What is RDA?

RDA is a content standard that ‘provides a set of guidelines and instructions on formulating data to support resource discovery’1. RDA both replaces and builds upon AACR2. No further developments to the AACR2 have been made since 2005. Instead, the work of rule creation, revision and amendment has been focused on RDA.

RDA is founded upon the entity-relationship model of Functional Requirements for Bibliographic Records (FRBR)2 and Functional Requirements for Authority Data (FRAD)3. This is reflected both in the stated purpose of the code and the structure of the code as text. In contrast to AACR2, which is based around types of materials and headings, individual chapters giving rules for specific material types, the structure of RDA is based around FRBR and FRAD entities and their relationships. This change is a significant departure and forces the cataloguer to think about the resources being described within an FRBR framework4.

The descriptive content of resource descriptions created with RDA supports the FRBR user tasks of ‘find’, ‘identify’, ‘select’ and ‘obtain’5. Controlled access points (previously headings in AACR2) describe an entity associated with a resource (for example, a creator) and support the FRAD user tasks of ‘find’, ‘identify’, ‘contextualize’ and ‘justify’6.

For a relatively young cataloguing standard, RDA has had a long history, being released in draft form for community review in late 2008. RDA was formally released as the web-based RDA Toolkit7 in June 2010 and went into official field test with selected US libraries from October to December 2010. This was followed by a report from the field test libraries8 and subsequent official response in May/June 20119. Whilst the field test revealed nine principal reservations, as the Joint Steering Committee for the Development of RDA (JSC) and the publishers had made progress to overcome these, the official ‘go live’ date was set for 31 March 2013. This date was adopted by the Library of Congress, the British Library and other libraries both in the US and internationally. More libraries, such as Die Deutsche Nationalbibliothek, Libraries and Archives Canada and the National Library of Australia, have set dates for their own adoption in the period up to 2015.

“RDA has deliberately liberated content from syntax.”

RDA prescribes changes to both descriptive content and controlled access points – what AACR2 termed ‘headings’ – within bibliographic and authority records. The intention has been to make RDA metadata compatible with legacy AACR2 metadata. However, there are significant differences between the two standards.

RDA has deliberately liberated content from syntax. Where AACR2 was wedded to International Standard Bibliographic Description (ISBD)10 punctuation, for example, RDA has separated the content, the metadata created, from its mark-up. Whilst ISBD punctuation can still be used with RDA, and in most current cases is probably the only syntax used with RDA, it is now a matter of choice or system context that would govern its use. ISBD instruction is now reduced to an appendix within the code11.

From the outset, RDA has aimed to achieve significant traction outside both the traditional library community and the English-speaking world. Outreach activities with the Dublin Core community via the DCMI/RDA Task Group have resulted in greater engagement with the broader metadata community. Representation from the non-English speaking world in the governance of RDA has been established with the appointment of a representative from Die Deutsche Nationalbibliothek on the governing Joint Steering Committee for RDA.

RDA as innovation

RDA has the power to disrupt, and the potential to improve, change and transform cataloguing. It is a departure from how cataloguers have previously understood and described resources. The introduction of the concept of entities and their relationships can deconstruct the single bibliographic description into a series of statements that can be represented as linked data. It is an innovation in descriptive cataloguing, conforming to Everett Rogers' definition of an innovation: “An innovation is an idea, practice or object that is perceived as new by an individual or other unit of adoption.”12

“RDA has the power to disrupt, and the potential to improve, change and transform cataloguing.”

Cataloguers' reception of RDA has not been universally positive as evidenced in list archives for the RDA-L discussion list, where attitudes have ranged from positive, to resigned, and explicitly hostile.13 In the adoption of RDA, libraries need to see the benefits that the new code brings; however, these benefits are not always clear to librarians. Rogers states: “A technological innovation usually has at least some degree of benefit for its potential adopters, but this advantage is not always clear-cut to those intended adopters.”14

Rogers' work on the diffusion of innovation can help us understand the adoption of RDA contextually. He also provides a framework for interpreting the pattern of adoption by libraries. Rogers defines diffusion as “the process in which an innovation is communicated … over time among the members of a social system. It is a special type of communication in that the messages are concerned with new ideas.”15 RDA, as an innovation in cataloguing, will be communicated in this way. The adoption of RDA will bring change to the cataloguing community: change to the rules applied in cataloguing and change to the way in which cataloguers work.

Moulaison recognizes the impact of the social context on the adoption of innovation within cataloguing: “Changes to the cataloging code … represent new practices that will have to be navigated by cataloging professionals. These innovations challenge the way librarians will think about data and metadata, and have the potential to be much more difficult to implement within the library social system than innovations … adopted by individuals or small groups of individuals”16.

However, the most entrenched and hostile opinions softened in the run-up to the 31 March 2013 implementation date and in its aftermath. It is worth noting that discussion on RDA-L has noticeably shifted from the negative and resistant to that of planning and, at the micro level, individual rule interpretation. This pattern is reflected in the published literature on RDA.17 Earlier articles, webcasts and presentations on RDA were concerned more with its theoretical nature, particularly in relation to FRBR.18 Now there is a noticeable shift towards the practical implementation of the code.19 As RDA is increasingly implemented, librarians are seeking advice from those with implementation experience. This shift in the published literature is mirrored by a similar move from an article-centred literature on RDA to a growing proportion of monograph publications that combine theoretical with practical issues.20

Rogers' criteria by which innovations can be measured can be adapted and applied to RDA21:

  1. Rate of adoption: the speed at which RDA is taken up.
  2. Relative advantage: the degree to which RDA is perceived to be better than earlier cataloguing rules (AACR2).
  3. Compatibility: the degree to which RDA is consistent with both cataloguing needs and the cataloguing community.
  4. Complexity: the degree of difficulty in applying RDA.
  5. Trailability: the degree to which RDA can be tested on a limited basis.
  6. Observability: the degree to which the output of RDA, as metadata produced following the rules, is visible.

Taken together, these criteria give a set of tools by which the success of RDA can be assessed and, in particular, the rate of its adoption.

“… the question for libraries is not so much ‘if’ but ‘how’ they will adopt RDA.”

RDA adoption

Libraries are not simply faced with a choice of whether to adopt or not. The situation is more nuanced, though only those libraries who act in complete isolation from others will be able to refuse to adopt the standard. Any library that derives data from external sources such as WorldCat, RLUK, shelf-ready vendors and PDA programmes will be required, to some degree, to deal with RDA metadata. Put simply, there is no ‘outside’ of RDA and its impact. Thus, the question for libraries is not so much ‘if’ but ‘how’ they will adopt RDA.

RDA implementation will follow the five adopter categories defined by Rogers. These categories help understand the pattern of adoption of RDA: “The individuals in a social system do not all adopt an innovation at the same time. Rather, they adopt in an over-time sequence, so that individuals can be classified into adopter categories on the basis of when they first begin using a new idea.”22 The characteristics of the individuals identified can be applied to individual libraries (as shown in Table 1):23


Category Category size Characteristic Definition

Innovators 2.5% venturesome Interest in new ideas leads them to leave the social system. RDA field test libraries.
Early adopters 13.5% respectful Integrated into the social system. This category has the highest proportion of opinion leaders. Adopt RDA before the majority. Libraries going live on, in lead up to, or immediately after, 31 March 2013. RDA is currently still in this phase.
Early majority 34% deliberate Adopt RDA just before the average. Libraries with a planned schedule for adoption within the short to mid term (e.g. before end 2013, early 2014)
Late majority 34% skeptical Adopt RDA just after the average. Currently waiting until they perceive that the time is right to adopt. Cautious approach to adoption.
Laggards 16% traditional The last to adopt. Resistant to change but will, inevitably, have to succumb. Libraries that hang on to AACR2 and do not believe that RDA offers a better alternative.

Table 1

The five adopter categories as outlined by Everett Rogers

RDA adoption categories

The mean point, expressed as the average time of adoption, lies at the boundary of the early majority and the late majority. Based on Rogers' assumptions, most libraries will have adopted RDA with the start of the late majority.24

RDA and hybridity

In adopting RDA, the decision facing libraries is twofold. Whilst adopting RDA for current cataloguing, legacy data also needs to be addressed. There are three options available:

  1. Leave all legacy data within the catalogue described with AACR2, or earlier, rules;
  2. Retrospectively convert all legacy data to RDA;
  3. Selectively convert critical parts of legacy data to RDA.

The third of these options is the most practical if the aim is to deliver a uniform discovery experience with finite resources.

Within the library catalogue, RDA will deliver two types of hybridity at the micro and the macro levels.

At the individual record (micro) level, controlled access points will be created in accordance with RDA rules which may be in conflict with those created under AACR2. The library needs to decide how to adjust to these changes. Libraries that create authority records or follow the Library of Congress Name Authority will see these changes immediately. They must then address the issue of harmonizing this new data with any legacy data. Failing to do so will deliver an inconsistent discovery experience as, for example, names fail to file together, thus creating additional index entries. If the older form of the access point is not retrospectively converted, there will be two index entries for the same entity. Retrospective conversion will result in hybrid records, combining AACR2 descriptive practices with RDA-controlled access points.

At the catalogue (macro) level, the library catalogue or discovery layer will also become a hybrid, combining records created under AACR2 rules, RDA rules, and a combination of both in individual, hybridized bibliographic records.

Conclusion

The acceptance of RDA, or at least the resignation to the inevitability of its adoption within the cataloguing social system, has the characteristics of a successful innovation.

It is assumed that the pattern of adoption of RDA will follow Rogers' general predictions. Libraries will inevitably adopt RDA. Clearly the best option for libraries would be to actively choose and plan for adoption at a time of most convenience, understanding what the issues are. The manner of adoption will be either positive – actively choosing – or in a spirit of resignation at having to adopt. The adoption characteristic will depend upon where in the adoption continuum a library is positioned.

“… the best option for libraries would be to actively choose and plan for adoption at a time of most convenience …”

RDA will leave the library with a decision as to the level of hybridity it is prepared to accept both at the micro and the macro catalogue level.

The bigger question for RDA is the level of its adoption outside its natural constituency. In these non-library environments, another, slower pattern of adoption is likely to emerge. Future systems and environments based on linked data that implement RDA will require that the flat MARC data structure be broken apart in favour of individual entities and their relationships. There will be a loss of the unitary catalogue record and the emergence of linked statements about resources. This will show how innovative RDA is if it is successfully implemented in these future environments.