The world beyond MARC: let’s focus on asking the right questions
6 likes1,193 views
The document discusses the evolution of library metadata standards from MARC to Linked Data frameworks like BIBFRAME. It argues that the focus should not just be on the end goal of fully implementing BIBFRAME, but on the important work of preparing existing legacy data to integrate semantic concepts and building the underlying linked data infrastructure that is necessary to realize the benefits of models like BIBFRAME. Specifically, it encourages libraries to begin enriching MARC records with semantic data and to work with national libraries and partners to develop the networks and relationships of trust between data that will support future metadata environments.
![The world beyond MARC: let’s
focus on asking the right
questions
Terry Reese, Head of Digital Initiatives
If you work with metadata in libraries – you
know that our world is changing. It seems like
most discussions inevitably turn to questions
about BIBFRAME and Linked Data. Our
metadata is evolving, our systems are evolving,
and how these changes will be impacted by the
work being done in the BIBFRAME community is
still an unknown.
Equally unknown is how the libraries’ need for
better discovery while still maintaining the
ability to do metadata work at scale impact
developments within BIBFRAME community.
These are interesting questions – and questions
that today, I think that everyone in attendance
would like to talk about. If library metadata
needs to evolve (and I’m not sure anyone would
disagree that it does [and always has been
evolving], will that evolution lead us to
BIBFRAME? I’m honestly not sure. I think that
somewhere, in the future, there likely will be
something called BIBFRAME that libraries will
utilize. Will it replace MARC, will it become the
new lingua franca for library metadata – who
knows. Go back to the library literature, and we
were having these same conversations around
MARC21XML and MODS. I think that in this
respect, the future is foggy. How we eventually
express metadata and the relationships within
our metadata is still likely to continue to change
as those working on BIBFRAME broaden their
lens from just MARC, to the various other
communities that leverage MARC cataloging
rules but not necessary the MARC framework
(I’m thinking of TEI, EAD, MODS) – these are
communities that will need to be brought along,
engaged with, and ultimately, will likely
continue to shape the discussion around
BIBFRAME. And while that is an interesting
topic of discussion, it’s not the path that I want
to go down. Our wonderful hosts have brought
together some great speakers that will talk at
length about how BIBFRAME is changing
metadata – and how we might implement it
within our local systems. I’m going to leave that
discussion for them. For my part, I’m
particularly interested in how we move this
conversation forward, and develop a plan for
ensuing that data we create today is able to
take advantage of these frameworks in the
future.
The Long Race
Whenever I hear conversations about
BIBFRAME, I often times think that our focus is
in the wrong part of the discussion. Make no
mistake, someone needs to be having these
discussions, and I’m glad that they are going on,
but BIBFRAME is like a race – and this race isn’t
a sprint, but a very long marathon where a fully
realized BIBFRAME represents the finish line.
I’m not sure how many of you are runners, or
how many of you have run marathons, but in
my own experience, it’s the work that happens
between the start and the finish that represents
the real work – and I think that is where we are
at right now. As the BIBFRAME community
begins to flesh out the model, and developers
and communities begin to think about how this
will impact their data – the work that needs to
happen in the middle, the development of the
library linked data networks, the entification of
our existing and legacy data – needs to be
happening right now. Without these networks,
without the entification of our data, the ability
to model data into BIBFRAME largely loses its
meaning.
The problem of a metadata schema or model
that moves beyond MARC isn’t found in RDF
triples, it’s found in the relationships that we
are able to build and express within our data.
It’s found in the relationships of trust that are
formed between our metadata and the larger
system of services that enhance and enrich our
metadata. While libraries should be thinking](https://image.slidesharecdn.com/speakershandoutreese-160314164000/85/The-world-beyond-MARC-let-s-focus-on-asking-the-right-questions-1-320.jpg)
The world beyond MARC: let’s focus on asking the right questions
- 1. The world beyond MARC: let’s focus on asking the right questions Terry Reese, Head of Digital Initiatives If you work with metadata in libraries – you know that our world is changing. It seems like most discussions inevitably turn to questions about BIBFRAME and Linked Data. Our metadata is evolving, our systems are evolving, and how these changes will be impacted by the work being done in the BIBFRAME community is still an unknown. Equally unknown is how the libraries’ need for better discovery while still maintaining the ability to do metadata work at scale impact developments within BIBFRAME community. These are interesting questions – and questions that today, I think that everyone in attendance would like to talk about. If library metadata needs to evolve (and I’m not sure anyone would disagree that it does [and always has been evolving], will that evolution lead us to BIBFRAME? I’m honestly not sure. I think that somewhere, in the future, there likely will be something called BIBFRAME that libraries will utilize. Will it replace MARC, will it become the new lingua franca for library metadata – who knows. Go back to the library literature, and we were having these same conversations around MARC21XML and MODS. I think that in this respect, the future is foggy. How we eventually express metadata and the relationships within our metadata is still likely to continue to change as those working on BIBFRAME broaden their lens from just MARC, to the various other communities that leverage MARC cataloging rules but not necessary the MARC framework (I’m thinking of TEI, EAD, MODS) – these are communities that will need to be brought along, engaged with, and ultimately, will likely continue to shape the discussion around BIBFRAME. And while that is an interesting topic of discussion, it’s not the path that I want to go down. Our wonderful hosts have brought together some great speakers that will talk at length about how BIBFRAME is changing metadata – and how we might implement it within our local systems. I’m going to leave that discussion for them. For my part, I’m particularly interested in how we move this conversation forward, and develop a plan for ensuing that data we create today is able to take advantage of these frameworks in the future. The Long Race Whenever I hear conversations about BIBFRAME, I often times think that our focus is in the wrong part of the discussion. Make no mistake, someone needs to be having these discussions, and I’m glad that they are going on, but BIBFRAME is like a race – and this race isn’t a sprint, but a very long marathon where a fully realized BIBFRAME represents the finish line. I’m not sure how many of you are runners, or how many of you have run marathons, but in my own experience, it’s the work that happens between the start and the finish that represents the real work – and I think that is where we are at right now. As the BIBFRAME community begins to flesh out the model, and developers and communities begin to think about how this will impact their data – the work that needs to happen in the middle, the development of the library linked data networks, the entification of our existing and legacy data – needs to be happening right now. Without these networks, without the entification of our data, the ability to model data into BIBFRAME largely loses its meaning. The problem of a metadata schema or model that moves beyond MARC isn’t found in RDF triples, it’s found in the relationships that we are able to build and express within our data. It’s found in the relationships of trust that are formed between our metadata and the larger system of services that enhance and enrich our metadata. While libraries should be thinking
- 2. about BIBFRAME and actively experimenting with the current evolving models and asking, “What does this mean for my data”, they should also be working to prepare the data that they have to be more linked data friendly, and begin pushing for and supporting organizations that are working to develop the underlying infrastructure that must be in place before the library can realize a metadata world built around semantic data. How does this impact me today? So, what can we be doing today? I think that is the most important question that we can be asking ourselves as we plan for the future. It’s a great question. Libraries pride themselves on their metadata – and if you look at cooperatives like OCLC; we have accumulated over a ½ a century of bibliographic metadata. This is an impressive amount of accumulated knowledge. However, as we think about how we reshape our metadata universe, this strength has also become one of the library community’s greatest obstacles. Our workflows, our tools, our education is tied to a very MARC-centric way of thinking – and if we are to make a break from this long history, we must be able to find a way to transition our present and legacy data and prepare it for a new bibliographic future. Fortunately, there is working happening now to make this a reality. The Program for Cooperative Cataloging, working with the British Library, the German National Library, and many representatives from library system vendors and academic institutions are starting to ask the questions about how we can begin the process of transitioning MARC from a metadata framework of strings, to one of strings+. The goal of this work is to encourage libraries to begin creating metadata that integrates semantic concepts into MARC records, and laying the groundwork for the transition to a more semantically capable metadata expression like BIBFRAME. And this is work that we can be doing now – we can be working at our institutions and with our national libraries and metadata partners to ensure that we are not only focusing on the future and BIBFRAME, but that we are working to build the necessary infrastructure to support the very linking and relationships that BIBFRAME will need to exploit to give it real meaning. It’s time to build roads So that is what I’m asking you to do today – to build the roads of tomorrow. And I think being in Rome, that is an apt metaphor. The Romans were famous for their highways. They developed roads that connected an empire, encouraged travel and trade. Today, as libraries and metadata specialists look to BIBFRAME and the future that it represents, we need build those roads – the linked data infrastructure, that will allow libraries to create references to semantic objects. We also need to do the hard work of entification – of turning our metadata from strings to strings+. And we need to experiment and engage in the conversation. Let us run this race together…Thank you. Presentation handout: Beyond MARC, presented March 15, 2016.