http://www.librarybytes.com/2007/02/band-aids-thoughts-inspired-by-chris.html