For many library users, the OPAC is the source of connection to their library – how they find books, movies, CD's, etc. And it's a well-known issue in libraryland that OPAC's frequently aren't all they should be, something that's increasingly in the spotlight because of how much better Web 2.0 sites like Amazon and LibraryThing seem to be at finding/browsing/cataloguing information – even though librarians are the “information professionals.”
Here's an example I came across tonight:
Search Results at RPL
“Who Killed the Electric Car?” = 0 hits
(not sure if they use cookies to track sessions so that link might be expired by the time you try it)
“Who Killed the Electric Car” = 2 hits
(one result for two copies at Central, another much more detailed entry for a copy at Sherwood Village – which is another issue I won't get into)
But what's the title for both entries in the RPL catalogue when you search for the title without a question mark? “Who Killed the Electric Car?” (with the question mark reflecting the full, actual title of the work.)
So if you're a patron searching for the film with the proper, full name and get 0 hits, you're very likely to not refine your search. In fact, the only reason I did was because I saw this film listed in RPL's “DVD's On Order” and knew other films on that list were in the catalogue already so this one should be too.
Somebody with expertise in cataloguing can maybe explain this – I suspect it's something to do with “?” being some sort of a reserved character in the Dynix software, similar to how Google apparently prefers dashes to underscores in its index, a seemingly simple distinction that can make a great difference in your PageRank.
Anyhow, this isn't a criticism of the RPL since it's symptomatic of problems in many library OPAC's across the country. In fact, one thing I have loved about RPL since getting back (and of course, a blog-entry list is pending on that very topic) is their extensive DVD collection on a wide-range of topics, non-fiction and fiction, and not just the National Film Board documentaries from the 1970's and German art films from the 1980's that are the bulk of so many other library's video collections.
One of the recurring battles in libraries is “give people what they do want” versus “give people what they should want”. RPL does a good job of doing both but definitely goes further towards “what people do want” in a way that many libraries don't which is great. For example, one new feature since I returned in that they now have extensive collections of TV show DVD's – you can get complete seasons of everything from Seinfeld to Alias to Corner Gas.
They even go further than that to reach people. Looking at their list of movies on order, one that caught my eye was “The Butterfly Effect 2” That's the direct-to-video sequel to an Ashton Kutcher film that didn't do great business or get great reviews the first time around (though the original is apparently a bit of a cult classic – currently a 7.7 rating on IMDB.)
And just to bring it full circle, who was one of the stars of the Butterfly Effect 2? That's right, Indian Head's biggest fan Dustin Milligan! (Light bulb! 💡 Maybe RPL's media specialist librarian didn't order this film after all. I'm guessing a patron, maybe 13, maybe female, really wanted to see this film and special ordered it. But what is cool is that RPL did order the film – some libraries wouldn't. And even though I just posted it yesterday, why not take the opportunity to use this screen grab of young Mr. Milligan sporting a very cool t-shirt? 😎)
Comments 14
There was a presentation at OLA (a couple actually) by Bibliocommons about the next generation OPAC. They are integrating what Amazon and LibraryThing do so well into the traditional OPAC system. It was rather interesting and is worth checking out.
The thing that gets me about this is that there is a really a pretty simple way that EVERYBODY can be happy: use traditional “exact” search first (so the librarians are happy) and if that doesn't reveal a lot of hits, or no hits, then offer up a “did you mean” section with results based on fuzzier logic (and go one step further and reparse the search string suggesting the “proper” way to search in the future). Libraries cannot use the Google way because web search engines are just that–they fundamentally rely on linking (and anchor text). Libraries will have to keep the traditional exact search methods but could use a little fuzzy logic to open things up. Honestly, its a wonder anyone finds anything in libraries.
And in addition to Bibliocommons, there were also presentations on the OPAC (and ILS) by several others, Like John Blyberg, Karen Calhoun and Ross Singer. You may have missed it, but I know their presentations will be linked on the conference site, so keep your eyes open!
Thanks for the tips – I'll keep an eye out for the presentations on the OLA web site.
Yeah, that's a good point. It's not the “Google way” as you say but if I didn't get any results on an OPAC search for “Who Killed the Electric Car?” and then a box popped up saying “We have 2 results for 'Who Killed The Electric Car' – is that what you meant?”, that would still evoke Google for people (which probably isn't a bad thing.)
What are the two things we heard over and over in library school? “Maximize access to information” and “It's all about the user.” This small change that you suggest would improve things vastly in both areas.
Quinn's solution is probably better than my preference.
This isn't as complicated as it sounds. If I were sketching wireframes long-hand, it would be a lot more clear. In the introductory search screen I would discretely but prominently add a two radio-buttons to define the default index. It will always default to “Title” but author is available as well. You can still have separate fields for Title Keyword or Author Keyword searches, this only applies to the “I'll take anything I can get” keyword search field.
That doesn't mean that you are only searching the title index, though. Your broad keyword search will act as normal. If there are 0 hits for your keyword search, however, it will jump to a browse view of the title index showing where the title would be in the list. If you made a typo in the 5th character or later, you will probably see what you were looking for anyway and are one click away from the record.
At the top of the screen in big prominent letters would be a link asking: “Do you want to browse the author lists instead, Dear?” On the downside you have to click one more time. On the upside you have a friendly Maritimer advising you about the best way to proceed if you don't see what you are looking for.
Actually a combination of the two would probably be ideal. Have the irate Ontarian barking “Did you actually mean this instead?” as well as the helpful Maritimer and you'll cover quite a few bases.
Also make sure that the de rigeur keyword search field is also up there at the top, or in the sidebar, so that typos can be quickly corrected. That probably doesn't need to be said, but sometimes designers drop old features in favor of new and this one shouldn't be dropped.
Cheers
Huh. I never quite realized it, but I *always* leave the question mark out when doing searches, I think because I do know it's an iffy character, and besides that, it's not like putting it in will help the search any. Not that I can extrapolate from what I do to what everyone else does. 🙂
I tried keyword searching for titles containing question marks at a few different OPACs, and it doesn't work for any of them. That's…not good.
(If you just do a title search on “who”, you get a listing of a bunch of titles that should end in question marks – looks like cataloguers usually leave the question mark out, though. Wonder if I was supposed to do that for all the titles I've entered with question marks? Oops. 🙂 Not that I'm going to change my ways. It looks *nicer* with the question mark in…)
According to the rules you leave in punctuation like question marks, exclaimation points etc. in the 245 field. Punctuation that will be redundant or contradictory at the end of an area (e.g. a period followed by a colon preceding the $b field, or followed by another period for the end of area punctuation. The usual exception applies, where if the offending punctuation in the title is integral to the meaning of the title, then you leave it in anyway.
If you check the LC title authority headings all of the titles that are questions beginning with “Who…” have the question mark in the authorized heading.
However, I have catalogued on systems (a long time ago, granted) where the “house rule” was to leave out any punctuation at the end of the 245 field because the system always supplied the “. –” So if there were no $b or $c (other title information or statement of responsibility) then question marks, exclamation points, etc. would be dropped.
This might be a legacy house-rule from earlier systems — or perhaps the current systems aren't up to speed yet.
That's another thing the RPL catalogue does that really bugs me – they default to Keyword search instead of Title search. I'm sure they have their reasons but to me, Title search is the most intuitive, especially in a public library (maybe not so much in other types of libraries.)
Your other point is a good one as well – I believe LPL does that “jump the nearest section based on nearby characters” so that your search term will get you in the general area if you have the first word or few characters correct. Not sure if this is a Dynix issue or not but I don't think I've seen any Dynix OPAC do this.
I'm sure there are lots of studies on “best practices” for OPAC's just as I'm sure that none of them agree on what those may be!
I don't always leave them out but because of my “techie” background, I at least have an understanding why asterixes and question marks and tildes and dollar signs other characters often have special meanings that can screw things up when you think you're using them normally.
Kathleen Crowley had a funny blog entry recently saying her first book would be titled “$b” to “piss off cataloguers and bungle up MARC records.” I think that's a hilarious idea. Maybe Corey Redekop's sequel to his ShelfMonkey book could steal this title?
http://linguistigeek.blogspot.com/2007_01_01_archive.html#2262436568125785436
Mike, all I have to say to your comment is “It boggles my mind that I understand (most) of what you're talking about.” Hooray for library insider jargon and knowledge! Thanks for the thoughts…
Yeah. I'm actually teaching my kids to do the ISBD chant “Title. Space-colon-space. Othertitleinformation. Space-slash-space. StatementofResponsibility. Period-space-dash-space.”
Well. Not really. But it would be cute. To some.
That's how we learned it in Library Tech school. Cataloguing at MLIS was good, and the Advanced Cataloguing class was excellent for moving towards an understanding of when and how the rules should be broken. But nothing will ingrain the rules better than chanting the forms in unison like a 6 year old.
I'm sure that Trixie and Renee will agree with me.
Cheers
I do that in my head sometimes when typing an address or whatever – space – dash – space.
It does have a nice rhythm – I bet kids would enjoy it. (Hmm, why are there no library-themed rock songs?)
Elisabeth Davies didn't exactly make us chant but you did get the into the rhythm of the rules by how she said it.
Library tech *did* make you chant? I'm jealous. Someone mad the analogy that the masters program is 70% theory and 30% practical whereas tech school is the opposite – does that ring true for you?
I am a maroon! Even after all the discussion around this topic in this thread, I went to order “What's Going On In There?” (full title: “What's going on in there? :”how the brain and mind develop in the first five years of life”) via ILL from Moose Jaw since it didn't show up in RPL's catalogue. Only the thought that Regina *must* have it if Moose Jaw does made me double-check my search and realise what was wrong. Grrr!
(Lauralee, it looks like the MARC record – at least in Moose Jaw – includes the question mark.)