This article consists ofinterviewswithpersonalities of Library and informationscience (LIS)field,realized asking three questions to five internationally recognized scholars. Throughthese essays,theAIB Study group on Cataloging, indexing, linked opendata and semantic web (CILW) wants to set up its activity starting by a detailed evaluation of the ‘classic' and ‘innovative' theories and principles, and of the views of scholars onthem.Itispossible,thus,toreviewcurrentdevelopmentofinformationandresources description and organization, in order to generate a fruitful debate and a necessary comparisonalsowithdisciplinesneighbortoLIS, startingfromtheorists andprofessionals that are mostfocused onthe issues of change, whose researchactivities are also linked to the ‘paradigm shift' warned in our disciplines.
So, questions developed by the group are intended to call attention to some fundamental questions. First, it would help clarify ideas about the role a catalog may stillhave,inlibraries andforthe society:ifitwill continue tobe a centraltoolforresearch activities,orwhetherits rolewill continue todeclineuntilitis alignedorovershadowed comparedtotheother currenttools forinformationresearchanddiscovery.The catalog, thus, is in front of the semantic Web, the linked data methodology, and the need to open up to other communities, in orderto be the preferred toolfor users.
The lack of effectiveness of catalogs in information searching is often due to the transformations ofthe search objects, namely the knowledge resources: itis necessary, then, wondering about what might be a definition forthe new object ofthe catalog, what are actually the resources of knowledge, and whattools are more suited to treat them. Resources useful to knowledge today are very different from each other and from what has so far seen as ‘document',their main quality is the interest users have inthem,they are placed in multiple webs of meaning and belong to multiple contexts. Finally, regardless of the technological tools best suited to the needs of society, it is necessary clarify policies and methodologies desirable for the treatment of resources, information and data, that remain the primary object of any knowledge organization and managementtool: itis increasingly importantthat data are correct and reliable, and this need grows as much as itincreases their possibility of growing and dissemination. Description should be decentralized, being confident ofthe work of others, fostering integration in the semantic Web, but without losing the data credibility.
In view of the possible answers, it is always essential to maintain an appropriate balance between‘conservation' and ‘revolution'. Seenfromdifferent perspectives,the basic principles, more or less classic as they are, can be reconsidered at a theoretical level, based on their strengths and critical points, in orderto set up a project that can be said to be truly innovative in terms of theoretical and methodological aspects for our area of investigation. A real progress and a real innovation are obtained not only by ‘overcoming' principles and practices already obsolete orjustinline withoutdated realities – as the wine becoming old and spoiling –, but also by developing and ‘reconsidering' that stable principles, unrelated to temporal or epochal shifts, which remainvalid forthemaintenance and development oflibrary and informationscience disciplines – as the wine aged in solid barrels.
If the underlying principles are not adequate, up-to-date and at the same time durable, it's no use proposing a new container, be it technological or conceptual, to try to present them as a real methodological innovation.
Catalog and the ‘search'
Is the catalog, as we know it today, destined to lose its central role among other knowledge and search tools pertaining to libraries?
New information search and retrievaltools allow the expansion of selecting and mediating activities far beyond the classical boundaries held by the library,towards the endless network of information and resources available via the Web. But, if establishing reliability and authority of many of its contents is not always easy, the idea behind the semantic Web is, however, the creation of directories from which to capture data through controlled and reliable sources. In this new perspective, can a ‘unique', shared and participated catalogue exist? Can an open source catalog exist, populated and maintained by experts with new skills, where cataloging and creation of ontologies and dictionaries will approach more and more?
A current concept of resource
In librarianship and library science the concept of ‘document' was a steady element for a long time. Now the International cataloguing principles (ICP)restrictthis definition to some archival resources. Moreover, in today's cultural scenario, it is necessary to include different kinds of objects that can be recorded, and, as a consequence, described and used as reliable information, and possible sources of knowledge. In this perspective, which could be a viable and functional definition ofthe new concept of ‘resource', and what could be its identity?
In short: whatis the object we have to deal with? How can we look at something ‘fluid' and objectify itinto a record, a catalog, a reference system?What do we consider as an information unity to be described and linked? A forum thread and/or a single post? A symphony or each of its movements? ‘Authoritative' data
By applying linked open data (LOD)to catalogs and otherinformation search tools, libraries and other cultural heritage institutions will have to become accustomed to not having the ‘ownership' of data with which to describe and mediate resources. This results in a true paradigm shift that renews the documentation tradition, and digital tradition on the whole, from the definition of the concept of information. In this scenario, how can the different cultural institutions maintain the role of authority in the process of production, dissemination,retention and maintenance of data, and being the gatekeepers for their quality?
Will the different institutions be able to integrate well into the semantic Web project, and develop a more accurate interpretation of their own identity?
The view of Marshall Breeding
In many respects,the catalog has already undergone a major shiftfrom its traditional form of an online search tool focused on the print inventory of a library. Today, libraries work to provide their communities with broader discovery services that address the full breadth of their collections, spanning all types of materials. Unlike online catalogs that returned titles of books or journals, it's essential now also to provide access to individual articles, book chapters, or digital objects and to display full text or visual representations whenever possible. These discovery services must pay close attention to user experience, providing access to library content using the interface conventions and techniques to which the public has become accustomed in their daily experience oftheWeb. These interfaces must also be optimized forthe mobile devices or tablets that represent the majority of use rather than the larger screens oflaptop and desktop computers. The genre ofindex-based discovery services has found a great deal of acceptance in academic libraries, with large proportions adopting them in addition – or increasingly instead of – traditional online catalogs. Public libraries demand discovery services or catalogs that provide equilateral access to e-books as well as print books. Through initiatives such as ReadersFirst, public librarieshave demanded and suppliershave provided dramatically simplified methods to discover, select, and download e-books to reading devices.
These new types of discovery interfaces represent only one thread of activity. Linked data and semantic Web technologies have also gained increasing interest. The universe of resource of interests to libraries exposed as linked data has grown considerably in recent years, through the majority of the body of scholarly content remains locked within proprietary publishers. Libraries have made important progress toward shifting from record-oriented bibliographic description, primarily though the various MARC formats to constructs based on linked data, especially BIBFRAME. Developers of many library management systems and discovery services have begun work to incorporate BIBFRAME,though we remain more in a time of experimentation and prototypes than operational products. I anticipate hybrid systems that will use relational databases and indexes fortransaction-oriented business processes butthat will increasingly tap into the universe of linked data to supplement discovery, access to content, and visualization of results.
Semantic Web technologies will also help libraries improve the discoverability ofresources of interestto their communities. We must acknowledge the reality that most users do not come to the catalogs orinterfaces thatlibraries provide, butrather rely on general search engines. The incorporation of semantic coding in the presentationoflibrary resources, suchas defined inschema.org, dramatically improves the discoverability ofresources beyond library interfaces. Since patrons don't come to the library,the library must work hard to make library resources discoverable and available in the places on the Web they inhabit.
This transition from record-orient catalogs to semantic Web technologies does not necessarily mean less controlforlibrarians in the information ecosystem. Libraries will continue to curate and describe local collections, butincreasingly using metadata structures based on linked data, RDF, XML, and less using library-specific records and protocols of little interest to the broader information ecosystem. While MARC has served libraries well in some ways, it has also led to some degree of isolation relative to publishers, the e-commerce sector, and other information-oriented industries. Adoption of the semantic Web may result in improving the relevance and positionoflibraries.It also means just as muchwork as everincreating descriptions forresources thatrelies on authoritative sources, butincreasingly expressed as links and relationships and less as self-contained records. The core roles of libraries have persisted throughout many phases of society and cycles oftechnology. How librarians carry outthat work evolves in accordance with changes in the broader environment.
The incredible diversity of resources
Library collections today include diverse types of materials. While print documents, monographs, photographs, and manuscripts constitute a part, libraries also collect many different types of digital content. Academic libraries, for example, generally expend the vast majority of their collection funds on access to scholarly and professional articles provided through subscriptions to content packages offered by commercial and non profit providers and take advantage of the growing body of open access materials. Digital objects form another vital part of library collections. Many libraries have longstanding programs in digitizing books, newspapers, manuscripts, photographs, and otheritems of historic or academic interest. Cultural heritage is increasingly represented in native digital form. Professionals in libraries, archives, and museums routinely describe materials in alltypes of media and formats, taking advantage of a variety of applicable metadata standards and schemas. Any formally definition of what constitutes a document or set of cataloging principles must encompass the incredible diversity of materials and their corresponding metadata conventions. No single static definitionof what constitutes a library resource can necessarily accommodate the continually expanding range of materials collected and curated by libraries and related cultural institutions.
Decentralization of bibliographic description
Increased involvement by libraries and related institutions in the creation of tools based on linked open data represents a positive step since it provides increased opportunities for the discovery of library-oriented resources. This transition may mean a more decentralized model of bibliographic description. The elements of bibliographic description may reside in RDF triples instead of MARC records stored in centralized databases, but the same type of intellectual effort will be required to support the management and discovery of collections. National libraries, regional consortia, and organizations such as OCLC will continue to play seminal roles following the anticipated transition to linked open data. Other players may also emerge. The concept of ownership of bibliographic data may well be further diminished. We are already seeing more freedom of exchange and less assertion of ownership as bibliographic records become more of a commodity and as expectations grow for collaborative exchange. Projects such as Europeana require contribution of bibliographic records under the most liberal Creative Commons license (CC0) regardless of whetherthey derive from OCLC or other sources that previously asserted some degree of ownership. The eventual transition to increased involvement in linked open data will not necessarily fundamentally change the role of libraries in creating and describing collections, but will hopefully provide a variety of benefits. The creation of high-quality metadata will likely continue to take place in ways that can be collaboratively distributed throughout the global library community. Even as the containers that convey bibliographic description change dramatically, the operations, strategies, and values of libraries will endure.