Issue 19, 2013-01-15
Editorial Introduction: A Peer Network
Code4Lib, and the Code4Lib Journal, considered as a peer network.
Building a Library App Portfolio with Redis and Django
The Tutt Library at Colorado College is developing a portfolio of library applications for use by patrons and library staff. Developed under an iterative and incremental agile model, these single-use HTML5 applications target multiple devices while using Bootstrap and Django to deliver fast and responsive interfaces to underlying FRBR datastores running on Redis, an advanced NoSQL database server. Two types are delineated: applications for access and discovery, which are available to everyone; and productivity applications, which are primarily for library staff to administer and manage the FRBR-RDA records. The access portfolio includes Book Search, Article Search, Call Number, and Library Hours applications. The productivity side includes an Orders App and a MARC Batch application for ingesting MARC records as FRBR entities using RDA Core attributes. When a critical threshold is reached, the Tutt Library intends to replace its legacy ILS with this library application portfolio.
A Comparison of Article Search APIs via Blinded Experiment and Developer Review
This study looks at perceived user preference between products that can provide a scholarly article search service via an application programming interface (API). The study set up a blinded review and asked users at Johns Hopkins to select the service that provided the most useful results. Few statistically significant preferences were detected, and some interpretation is provided of what the results might tell us. The specific products evaluated for this study are: Serials Solutions Summon, Ex Libris Primo, EBSCO EDS, EBSCOHost ‘traditional’ API, and Elsevier Scopus. Re-usable open source tools for implementing article search were created to support the study and future development, and a developer review of the APIs is included based on the developer’s experience in this implementation.
Providing Information about Reading Lists via a Dashboard Interface
As developers of the open source LORLS Resource/Reading List Management System we have developed a dashboard to better support academic staffs’ understanding of how their students use reading lists. This dashboard provides both graphical and tabulated information drawn from LORLS and the Aleph Integrated Library System.
Development of the dashboard required changes to back-end functionality of LORLS such as logging views of reading lists and caching of loan data. Changes to the front end included the use of HTML5 canvas elements to generate pie charts and line graphs.
Recently launched to academic staff at Loughborough University, the dashboard has already garnered much praise. It is hoped that further development of the dashboard will provide even more support for academics in the compilation of their reading lists.
Visualizing Library Statistics using Open Flash Chart 2 and Drupal
Libraries continue to need to demonstrate their value to stakeholders, and while statistics alone do not represent value, they are an important element. We found ourselves, and our stakeholders, uninspired by our infrequently updated bulleted list of statistics on our website and so set out to create a more dynamic and visually appealing look at our statistics. This article outlines how we used our content management system, Drupal, Open Flash Chart and custom programming to convert library statistics into Flash charts, including how to populate the graphs with dynamic data from external sources. The end result is our Library Statistics Dashboard (http://library.uncw.edu/facts_planning/dashboard) that visually demonstrates the use, activity and resources in the library via interactive and visually interesting graphs.
Library Widget for Moodle
Any course within a course management system is generally considered the intellectual space of the professor teaching it. Research tools and guides, such as search boxes for discovery services or links to course-specific and subject-specific guides, are created and maintained by librarians. In trying to get our tools and services closer to where students spend their time devoted to coursework, Oakland University libraries have developed a library widget – a self-serve code generator that allows professors to select what tools and services they want to bring into their course space. This approach has proven to be flexible, because it does not depend on a library presence within the course management system. It also offers persistent presence within courses since professors can archive courses, including the library widget, at the end of a semester and restore them in the system in future semesters. We are using the library widget as a pilot to inform decisions on future full integration of such functionality into Moodle.
Open Source Library Software Development in a Small Rural Library System
Using the Crawford County Federated Library System’s development of an open source web kiosk management system, as an example, this article will illustrate how an open source library project is defined, specified, written, tested and rolled out. The article will also discuss how the project was released as an Open Source project and future development of the project. The web kiosk project is called Libki and was written to authenticate users and allow access to the Internet kiosks based on time limits. Libki is a completely Open Source project and is now used by multiple libraries across the US. The client side of Libki is cross platform and supports multiple operating systems including Microsoft Windows and Linux. The administrative side of the program allows access to user logs, controls time and access and allows the librarian to log a patron off the system in real time. Libki was completely developed and written by staff members of the Crawford County Federated Library System.
Determining Usability of VuFind for Users in the United Arab Emirates
In late 2011, the Higher Colleges of Technology, a higher education institution in the United Arab Emirates, implemented Vufind as the search interface for the libraries’ resources. Before launching Vufind in the 2012 academic year, usability testing occurred across three campuses to test the functionality of the search interface features. Twenty-one participants, including Emirati students and expatriate faculty, were tested using a performance based assessment along with think-aloud protocol, which was recorded using Camtasia screen capture software. As a result of the testing several features of Vufind were customized including language, layout and prioritization of results. The current study builds on the limited existing body of literature on Vufind, which has previously indicated a number of design elements and practices which should optimize user experience. Several key findings are consistent with and confirm results from prior studies with findings from this study adding to the literature by observing how or why linguistic orientation affects user behavior in search systems.
Using XSLT and Google Scripts to Streamline Populating an Institutional Repository
The College of Wooster has created a process that allows library staff to quickly populate institutional repositories. An XSLT script is used to transform RefWorks citations into Dublin Core XML and batch load those records into the institutional repository. A second script in a Google Docs spreadsheet then looks up publisher permissions in Sherpa/RoMEO. The resulting workflow has dramatically reduced the amount of time necessary to populate an institutional repository with faculty scholarly articles.
Indexing Linked Bibliographic Data with JSON-LD, BibJSON and Elasticsearch
Linked Data is a powerful tool for sharing bibliographic metadata. By combining the decentralization of the web with the use of globally defined metadata vocabularies, data from many sources can be treated as a single, aggregated graph. Supporting search across these distributed data sources within the same application, however, requires considerable work in vocabulary alignment and data transformation. Aggregate systems must convert data into a unified model which must (almost inevitably) be generic at the expense of the structure and granularity of the original data. This paper presents a novel solution for representing and indexing bibliographic resources that retains the data integrity and extensibility of Linked Data while supporting fast, customizable indexes in an application-friendly data format. The methodology makes use of JSON-LD to represent RDF graphs in JSON suitable for indexing with Elasticsearch. BibJSON is used as a common index format capable of handling a wide range of library resources. Since all three technologies (RDF/JSON-LD, BibJSON and Elasticsearch) share an emphasis on extensibility, it is possible to create an index of bibliographic data that is both generalized and flexible enough to handle Linked Data from multiple sources.
Metadata Analysis at the Command-Line
Over the past few years the University of North Texas Libraries’ Digital Projects Unit (DPU) has developed a set of metadata analysis tools, processes, and methodologies aimed at helping to focus limited quality control resources on the areas of the collection where they might have the most benefit. The key to this work lies in its simplicity: records harvested from OAI-PMH-enabled digital repositories are transformed into a format that makes them easily parsable using traditional Unix/Linux-based command-line tools. This article describes the overall methodology, introduces two simple open-source tools developed to help with the aforementioned harvesting and breaking, and provides example commands to demonstrate some common metadata analysis requests. All software tools described in the article are available with an open-source license via the author’s GitHub account.
The Format Registry Problem
File format identification is an important issue in digital preservation. Several noteworthy attempts, including PRONOM, GDFR, and UDFR, have been made at creating a comprehensive repository of format information. The sheer amount of information to cover and the constant introduction of new formats and format versions has limited their success. Alternative approaches, such as Linked Data and offering limited per-format information with identifiers that can be used elsewhere, may lead to greater success.
SPRUCE Mashup London
SPRUCE digital preservation mashups are a series of unique events that are being organized in the United Kingdom to bring together digital preservation practitioners and developers to work on real-world digital preservation challenges. During the 3-day event the digital preservation developers work to create practical solutions to real-world challenges the practitioners are having related to digital preservation. Meanwhile, the practitioners work to create compelling business cases for digital preservation at their institution. This article describes the SPRUCE Mashup London event held in September 2012.