V-Ref and the Spectre of Transcripts

V-Ref and the Spectre of Transcripts

More ideas-swapping, this time from lbr:

I am a firm believer that text transcripts are one of the most revolutionary aspects of virtual reference. They open a whole new world of possibilities for resources that would be difficult if not impossible in traditional reference interactions:
detailed self-review and peer-review for quality control and improvement
ongoing modeling of good (and not-so-good) reference techniques for new and future librarians
– construction of knowledgebases of past interactions so that librarians can benefit from their colleagues’ knowledge and discoveries when answering the same questions later
– large-scale aggregated analysis of patron questions and needs to inform administrative decisions about staffing, training, collection development and resource acquisition.

I could not disagree more. I say this with a background in educational technology, virtual environments, and as a person active in a variety of online communities. Admittedly I have the default reaction of cringing every time I hear about someone collecting transcripts from IM conversations, but even aside from that, I don’t think transcripts are what make v-ref a good thing.

Let’s say we agree that recording the interaction between patrons and library staff is something we’re interested in doing. For all the reasons listed above, why wouldn’t we just install cameras and microphones at the reference desk? Why not at least record the audio of all reference interviews, and since voice-to-text transcription is reasonably accurate, why don’t we just keep a text transcript of all interactions that way?

Well, that’s easy to answer. What patron would possibly want to talk to you if they knew everything they said to you was being logged, for whatever reason?

Yeah. You see where I’m going. The fact that transcripts are kept and held by off-site vendors without explicit warning creeps me out as it is. On one hand people talk about wanting to implement v-ref in, say, public libraries in order to reach a different audience (read: teens and tech-savvy young adults), and on the other they’re talking about logging transcripts and storing them in databases. What teenager would ever use a service that had a big pop up window that said “Hey, we’re recording everything you say here, btw, and we share it with our friends”, an extra and unpopular step that ethically MUST be present if you want to collect transcripts? Honestly, I’m not sure I even would use such a service. I’d be tempted to claim copyright infringement and sue the pants of the library, but I’m not that sort of person. Usually.

I will attempt to keep my personal distaste of transcripts to one side as I break down some of the “possibilities” of transcript collection as described by lbr:

detailed self-review and peer-review for quality control and improvement.

Transcripts are hardly the greatest way to do self-review or peer-review. Pulling out the transcripts and going over them presumes that the staff member is somehow unaware of the kinds of conversations they have in v-ref. Ideally, reference managers have open lines of communication with their staff wherein there is a constant conversation about what’s going on with v-ref. I’d be happier if the request were instead to have staff post to a v-ref blog at the end of every shift, detailing the sorts of questions they had that day and the problems they encountered, sharing ideas for good answers to difficult questions and fielding comments from other staff members.

In you’re really interested in using v-ref transcripts in performance reviews, you can just have staff print off a few transcripts for personal staffing records rather than compiling a universal database. To me this feels like trying to grab a bull by the horns and reaching for his tail.

As for improvement and “quality control”; the best ways to move in both of those directions is to do regular and thoughtful staff training. In my experience (as a v-ref user and on the other side of the fence), most library staff come to v-ref as their first experience ever with IM; they are clumsy, awkward, and often come across as brusque and/or too busy to really investigate the patron’s query. They end conversations too early, are in my opinion overly concerned about “appearing professional”, scared of l33t (net speak), intimidated by the speed at which v-ref moves, and generally uncomfortable with the medium. If you’re interested in improving the quality of v-ref service, install an internal IM system and let staff get used to communicating using it. Bring in someone who knows how to IM and knows how to talk about using IM. Reading transcripts is not going to change anyone opinions or make frustrated, frightened staff better v-ref service providers. It may in fact only make them perform less well, given the pressure of being permanently recorded.

construction of knowledgebases of past interactions so that librarians can benefit from their colleagues’ knowledge and discoveries when answering the same questions later

I’m profoundly uncomfortable with the idea of a database of “good” reference answers. If someone comes to me at the reference desk looking for some really good articles on the American Civil War, I am not going to go first to a database of reference answers handed out over the last few years by other reference librarians. Any such database (after all the scrubbing of personal information and uploading from one institution to the main database) would be out of date. Sure, it might help me find something we may or may not have access to at my library, but it would be missing articles that were the most recent, and would direct me to listen more to the “popular” answer rather than the patron’s needs. I have been trained on the resources at my own library. Most library staff are extremely knowledgeable about the resources available, and if you fear your staff are not knowledgeable enough, this is another moment to look at staff development, not collecting transcripts. We have spent a lot of money making databases accessible and relatively easy to use. Why this sudden distrust of library staff? Two heads are better than one, I’ll agree, but I’m not sure 1500 are better than two. Instead of implementing a database, get your reference staff to rely on each other for help. Have a gov docs question and you feel out of your depth? Isn’t that why we hired a gov docs librarian?

And more to the point: your reference interactions should not be the means through which you staff communicate with each other. In my experience the reference interactions that really require mediation are specific class assignments; these are best dealt with through a staff meeting, wherein the subject librarian who is aware of the assignment details the question and gives some good directions for answers. Support in that case should really not be mediated by technology. Unless that technology is a blog where a subject librarian posts the details of assignments and the best databases or books where sources can be found. More direct communication between librarians, even from campus to campus or across the country, is a better method of knowledge-sharing than breaching patron confidentiality by storing v-ref transcripts.

large-scale aggregated analysis of patron questions and needs to inform administrative decisions about staffing, training, collection development and resource acquisition.

There are only two statistics administrators need to know for the purposes of understanding the importance of v-ref. How many patrons logged in to the service, and how much time was logged by staff. Libraries are staffed by knowledgeable, thoughtful people. We don’t need a database to tell us if patrons are pointing out that our collections are suffering. Having open lines of communication between front-line staff, acquisitions librarians, and administrators will render such a database obsolete. If staff are routinely encountering questions they feel ill-suited to answer, they need a venue to voice that concern. Is there a library staff member who would not report the lack of an important source once it was brought to their attention?

Personally, I would prefer to see no v-ref transcripts held by any library anywhere. Offer the transcript to the patron if they want it, and to the individual librarian if they want to keep it, but the institution should not be holding transcripts of reference interviews. We still have issues around holding borrowing records, why on earth are we so flippant about storing actual conversations between library staff and patrons?

I am all for using technology to make library services better, but just because you can record something doesn’t mean you should. Refocus on the problems you think transcripts will solve and work out better ways to do so. Nine times out of ten, the problems can be solved by improving internal communication, staff development, providing outlets for concerns, and encouraging staff to talk openly and honestly about any problem they are encountering in the workplace.

Recording everything that happens between reference staff and patrons is not only a turn-off for patrons. It sends a rather unsavory message to library staff as well. In recording everything library staff say, are you making their workplace uncomfortable? Sure, we expect all interactions to be upstanding and ideal, but what does recording actually accomplish? We expect teachers to do their jobs well, but we don’t record everything that goes on in the classroom. That would be a breach of privacy and a mark of pretty profound distrust between teachers and administrators. In recording everything reference staff say, are you preventing them from uttering those painful words, “I don’t know, but let me find out who does”? They are typing words directly into their performance review, after all.

I don’t like transcripts. I’ll be honest. I’ve experienced that sick feeling when you realize that a delicate conversation you had with a friend has been helpfully emailed out to a few others for the sake of keeping everyone in the loop. IM has as relationship with speech, with a face-to-face interaction, and I believe we should grant it the same respect we grant voice-delivered reference questions. When you start up an IM session, you are speaking to one person, not a whole host of librarians across the globe.

The history of IM is a far more personal one than email; email was always more professional in purpose, a tool for exchanging ideas and documents rather than sensitive information. From the beginning IM has been a way to connect with one person in a largely undocumented and undocumentable way. It is more personal and off the record. We should be honest and upfront about what’s going on in that interaction and not bait and switch our patrons, who are not used to the idea that IM leads directly to recording and database population. V-ref is not an opportunity to milk patrons or library staff for information to help us improve our services globally. V-ref is an extension of traditional reference, and is a service libraries should provide to patrons who need alternative means to connect with library staff.

Once we’ve got a more complete and comprehensive archive, we can really start to leverage it as a knowledgebase with the ability to search previous transcripts for keywords and have them help us find resources for the session we’re in right now.

It all boils down to a simple information literacy rule: don’t let other people do your thinking for you. Reach out to your colleagues when in a bind. Don’t let staff do v-ref alone in their offices, or in a generally isolated place. There are so many skills in a library reference office; encourage staff to understand the skill set of their colleagues, and call on them when needed.

Let’s not outsource the expertise that’s already there.

Comments are closed.