Table of Contents
Virtual VuFind® Summit 2021
Dates and Location
Following the success of the prior year's virtual Summit, and due to ongoing global concerns related to the COVID-19 pandemic, the 2021 VuFind® Summit was held as an online-only event, combining the annual German and American VuFind® Summits into a single conference. The event was held online from November 15-17, 2021, with events taking place from 13:00-16:00 UTC/GMT on each day.
Video Recordings, Documents and Social Media
The event was held as a Zoom webinar. Recordings of sessions are available through the VuFind® Summit 2021 YouTube Playlist.
Some planning activities took place through the conference planning document.
Discussions of the event on social media used the hashtag #VuFind2021.
All times are in UTC/GMT.
Day 1 (November 15, 2021)
1:00 - 1:15pm Lobby Time (feel free to join early and test equipment) 1:15 - 1:45pm Demian Katz - VuFind: State of the Project 2021 1:45 - 2:15pm Mario Trojan - IxTheo Authority Search 2:15 - 2:30pm Break 2:30 - 3:00pm Parthasarathi Mukhopadhyay - Automatic Geographic Name Authority from Scratch... 3:00 - 3:30pm Galen Charlton - Updating the VuFind® ILS Driver for Evergreen 3:30 - 4:00pm Demian Katz and Ere Maijala - VuFind® / Koha Integration
Day 2 (November 16, 2021)
1:00 - 1:15pm Lobby Time (feel free to join early and test equipment) 1:15 - 1:45pm Lionel Walter and Christoph Böhm - swisscollections.ch... 1:45 - 2:15pm Mohan Raj Pradhan - Harvesting of Data from DOAJ and OJS in VuFind® with Reference to SDI Service 2:15 - 2:30pm Break 2:30 - 3:00pm M. Jayakananthan - Integration of DSpace 7 with VuFind® Using the OAI-PMH Protocol 3:00 - 3:30pm Dibyendra Hyoju - Indexing DSpace, Koha and Wordpress Website... 3:30 - 4:00pm Parbati Pandey - Use of Scripts for Harvesting and Web-indexing...
Day 3 (November 17, 2021)
1:00 - 1:15pm Lobby Time (feel free to join early and test equipment) 1:15 - 1:45pm Cornelius Amzar, Dorian Merz, André Lahmann, Stefan Winkler - Result Grouping with Match Keys 1:45 - 2:15pm Lightning Talks 2:15 - 2:30pm Break 2:30 - 4:00pm Roadmapping Session
Detailed Talk Descriptions
Demian Katz, Villanova University - VuFind: State of the Project 2021
This talk will discuss developments within the project (both technical and organizational) over the year since the previous VuFind® Summit while also highlighting some possible future developments for the next major release.
Mario Trojan, University Library of Tübingen - IxTheo Authority Search
This presentation will focus on the customization of VuFind®'s authority index and search functionality for IxTheo as well as RelBib and KrimDok.
The presentation will cover topics like differences between normalized / non-normalized authors, Personal relations, Different kinds of title links (by, about, related authors / joint titles) and a Controller plugin mechanism to resolve pictures from Wikidata, as well as an overview of certain VuFind® specialties which we had to deal with.
The presentation will also give a short prospect on the IxTheo Green Road Open Access Publication Service which is about to be implemented next year.
Parthasarathi Mukhopadhyay, University of Kalyani - Automatic Geographic Name Authority from Scratch for Developing Geodetic Search Interface: Where VuFind® Meets Data Carpentry
Geographic name authority datasets when added with latitude/longitude or boundingbox data may lead to map based retrieval systems in VuFind® through geodetic search mechanisms, but unfortunately, global geographic name authority datasets (LCNAF, FAST, VIAF and Wikidata) are not providing geospatial data for place names in general, and also coverage for countries in Asia is not adequate. The geodetic search feature in VuFind® depends on the availability of geocoordinate values in records (tag 034 for MARC records and DC.Coverage.spatial in DCMES) but at the same time hand crafting of geocoordinate data values for thousands of places is an impossible task. This deliberation is a quest for an alternative path to develop MARC based country-specific (here India) authority datasets for geographic names (with geospatial data) from scratch by utilizing an array of linked open geospatial datasets, by applying data wrangling techniques and by deploying open source data carpentry tools. The immediate utilities of such geographic name authority datasets are – a) use of MARC formatted geoname authority datasets in local ILSs; and b) in enhancing retrieval features of VuFind® through developing geodetic search interface by using such authority datasets with geocoordinate data. The mechanisms as proposed here may later be extended to generate corporate name authority, personal name authority etc in large-scale for libraries in countries under the developing block of the world.
Galen Charlton, Equinox Open Library Initiative - Updating the VuFind® ILS Driver for Evergreen
Equinox Open Library Initiative is working on updating the 2009-vintage Evergreen ILS driver to support recent versions of the Evergreen ILS. During this presentation, Galen Charlton, Implementation and at IT Manager at Equinox, will
- Present the updates made to the VuFind® Evergreen ILS driver
- Outline a future direction for the driver to have it use Evergreen's API
- Discuss the implications for the VuFind® driver of Evergreen's approach to managing access to sensitive patron information
- Share our experience as a nonprofit organization that supports open source software for libraries and invite feedback on our contributions as we branch out into VuFind
This presentation will present VuFind® as a front end for Evergreen and is suitable for developers seeking technical information on developing ILS drivers for VuFind® and all interested in learning about using VuFind® with Evergreen.
Demian Katz, Villanova University and Ere Maijala, National Library of Finland - VuFind® / Koha Integration
The originally scheduled presentation in this slot was unable to be held due to the unavailability of the speaker. An impromptu substitute was provided, highlighting resources for learning about VuFind® / Koha integration and covering some of the common challenges of integration.
Lionel Walter, University of Basel Library and Christoph Böhm, Outermedia Company - swisscollections.ch, the Gateway to Historical and Modern Collections in Swiss Libraries and Archives
VuFind-based swisscollections.ch is a new portal developed by the University of Basel Library and the Outermedia company. In this talk, we will highlight and explain some features of this portal. On the one side, there were some specific requirements: hierarchic displays of archives (based on VuFind®'s “Hierarchies and Collections”), specific detailed displays for manuscripts, custom browse indices. On the other side, we will describe the mechanisms we use to deploy VuFind® on productive/staging/test servers. It is based on GitLab, Docker images and a Kubernetes cluster.
Mohan Raj Pradhan, HealthNet Nepal - Harvesting of Data from DOAJ and OJS in VuFind® with Reference to SDI Service
This presentation discusses the installation and harvesting of DOAJ and OJS metadata in VuFind. Also discussed will be a new service under development at HealthNet Nepal: after harvesting is done, data can be sent to users as per their interest periodically through the SDI services of VuFind.
M. Jayakananthan, International Water Management Institute - Integration of DSpace 7 with VuFind® Using the OAI-PMH Protocol
Traditional online catalogues, repositories, and journal databases provide separate interfaces and platforms for users. In contrast, federated search engines and metadata harvesters enable searching from a single interface. This paper documents the success and challenges faced in introducing DSpace 7 with OAI-PMH enabled and integrated with VuFind, an open source resource discovery tool.
Dibyendra Hyoju, Tech-Out Solutions Pvt. Ltd. - Indexing DSpace, Koha and Wordpress Website of Tribhuvan University Central Department of Library Information Science
This presentation discusses the pilot project of Tribhuvan University Central Department of Library Information Science for the implementation of VuFind® to index its digital repository managed in DSpace software, catalogues managed in Koha ILS and the website developed in WordPress and make VuFind® its library resource portal.
Parbati Pandey, Central Department of Library and Information Science,Tribhuvan University - Use of Scripts for Harvesting and Web-indexing of Records from Koha, DSpace, Moodle, and WordPress in VuFind®
To date, the command-line has been used for harvesting records from Koha, DSpace, Moodle, and WordPress. Library professionals who have limited computer know-how find it difficult to use the command-line configuration system for harvesting records into VuFind. So, in this paper, scripting is being used to automate harvesting records from Koha, DSpace Moodle, and WordPress. A usability test is being done among the participants of a basic digital library training course jointly organized by the Central Dept. of Library and Information Science, TU, HealthNet Nepal, and VuFind.
If this usability test is positive, the system will be further improved and tested in other training courses of the digital library being organized in Sri Lanka and Namibia.
Cornelius Amzar, Dorian Merz, André Lahmann and Stefan Winkler (BSZ, Leipzig University Library) - Result Grouping with Match Keys
Deduplication in library catalogues is often considered to be an error-prone, time-consuming and complex procedure. That's why many bibliographic systems don't (fully) implement deduplication and show multiple instances of more or less similar records. This assumption, however, is only true when aiming for a “perfect” solution which works with string distance algorithms on each bibliographic data field, thus resulting in a very complex decision and merge logic.
By using a more pragmatic approach we were able to achieve good deduplication results utilizing the Apache Solr grouping feature applied to match keys based on record title, author, year and format. Records with the same match key are not fully merged but are grouped together in the result list showing only the first matching record, hiding duplicates by default. This approach leads to a significant reduction of the result list while offering the user the same result set size without noticeable drawbacks to search performance.
The development took place in several steps, finally yielding a standalone configurable VuFind® module that can be included by composer VuFind® 8.0 or higher. In this talk we will give a short walkthrough of the preliminaries necessary for indexing and configuration of the module and will show some VuFind® catalogues already using this module in production.