April 15th and 16th, I attended NISO’s BIBFRAME Roadmap Meeting in Baltimore. The meeting was intended to bring together different people working in the information space, not just librarians but publishers and organizations involved in scholarly communication, to talk about a path forward as we begin to move beyond MARC. The Library of Congress has been busy working on a new ontology and data model for bibliographic information, BIBFRAME, but unfortunately, no one from LC attended this meeting to join in the conversation, which was a little disappointing.
I will admit, I was also a little disappointed by some of the things my fellow librarians said over those two days. Unfortunately, I had to leave early on Tuesday to catch my flight home, and while I tried to keep up on Twitter, I don’t know what I missed, and things were just starting to get really interesting when I left. Maybe the conversation moved forward after I left, and the statements I found frustrating were clarified afterwards. Maybe with more information, I wouldn’t have found them quite so off-the-mark.
One of the pieces of the conversation that I think was really missing was a frank discussion about the role and purpose of libraries, and the role and purpose of library metadata. If we aren’t on the same page about these ideas, then we won’t be working together as well as we could be, and we might even be working at cross purposes. I don’t think I’m alone in this. The head of NISO, Todd Carpenter, said in his opening talk that “We as a community don’t know where we’re going,” and I agree. Or at least I think that some of us know where we want to go, but we don’t all want to go to the same place.
There were several major themes that arose throughout the meeting, and lots of smaller and larger group discussions around these themes: Openness of data and standards, the goals and strategies of a new data model, business implementations and aspects of adoption, prototypes and technical issues, and users/user needs. I think it became clear to everyone that these themes are not truly separable; they are all deeply enmeshed and decisions can’t be made in one area without affecting the others.
To my mind, the goals have to come first. The Goals and Strategies subgroup did raise some important ideas, but I think they are all very vague: We should enhance collaborative and shared infrastructures, provide better usability of library data for end users, minimize the cost of change, identify pain points in moving forward, and identify new roles for standards organizations. To me, all of these raise further questions: Why? To what purpose? Better usability of library data for our users to do what? Collaboration to do what?
The moment when I found myself stunned by how much I disagreed with someone was when I brought up the idea of more tangible goals. In my opinion, the purpose of a library is to make resources available to people in our user communities. In order for those resources to be used, people have to know about them, and we have to make the resources easier to find. I believe the primary reason to move to a new data model is to make our library resources more easily discoverable on the open web, through the means people are predominately using to find information: search engines and web apps. I think it’s not enough for WorldCat to put their bibliographic information on the web. Libraries need to tell the web what we own, what we can make available, and our current OPACs fail at this, hard. My first goal is to ensure that when people search the internet, their libraries’ resources come up front and center in results lists. I want to stop making people come to my OPAC to find out what I have.
When I mentioned this goal, someone actually grimaced, and said that she didn’t think we should be doing all this work for Google. I saw in that moment a fundamental difference in understanding in terms of what our purpose is, and what the purpose of our metadata is, and of how the information world works right now, for our users.
Why do we create metadata, if not to allow people to find the things that the metadata describes? Do we create metadata for itself, just so that it exists? Do we create metadata just so that it will be really good? Do we create it so that it can sit in our little, lonely, isolated buckets, on the web but not of the web?
I think it’s obvious what I think, but…maybe I’m wrong? Maybe I’m not seeing the whole picture?
But if my thoughts about library metadata have any validity, and I think they do, we should be working now to get our holdings data out there on the web, indexed by search engines, linked to other web properties, and more easily discoverable by everyone who can access it.
In addition, I think we should make that data available to app developers and to people who use other systems like GoodReads, LibraryThing, Facebook, Mendeley, and whatever brilliant app doesn’t exist yet. We should make sure that if someone wants to create a tool that will allow users to find things at their local library, they can do that.
We do need to make sure that our systems work together, but it’s just as, if not more, important that our systems work with external systems. It’s important, too, that we not box ourselves in to working only with the vendors we’re used it, waiting for them to do the work for us.
There was a lot more that came out of this meeting, and I have notes and notes and notes to go through. I intend to write another post, with some more positive and practical things that came out of this meeting, but this was where my mind kept circling back during the meeting: Why are we creating metadata in the first place?