Diane Hillmann is Director of Metadata Initiatives and the Information Institute of Syracuse (formerly of Cornell)

There are several converging trends:

• More catalogers work at a support staff level than as professional librarians
• More cataloging records are selected by machines
• More catalog records are being captured from publisher data or other sources
• More updating of catalog records is done via batch processes
• Libraries continue to de-emphasize processing of secondary research products (books and serials) in favour of unique, primary materials

Options:

• Extinction
• Retool

Extinction:

• Keep cranking about how nobody appreciates us
• Asert over and over that we’re already doing everything right – why should we change?
• Adopt a ‘chicke little’ approach to envisioning the future “the sky is falling”

Retool

• Consider what cataloger do, and what they will do, and map training
• Look for support for retraining at many levels
• Find a new job title – catalogers do a lot of other things

What do ‘metadata librarians’ do (as opposed catalogers – the retooled cataloger):

• Think about descriptive data without pre-conceptions around descriptive level, granularity or descriptive vocabs
• Consider the entirety of the discovery and access issues around a set or collection of materials
• Consider users and uses beyond an individual service when making data design decisions

The metadata librarian is

• aware of changing user needs
• understands the evolving information environment
• works collaboratively with technical staff
• familiar with all metadata formats and encoding standards

The metadata librarian skill set is:

• Views data as collections, sets or streams
• Familiar with a variety of metadata formats (DC, VRA Core, MODS etc.)
• Understands basics of data encoding (XML, RDF etc.) but is generally not a provrammer
• Understands the various ways that data can be created (by humans or machines) and manipulated (crosswalked etc.)

Characterisitics of the New World:

• No more Integrated Library Systems
• Bibliographic utilities are unlikely to be the ‘central node’ for all data
• Creation of metadata will become far more decentralized – not all library data
• Nobody knows how this will all shake out
• But: Metadata Librarians will be critical in forging solutions

Disintegrated Library Systems:

• All metadata will not be managed in and delivered from one central store
• Discovery is the first function that is being disaggregated from the ILS – there will be others
• Metadata may be managed in a variety of databases, structures and systems
• …

Role of bibliographic utilities:

• Optimized to be the middleman of the traditional data sharing system
• Currently limited to handling MARC data – not sure whether or when that will change (RDA will be firths challenge here)
• New services are contemplated

(as an aside OCLC getting a hard time here today – feel a bit sorry for Roy!)

New models of creation and distribution

• All data will not be created by librarians
• some will originate from machine processes
• We need to exchange data based on a more open model – on the web
• Broader use of OAI-PMH is a good start towards opening data beyond applications and bespoke portals
• Need to avoid commoditizing DATA instead base business model on building necessary SERVICES

Not sure about OAI-PMH – why not just published the stuff on a webpage with semantic markup to give structure?

Open data:

• Nobody knows how rich our data is unless we make it fully available – we can’t compete as data providers unless we do this