Editorial Introduction to Issue 67: Changes Afoot
For readers who might have been wondering, I shall resist Mark Twain’s remark about reports of his demise being exaggerated, and reassure you that while Ariadne has been undergoing changes to the way in which it will be delivered to the Web, it has been business as usual in the matter of the content, as you will see from the paragraphs that follow. Issue 67, while currently not looking any different, is in the process of being migrated to a new platform developed to enhance functionality and give a more user-friendly look and feel to the publication. These developments are being approached with uninterrupted service uppermost in our mind while attending to the not inconsiderable legacy material the publication by now holds. Our intention is to ensure the current and previous content remain available without any break in accessibility to the Ariadne site.
In their article Towards Interoperabilty of European Language Resources, Sophia Ananiadou and her colleagues explain how a recent survey identified the growing trend of the Internet to exclude European citizens who cannot understand or speak English since so many transactions and information are in English. They provide an overview of the META-NET Network of Excellence (NoE) and its work to support European citizens through enhanced communication and co-operation across languages. A major outcome of this NoE will be META-SHARE, ‘a searchable network of repositories that collect together resources such as language data, tools and related Web services, covering a large number of European languages.’ The authors also describe the Unstructured Information Management Architecture (UIMA) Framework, which ensures interoperability of resources, and its U-Compare platform above it. Together they are designed to support the rapid construction of natural language processing and text-mining applications without the need for extra programming. In this way they can support automatic machine translation. In this way, the META-SHARE system serves not only to locate language resources, but also supports rapid prototyping and testing of applications destined to use such resources. The META-NET will therefore serve to preserve European cultural and linguistic diversity.
The authors introduce the Multilingual Europe Technology Alliance (META) which comprehends a wide range of stakeholders and which will undertake a ‘technology visioning process’ to innovate in the area of language technology applications. The META-SHARE platform will hold language data, tools and Web services for a large number of European languages. Work has begun on the preparation of resources and the aim is to encourage a diversity of innovation by accepting both toll-free and chargeable resources. The authors highlight the role of high-quality metadata in permitting users to locate the broad range of resources across a wide range of languages. They stress the importance of interoperability and comparability in this venture. They are concerned about the effect that different programming languages, formats and data types can have on compatability.
The authors describe the nature and benefits of the U-Compare platform including the GUI, workflow canvas, annotation functionality and a library of components. The platform also offers facilities to evaluate the performance of workflows and the option to substitute more effective components to improve performance.
The authors then go on to explain the role of the Unstructured Information Management Architecture (UIMA) in their system. This UIMA framework also serves to wrap resources as its own components, thereby rendering their original programming language irrelevant. They then proceed to describe the role of the Type System, the set of annotation types employed. They also explain how the use of a sharable type system serves as a partial solution to fostering interoperability across working groups at least at an intermediate level of hierarchy. The authors point to the commonality of processing approaches of these language technology applications, and the likelihood of a rapid increase in the availability of new applications without the need to write large quantities of new code.
Despite a decidedly philosophical tone to the opening of Christopher Blackwell and Amy Hackney Blackwell’s article, they assure us that Image ‘Quotation’ Using the C.I.T.E. Architecture is concerned less with the metaphysical aspects of quotation than with the practical considerations involved. More importantly, they are concerned with the quotation of images, a tougher undertaking if ever there were. They provide in their introduction a useful illustration of the ease with which it is possible to quote from an established textual work, such as the Hebrew Bible, as compared with the pitfalls involved in attempting to refer to a fragment of an image. Yet the increasing distribution of images within scholarly activity only exacerbates the failings currently encountered. The authors not only discuss the theoretical assumptions underlying their solution to image quotation but also detail technological implementation accompanied by examples from different disciplinary domains.
The authors introduce the work of the Homer Multitext Project and the nature of the data with which it works. In particular they point to the pre-existing high level of scholarly commentary relating to the primary sources with which the project members work. They explain how considerable attention will have to be paid to the scaling-up of the underlying digital service operating on these materials, with equal regard given to handling the technological changes to which all such systems are subject.
What perhaps I should highlight for readers at this juncture is the ground-breaking development that the HMT and similar initiatives represent. For the first time they offer scholars the opportunity not only to examine in depth and at leisure, manuscripts and commentaries which previously required enormous effort to access, but they place side by side sources which have never and probably never will be seen together physically.
They also explain the components of the infrastructure underlying the HMT Project, which is called CITE (Collections, Indices, Texts, and Extensions). It becomes clear that it is important to appreciate ‘the basic mechanism for linking in this digital library architecture: citation ‘¦’ and which provides the capacity to manage objects both concisely and in a very granular fashion. The essential point to appreciate is the use of URNs with which to express citations which, looking at the textual examples the authors supply, can be very concise and effective. In the instance of the Iliad, it can be used to retrieve a single line without difficulty. By defining an URN-structured citation for each scholarly object, whether primary source, commentary text, etc, the HMT can link and publish a comprehensive range of resources for users.
The authors further explain that the Indices (i.e., the ‘I’ in CITE) are the keystone of the HMT digital library infrastructure in that they list all the links between URNs or even a URN and an unformed element of textual data. Consequently, text-to-image linking becomes straightforward. Not only does the system have little difficulty in linking to images or fragments of images, but it does so in a manner that is independent of format or technology which makes it so flexible in the context of technological obsolescence. Moreover, by employing percentages in the URN, it is possible to ensure that URNs expressing ‘regions-of-interest’ in an image remain valid, irrespective of how the cited image is scaled.
Just as it was apparent that the placement of Homeric texts and commentaries in the virtual domain has generated new and intensified research, the application of such technologies in the biological domain has a potential to resolve the problem of an enormous backlog in the description and analysis of historical specimens. The fact that the number of specimens awaiting analysis far outstrips the capacity of the relevant scientific communities to attend to them is very amply demonstrated by the authors’ reference to the time it took to discover and analyse botanical specimens from Darwin’s voyage on HMS Beagle of 1831-36 - some 175 years! The principal difficulties revolve around physical access and the retrieval and management of physical artefacts. The potential of moving to a digital solution is consequently very important. While perhaps less mould-breaking than the potential for scientific discovery, the opportunities that such technology offers the teaching of biology are no less revolutionary.
Quite rightly the authors point to the importance of the persistence of the data contained in a CITE-URN: ‘the final result of this process is a canonical reference, abstracted from any ephemeral manifestation of files, formats, and services. Even if the CITE architecture falls by the wayside, the information in a CITE-URN would admit mechanical reproduction into a new useful format, provided only that the images continue to exist somewhere.’
Jez Cope and Geraldine Jones point to the growing appreciation of the potential of social media to support all aspects of the research cycle as grounds for the establishment of the Connected Researcher initiative. In Connecting Researchers at the University of Bath they assert the usefulness of social media tools in enabling researchers to network with like-minded people. They do emphasise that the use of such social media can only be effective in ‘the specific context or community’ of the user. The authors describe the aims of the Connected Researcher initiative in terms of awareness-raising, spreading of good practice, providing support to researchers in their decision making and in promoting standards, curation and preservation. But in so doing, they indicate the dangers inherent in an over-prescriptive approach which can stifle innovation. Instead, they adopted the approach of social learning, bringing more and less experienced users of social media together at their events to support the sharing of digital networking techniques. The authors then describe how they linked the Connected Researcher events to the research cycle associated with the four stages of the cycle as described by Cann et al. In describing the events to support the initiative’s aims, the authors highlight the guidance offered by Dr Tristram Hooley, Head of the International Centre for Guidance Studies (iCeGS) at the University of Derby and co-writer of Social Media: A Guide for Researchers. His advice covered the value of online networking including Twitter.
Whether psychologists would agree with the authors that conference delegates who interact with the event via transmissions over Twitter are automatically ‘more engaged’ than all those who do not, assumes a universal passivity on the part of the latter that does not appear to be justified. However, there is no doubting the usefulness of Twitter as a collaborative and supportive tool, and nowadays a common form of event amplification. Moreover, the authors are in my view entirely correct in their assessment of Twitter as a means of creating and maintaining contacts. The initiative’s second event concentrated on the value of RSS to researchers and the considerable power it holds to help them filter information.
The authors go on to describe their marshalling of four of the University of Bath’s bloggers to support their view of the value of blogging to researchers. They highlight the particular value of blogging at the initial stages of one’s research where the act of writing, albeit informally, forces one to put down ideas in a coherent fashion, which not only allows for more objective review by the author, but also by peers in a less formal context. It is interesting if not surprising to note, moreover, that the team working on the Connected Researcher initiative enjoyed far greater success in garnering support for their activity over the very social media they wished to promote than they did through more ‘traditional’ channels. They also raise the useful point in respect of the sustainability of a blogging platform for researchers, explaining that while institutional platforms offer greater persistence, there is the problem of what to do when one moves from that institution.
They are also to be congratulated in my view by refusing to be seduced by the technology, concentrating instead on the human behaviour that determines usage. They do well to underline the importance of disciplinary links and relationships to researchers, something forgotten at one’s peril. In their conclusions, they describe the stage along the road of technology adoption where many researchers find themselves: aware, even interested, but adopters are still in the minority. They identify the means of encouraging greater usage as the resolve to dismantle the barriers to adoption, principal among which is the time required to find and learn such tools.
In MyMobileBristol, Mike Jones, Simon Price, Nikki Rogers and Damian Steer describe a project with ambitious aims to support the delivery of time- and location-sensitive data in the Bristol area to mobile phones. It benefits not only the students and staff at the institution but also operates in partnership with the municipal authority. The authors describe MyMobileBristol in the Mobile Campus Assistant Project which had similar aims but narrower scope since it did not require access to restricted components of mobile phones. They also highlight the importance of Semantic Web technologies in their work and their involvement in this project where they would serve to aggregate content on a more ambitious platform.
The answers they intend MyMobileBristol to offer users were highly practical ones such as bus times to halls of residence, nearest Wi-Fi hotspots, etc. Unsurprisingly they achieved rapid take-up. Initial stakeholder analysis of the system provided the project with a new set of use cases which would only increase its popularity. They also describe the opportunity for collaboration the project has offered in the context of innovation in online markets. They also point to the value of the one-day workshop co-organised with the UKOLN-based DevCSI Project in furthering the discussion of use cases, problems and solutions. In addition to the long-standing methods of communication with its stakeholder community, the authors describe the features of the collaborative platform implemented to increase that contact.
They also describe the issue of handling the public and more private aspects of the mobile platform’s development in this co-operative context, as well as the contribution JISC Techdis has made to the development of its stakeholder engagement plan. The authors also explain the importance of Bristol City Council to the project and how the latter can contribute to the Council’s plan to reduce its carbon footprint radically through the adoption of smart applications. These trends are further supported by local and national authorities to provide more data to the public, and despite inevitable difficulties surrounding geo-location data, the City Council is keen to collaborate with the project.
The authors go on to describe the software with which the project is being implemented and provide an overview of the architecture developed. They also describe the continuing consultation with stakeholders, including interviews and deployment of a usability expert. Despite its beta status, the MyMobileBristol platform continues to extend the data it offers and to receive requests for more personalised content such as lecture timetables. In their conclusion, the authors identify the potential difficulties handling and reusing data from various provenances may produce, but point to the considerable benefits this project will provide, including its potential value to JISC itself.
In Piloting Web Conferencing Software: Experiences and Challenges, Julian Prior and Marie Salter introduce Elluminate Live! as a Web conferencing application which may represent an attractive option for institutions keen to make efficiency savings. They point to the most obvious usage and how Web conferencing has emerged as a firm favourite for users eager to see and hear colleagues in a face-to-face situation, but over the Web, with attendant cost-benefits. They then go on to the amplification of events, pointing out that Web-conferencing software can also support a comprehensive webinar process for users. The authors describe the pilot activity begun at Bath which operated in the context of CPD (Continuing Professional Development), but which soon was looking at exploiting the Access Grid which offers robust support for live broadcasting. They also describe the drawbacks. The authors go on to draw comparisons across other synchronous Web-conferencing applications using standard criteria and explain their reasons for choosing Elluminate.
They then provide details of their usage across the tool’s wide functionality, including use scenarios such as interviewing, conference activities and a virtual open day (in order to satisfy the demand for information on a new teaching programme), virtual team meetings, off-site evaluation of campus-based students’ work, multi-located seminar discussions and group work. The authors then provide a very useful insight into the difficulties encountered in the pilot stage, many of which did not emanate from the application, but the usual problems when adding hardware and software, to which they provide some useful and commonsense solutions.
In their article Retooling Special Collections Digitisation in the Age of Mass Scanning Constance Rinaldo, Judith Warnement, Tom Baione, Martin R. Kalfatovic and Susan Fraserdescribe the work of the Biodiversity Heritage Library (BHL) on the development of a cost-effective and efficient workflow for the digitisation of its special collections. It becomes apparent in this article the value that is attached by researchers to the historical data to past biodiversity research and the importance of making them digitally accessible to researchers. They note that the heterogeneity of special collections can impede the scaling up of mass scanning operations. In addition to effort on the workflow aspects, the group would also seek ‘to explore ways to enhance metadata for library materials that are designated as “special collections”.’
The project relied considerably on meetings to provide comparisons not only of digitisation costs but also techniques, and it invited librarians from a range of institutions. Its second meeting established the parameters to begin small-scale tests on same or similar items across the institutions. In its third meeting the group was able to discuss the test results by vendor. The authors explain the characteristics of books which render them unsuitable for mass scanning. They explain the partners had developed a detailed set of criteria in this respect and that the rejection rate for such specialised items was higher than initially anticipated, particularly for natural history materials. Reasons ranged from the mundane (oversize dimensions) to the worrying (items of such high value as to occasion security concerns).
In respect of the outcomes of the scanning tests, the authors indicate a very discernible gap between institutional and commercial costs of digitisation.
They offer comparisons between the cost and performance of in-library and institutional scanning and those of the ‘institutional boutique’ solutions with state-of-the-art technology but higher costs. They also describe their investigation of the potential benefits from Web services and social networking and the involvement LibraryThing. They determined that adding content links in Wikipedia did increase the number of referrals to the BHL site and that Wikipedia and LibraryThing were the most promising locations for user-generated feedback and tagging.
The authors then turn their attention to the current situation, where much of the biodiversity literature which could easily be scanned has now been processed. What remains is the more difficult material which is nonetheless of equal importance to the scientific community. They do emphasise the value of good bibliographic records in building a shared catalogue, and that enhancing metadata ‘is a better investment than packing and shipping special collections materials to off-site facilities.’ They also point to the benefits of a portable scanning unit in dealing with fragile items from their special collections.
Faced with the problem of some 1,200 non-functional links on the Institute of Education’s library catalogue, Bernard M Scaife began his work there as Technical Services Librarian by setting about solving the link rot problem. In From Link Rot to Web Sanctuary: Creating the Digital Educational Resource Archive (DERA) he describes how tackling the affected links manually proved quite instructive, and he found a series of characteristics behind these link failures. One particular group of failures related to missing documents of which many, it turned out, could be classified as official publications frequently untroubled by any familiarity with persistent identifiers, thus exacerbating the problem. It soon became clear that a more structured solution was needed, and so the Library’s own research repository with EPrints software, hosted by ULCC, was pressed into service where the persistence of a citation which linked to a record in a Web archive looked more resilient. As is so often the case, the matter of IPR soon raised its head in such a venture, and the increased threat to the longevity of government content occasioned by the Coalition Government’s announcement of public sector cuts meant that many content holders welcomed the project’s offer of assistance. With the likelihood of a stream of documents suitable for inclusion in the Digital Education Resource Archive (DERA) assured, the author and his colleagues turned their attention to the adoption of suitable metadata, beginning with their EPrints repository fields. While many were retained as useful, among new fields introduced was that of ‘Organisation’ since the team was extremely conscious of the possible demise of content-holding entities and all that that would involve. Bernard describes how EPrints is capable of flexible handling and how he and his colleagues altered their approach to cataloguing documents which might no longer exist. They developed a process involving retrospective data loading. The first significant instance of a content-holding agency’s disappearance proved to be that of BECTA whose electronic document archive was entrusted to DERA. By February 2011 the DERA was at a level of operation such that it could be officially launched.
In Looking for the Link between Library Usage and Student Attainment Graham Stone, Bryony Ramsden and Dave Pattern begin by giving the background to their successful bid in the JISC call through the Activity Data programme, explaining the data collected informed a co-relation between library usage and degree results. Were Huddersfield’s findings an anomaly or the norm? If the latter, how might they be used to increase student attainment? Huddersfield and seven partners were awarded funding by JISC to investigate these questions. In stating their aims and objectives, the authors maintain that library use is a cause of increased student knowledge and so attainment, but recognise that other factors have an influence.
The authors go on to describe how the Library Impact Data Project (LIDP) took a critical stance towards the hypothesis that library usage influences attainment, pointing to circumstances which indicate that library usage is not the only factor. Degree courses differ, the availability of resources inside and outside the library likewise. They also point out a number of other variables, including digital information literacy, as factors that must be taken into consideration. In discussing the nature of data collection, the authors point out the value of collecting data on multiple years of library usage, since results could be obtained across the lifetime of that degree course, where behaviour, unsurprisingly, altered over time. They are very interested to compare findings with their partners. They describe the data acquired and explain shortcomings. While some partners are experiencing problems in the collection of certain types of data, the project does seem confident of sufficient information to fulfil its aims. On the other hand, one partner has extended the analysis by including PC log-in data.
The authors go on to discuss the variability of releasing their anonymised data for reuse, but recognise that there could be difficulties which might limit their ability to do so. They emphasise the care with which all data are being anonymised, right down to excluding data from small, and so identifiable, courses. Added to these quantative data, all partners are seeking qualitative data in focus groups of students. While the authors note that their project’s duration inevitably limits the scope of their findings, they are confident that a number of approaches could be adopted in any subsequent activity.
As usual, we offer our At the Event section, as well as reviews of: a timely collection of essays that highlights the values of institutional leadership and resourcefulness in academic librarianship’s engagements with Web 2.0; a recently published book on the selection and preparation of archive and library collections for digitisation; and a work which takes a hard look at academic libraries, how they are being redefined and what skills will be required of the staff who will move them forward.
I hope you will enjoy Issue 67.
Author Details
Richard Waller
Ariadne Editor
Email: ariadne@ukoln.ac.uk
Web site: http://www.ariadne.ac.uk/