Library IM Bots
Technobiblio posted about the possibility of creating IM bots for libraries, that could (using an automated service) have canned responses for common library IM inquires, like "what are your hours" or "what events are happening at the library." Rochelle over at the Diary of a Subversive Librarian took this idea to task. She’s a self-proclaimed "keyword bot specialist," and explains why this approach won’t work well for library IM services.
I’d say the key here isn’t to develop a bot to answer those questions. Take that knowledge and display that information more prominently on the library’s website instead. Make the information easier to find, or include it on the same page as the IM service in addition to its regular place. Add a FAQ if you can’t work out better ways to answer obvious questions on the front page. I think limiting v-ref by creating databases of generic answers is a really good way to kill the service rather than promote it.
Right on Rochelle. Of the 100 or so IM Ref questions I’ve answered so far, I would say that only one (an inquiry about the phone # of one of our branches) would have been feasibly bot-answerable. Bot-ing our IM services would surely, as Rochelle says, kill the service. Nobody, and I mean nobody, likes to talk to a machine.
April 6th, 2005 at 2:31 pm
I would point you to my comments on the post over at Rochelle’s DoaSL as an attempt at better explaining what I was proposing. It was not in any way to replace v-ref IM or reference librarians. Sorry that I was not more clear about my meaning. Hopefully the comments will explain what I was aiming at – chris
April 6th, 2005 at 4:44 pm
Poor Chris. He threw an idea out there and I ran with it in the direction I thought he going.
But the idea of replacing live library staff with a bot isn’t that far-fetched; I was dabbling in v-ref last summer, and the vendor who supplied the service at that library was storing all the IM conversations we had. As their literature explained, they were keeping the transcripts with an eye to creating a database of answers that could be, presumably, called upon as a resource for all v-ref staff continent-wide.
I’m extremely wary of vendors having a copy of reference questions and answers, myself. And I’m even more wary of some unsuspecting soul trying to sort through my friendly banter trying to find the perfect answer for whatever question the patron had.
Chris was thinking of inserting a jabber server as another service in addition to v-ref, and that’s interesting. I think we don’t really have enough information like that to justify an entire database, but there’s something to be said for creating cellphone accessible content, since that’s one direction technology is heading. I suspect the answer for that is probably in a webpage rather than a database. That might work something like an RSS feed that’s updated daily with hours and events of the day and the next day. You can subscribe to feeds with a sidekick or a blackberry, as far as I know.
Ah well, lots to chew over in the next few years.
April 7th, 2005 at 6:58 am
What I’d like to see would be common questions answerable by text messenging. Send “Library hours” and get “8am-11pm M-Th…” etc. Send a book title and get a call number (or a “Sorry we don’t have that. Would you like to 1) request it by ILL 2) recommend it for purchase by the library?” or maybe even “Did you mean…”)
That’s the sort of question that doesn’t really need a human, especially if the reply came back with the IM and phone number for the reference desk. If I can txt Google and get the Amazon price of a book, why can’t I txt a library and find out if I don’t have to buy the book?
April 7th, 2005 at 8:46 am
Oh dear. Yes, poor Chris!
I echo Rochelle’s concern about being wary of third party vendors having a copy of all reference interactions. I’m almost positive that would violate our library’s patron privacy policy…
And Rebecca’s question is a good one: “If I can txt Google and get the Amazon price of a book, why can’t I txt a library and find out if I don’t have to buy the book?” I don’t know that I have the answer though. We, as librarians, would need to pressure the ILS vendors to develop something that would allow an IM client to interact with the records database…or some hacker-inclined librarians could whip something up. Unfortunately, that’s beyond my skill set.