Robyn Holmes
Kaye McIntyre
Publication date

This paper examines some of the strategies used by the National Library of Australia to build a collaborative music service on a national scale, and in particular the challenges faced in building online audio content.  It demonstrates the opportunities that a government-business partnership provides to break through some of the existing barriers in providing access to resources.  And it reflects upon the volatility in the music industry that creates ongoing challenges for archiving, resource discovery, online access to and delivery of musical sound.

What is Music Australia and how does it work?

Music Australia is a free resource discovery service to Australian music, hosted by the National Library of Australia and representing more than 50 cultural organisations around the nation. The service crosses the boundaries of formats, and institutions, to present a comprehensive showcase of Australian musical culture from the earliest published music to the latest recorded hit. This includes all types, styles and genres of Australian music, information and research about Australian music – whether created, performed or published in Australia or by Australians, or associated with Australia. In this way, Music Australia creates and displays to the user a ‘virtual’ national music collection.

We harvest and upload metadata from multiple institutions into the Australian National Bibligraphic Database (ANBD), from which the Music Australia resources database is derived. Music Australia stores and displays the bibliographic records, but the contributing organisation holds and delivers the resource. If a resource is available online, people can view, listen, download, or print. If an object is not in digital form or online, the user is guided from the bibliographic record to where they can locate or ‘get a copy’, whether by borrowing through an institution or buying from an institutional copying service or commercial supplier.

To be sustainable, Music Australia is automated as much as possible so organisations can contribute to the service largely from within their existing staffing, resources and programs.

Most importantly, the service is not a single format site akin to the growing number of commercial digital download music or sheet music sites. From the beginning we wanted to ensure that music was represented in its inherent complexity as a symbolic and performed artform documented in multiple formats and contextualised as part of an historical continuum in Australian cultural life. We wanted to provide a musicsubject-basedview displaying the range of resources together with information relating to musical works, composers, performers and organisations. All this is delivered seamlessly through a single point of access.A simple search from the home page using the term ‘kangaroo’ will suffice to demonstrate the wide range of formats, functionality, institutions and resources – from heritage to contemporary.The search returns a wonderful mixture of music material associated with the term ‘kangaroo’, including sheet music from Paris in the 1800s, recordings of indigenous dances and children’s songs, through to orchestral suites and football anthems. Formats include pictures, scores, manuscript collections, books, sound recordings, moving image, archived websites, educational kits, as well as biographical information pertaining to people and organisations.The search even reveals an original manuscript drawing of Percy Grainger’s ‘free music’ - a Kangaroo pouch method of playing and synchronising 8 oscillators – held at the Grainger Collection in the University of Melbourne.

Strategies for building sound content in Music Australia

Music Australia has been developed in four stages, the first two stages in management partnership with the National Film and Sound Archive.

  • Pilot demonstrator project: November2001-August2002
  • Pilot to Production development: March2003-February2005
  • Release of Music Australia 1.0: March 2005
  • Release of Music Australia 2.0: April 2007

A key question from day one of the service development has been: how could we generate Australian music content online when there was virtually none at the beginning of the pilot? Clearly if we wanted comprehensive coverage of the musical landscape, we would have to expand across a number of sectors.

The approach we took was systematically to build partnerships with organisations across four arenas:

  1. Libraries, including state libraries and small specialist music libraries and archives (eg Grainger Museum at the University of Melbourne, Australian Music Centre);
  2. National collecting institutions (eg National Library of Australia, National Film and Sound Archive);
  3. Research projects and linkages to build university infrastructure and unique musical content (eg Australian Sound Design, Sound Footings, Australian Jewish Music Archive, National Recording Project for Indigenous Performance in Australia);
  4. Arts projects and arts funded organisations (eg Australian Music Online, funded by the Australia Council for the Arts).

The key principles and strategies we used through these partnerships to build content were to:

  • develop guidelines for digitisation of scores and sound and coordinate content selection cooperatively across different organisations;
  • assist organisations to identify unique Australian resources and catalogue and upload these to the Australian National Bibliographic Database (ANDB);
  • develop a methodology for harvesting, mapping and converting disparate types of data into a central national database of MARC records (the ANBD);
  • prioritise relationships with agencies or projects that could manage relationships with artists and the associated contractual issues and rights obligations;
  • support and influence initiatives or projects from the early stages of development so that appropriate architecture, linkages, protocols and permissions for delivery through Music Australia were established at the outset;
  • alert an ‘unknowing’ community of creators, performers, producers, publishers and audiences to the Library’s role in building this national corpus and service online;
  • identify Australian web sites with rich music content for distributed archiving by different institutions into PANDORA, Australia’s web archive.

Such partnership building, mainly with government funded organisations, has dramatically increased the representation of music in the ANBD and Music Australia since the first stages of its development. Nevertheless, it would be fair to say that, despite these very significant initiatives, the desiredflow-oneffect of generating digitised or capturingborn-digital Australian music content and making this accessible to users online continued to remain small in quantity and limited in scope.

A few issues acted as significant inhibitors to creating and sustaining access to online music content, and especially sound:

  1. The problems associated with archiving and making accessible born digital objects and deep web content. This is made more difficult without national legal deposit legislation in Australia foraudio-visualrecordings and electronic media.
  2. The copyright barriers, which have inhibited online access to the legacy content of major national projects such as the Anthology of Australian Music on Disc which had been an initiative of the Australian National University.
  3. The legacy and sustainability issues that remain withone-offresearch or arts grants, funded competitively by external agencies of government through peer review. Who archives the work, who updates the data and who sustains the access to source material when the funding and project expertise runs out?
  4. The lack of systems in institutions forlarge-scaledelivery of sound recordings and other media. This has meant that the leading work of sound archives in digital preservation and storage has not as yet systematically flowed through to delivery and online access. The great news for the National Library has been the recent launch of the Library’s audio delivery system. The first of the Library’s substantial oral history and folklore archival recordings related to music are now accessible through Music Australia. A delightful example is 90 year old Harry Bormann, born 1901, and interviewed and recorded in 1991 by field collector John Meredith on Harry’s verandah in Claypans South Australia. No matter how hard Meredith tried to capture Harry’s tunes, his dog insisted on singing along to a soundscape of humming bees! The background detail in such field recordings is often as important as the songs and tunes, a window into the world and life of the singer.

You can listen to this item directly in Music Australia or retrieve the item here.

This will take you first to a Licence Agreement covering conditions of access and terms of use and then to an mp3 for listening or downloading.

If there is a timed summary or full transcript of the recording, the sound is matched automatically to the text, thus enabling the user to search for exact moments in which songs or tunes may be embedded and contextualised within lengthy interviews. Subject to permissions, a Music Australia user can also order an archival or broadcasting quality copy via the National Library’s Copies Direct service simply by selecting the “Want to Buy” option from the record.

Music Australia 2.0 and a government-business partnership

By the end of December 2005, the first year of the production service, the MA resources database included more than 153,179 bibliographic records. Of these only 11,211 musical resources were fully available online or had an online component. Most notably, of the almost 37,983 Australian sound recordings catalogued in Australian institutions, only 557 (1.4%) were available as online sound. By comparison almost 10% of Australian music scores were available in digitised or digital form, largely due to the extensive program by the National Library to digitise and deliver Australian sheet music and to cooperativelykick-startprinted music digitisation in the six State libraries.

Moreover, most strikingly, approximately 80% of the online content in Music Australia was out of copyright in the public domain. The service was thus seriously skewed to heritage music and perceived by users to represent primarily an historical not a contemporary corpus, even though the vast majority of the bibliographic records were for musicpost-1930.In the meantime, the quantity of music being offered online in the marketplace either by artists, publishers or distributors was suddenly beginning to grow. When we started building Music Australia, there was almost no Australian music online. By 2006, and before iTunes was available to Australians, a few online retailers had begun to emerge, the mobile phone companies were moving into the music business, individual music web sites were spawning, file sharing was rife, and social networking around music spaces was gearing up.

We therefore started to consider the vexed issue of how we might grapple with music that was disseminated in the democracy of the web, well outside of the institutional spaces and collecting processes we occupied. We began to think more widely about engaging directly the Australian music industry.An opportunity presented itself in 2006 in the form of an Australian music aggregator, destra Music, at that time a small company that had begun its life as a wholly Australian owned and focussed digital music business. It was actively aggregating Australian music from publishers and streaming to radio, and had also initiated <> as afree-to-air‘deposit yourmusic’ site for emerging Australian artists.

What was particular about destra that made it stand out as a potential industry partner?

  • Destra had been an early entrant into the digital music business and management of online rights in Australia.
  • They had created rich data for their digital music, with descriptive fields of data at album and track level that could be exported and mapped to MARC records for inclusion in Music Australia.
  • They were willing to generate for Music Australia an ‘Australian’ subset from their global database, including artist information
  • They were prepared to share the intellectual property in their data via Music Australia so other libraries and archives could access it.
  • They offered support for the delivery of WMA (with Digital Rights Management) with 30 second clips and cover art, and were moving towards the delivery of mp3 formats (withoutDRM) in line with a shift in the music industry. (They also hosted a free download/networking site for unsigned and independent artists whichwe were discussing as a possible future content source).
  • They were able to host aback-ende-commerceservice for Music Australia, which would assist our users to move seamlessly from sampling to downloading full tracks and albums online.

What benefits did destra Music foresee in participating in a national, but non-commercial, service?

  • They saw benefits to their marketing in agovernment-businesspartnership, and the potential for reaching a different type and range of users.
  • They believed it would help to leverage other research and development, including industry innovation grants for their imaginative software development ideas.
  • Music Australia could provide an historical context for their aggregation of Australian contemporary music and, with the future prospect of archiving their work, a sense of permanency about its value.

The National Library decided to take a serious business decision: in late 2006 we negotiated a contract to purchase the Destra data in perpetuity, with daily updates for two years, and to reengineer the service to accommodate a digital download facility. The National Library launched Music Australia version 2.0 in April 2007, two years after the first release.This music industry partnership therefore enabled us to

  • Provide online access to a growing ‘virtual’ contemporary audio collection that was being created outside institutional collections, generated directly from the publishers and artists.▪Develop the means to deliverin-copyrightcontemporary music, with our industry partner managing all the artist and publisher rights and royalties through digital rights management and licensing agreements.
  • Provide a Music Australiae-commerceservice to enhance users ‘getting’ options, with the industry partner managing the commercial download transactions.
  • Aggregate more artist information about contemporary music, often unavailable in traditional sources.
  • Move towards a new collecting and access model, to ensure we could capture both a permanent bibliographic record of Australian ‘hot off the press’ music as well as conceive a future mechanism for direct contributions to the service by artists, creators, publishers and distributors.

Enable the user to navigate between track and album level bibliographic records, as well as people and organisation records, a level of granularity that had not previously been possible.

Immediately before the launch of the new Music Australia version 2.0 there were:

  • 167,656 resources
  • 13,080 online resources
  • 4392 people and organisation records

With the release of MA 2.0, April 2007, with Destra content1, there were:

  • 211,889 resources — an increase of 26.3% (44,233)
  • 55,954 online resources — an increase of 327% (42,874), mostly representing an increase in online sound
  • 4798 people and organisation records — an increase of 9.2% (406 records)

Additionally, the number of requests from artists and other suppliers to participate in Music Australia has encouraged us to consider the deposit of music from independent artists and multiple suppliers.

The most frequently asked questions that now come to Music Australia are “Why am I not represented in Music Australia?” and “How can I contribute?

The surge in usage from late March to May 2007, after the official launch on April 6, is striking.

Recent events: termination of contract by business partner and its implications for Music Australia

This paper, however, has an unfortunate coda. Earlier in 2007, the parent company destra Media, which had rapidly expanded and diversified its business enterprise since the period when we negotiated ourcontract, was taken over by the media entertainment group Prime. A business review led to Prime closing the door on the ‘destra Music’ arm of the business, and to this end destra terminated its contract with the Library at the end of October 2008.

What reflections might we offer about this?

Reflection 1 – Do not give up though mitigate risk against external factors and exit if necessary.

The current volatility in the music industry and financial markets means that we are
challenged by factors operating globally over which we have no control.

  • Australian music is just as subject to global issues and trends as elsewhere, with the market dominated by the four transnational majors (about 70‐75% of sales) and by iTunes at c.90%. The implications for finding an aggregator able and willing to create an Australian sub‐set of recorded music that includes all the Australian independents and small music producers remains problematic.
  • Business models are in transition. The shift away from delivering downloads in WMA files with embedded DRM towards ubiquitous mp3 format is developing pace, partly in response to poor uptake by people frustrated with the limitations upon downloading. Australia’s small music distribution agencies are engineering new web services models towards this end. For Music Australia, this is a wait and watch game.
  • The battleground over file sharing, piracy, legal downloads and online licensing is rife, and is indicative of the turbulent industry environment. As well, there is a strong push in the creative industries and new music towards creative commons licensing, in direct opposition to the recording industry’s position.

No‐one in the Australian music industry is prepared to foretell the shape of the sector here in the short or even in the mid‐term. While the Library is still determining the way forward about the immediate problems posed by the terminated industry contract, we have simply had to exit one part of the business by turning off the e‐commerce facility. However, the ecommerce, not entirely successful, is the least of our concerns. Rather, our key focus is to work out an ongoing strategy for harvesting contemporary Australian music content directly into the service3, and to continue the major breakthrough we have achieved in being able to deliver in‐copyright music.

Reflection 2. Do not back away from risk and innovation!

There was always significant risk in aligning the service with a single commercial provider, but this decision was a well thought out innovative test‐case for the Library to cross partnership and copyright barriers that could have flow‐on effects for all our services. We released Music Australia 2.0 during a period of intense and rapid technology change, growth of online music, web 2.0 and change in user behaviour towards active interaction and social networking. But could we realistically expect that such a world will slow down or stabilise any time soon to give us time to catch up? Institutions like the National Library have had to learn in this environment to be more nimble, flexible, responsive, willing to innovate, able to discard or modify ideas and processes if they prove unworkable or as the patterns of use or the technology changes. We are currently working towards bringing all our web discovery services like Music Australia together under a single architecture, precisely so we can implement changes to our services more quickly and efficiently.

Reflection 3. Think ahead!

Sustain the vision and imagine the future but take small steps along the way.

Our vision for Music Australia includes:

  • Multiple suppliers and aggregators for music content, including the capacity for individuals to contribute music and update information (Note: user capacity for tagging, correcting and updating is already implemented in the Library’s new digitised Newspaper Beta service4 and has been remarkably successful).
  • Collecting and harvesting directly from the supply chain, and archiving with a flowthrough to end‐user access both from institutional repositories and via national discovery services.
  • Access to full content music resources, using various technologies including e‐commerce. The aim is to ensure that users can seamlessly get online the music they discover, while allowing for multiple commercial options.
  • Use rights management, licensing solutions and authentication processes to facilitate delivery of in‐copyright material to the public, especially as the ‘long tail’ becomes even longer.

This industry partnership setback in Music Australia has required us to pause, find an interim
solution while larger‐scale shifts take place in the marketplace, and continue our quest to
break through barriers to access.

Our message remains: be alert to strategic opportunities; catch these early; align goals with
partners who can build capacity; exit if necessary; find alternative business solutions. So,
watch the Music Australia space for the next variation on a theme!