CollectionBuilder is a lightweight, flexible tool for creating digital collection websites that are driven by metadata and powered by modern static web technology. Using three primary components—a spreadsheet of metadata, a directory of assets, and a configuration file—CollectionBuilder gives information professionals and digital scholarship practitioners agency in building and customizing digital collection websites for free.
CollectionBuilder is designed and maintained by librarians at the University of Idaho according to the Lib-STATIC approach, a methodology committed to leveraging static-web technologies and librarians’ specialized skills in metadata and classification to create engaging web publications.
In 2019, CollectionBuilder’s development team received an IMLS-sponsored National Leadership Grants for Libraries Planning Grant to refine and promote CollectionBuilder.
Learn More About:
The Tool - CollectionBuilder
Over the course of our years designing digital collection interfaces, it has become apparent that the metadata itself is the most important aspect of a digital collection. As such, we (a team of librarians at the University of Idaho) have built a tool that puts the metadata forward, and uses it to build visualizations and browsing features.
CollectionBuilder uses static web technologies to build digital collection and exhibit pages from metadata. The tool generates item pages and all its visualizations and browsing features from a metadata spreadsheet and config and theme files.
The tool uses 4 main components:
- Jekyll - a static website builder (i.e. NO SERVERS!!!) that builds websites from data files using the Liquid templating language (which you don’t need to know, but which is easily readable and usable) and Markdown files for content
- Git/GitHub - a way to collaborate, track changes, and import pre-built services
- Data Files - We use Comma Separated Values (CSV) files that are just simple version of a spreadsheet and text files written in YAML (.yml), which are basically lists formatted in a specific fashion.
CollectionBuilder builds digital collection sites that encourage investigation and discovery. Our item and visualization pages are built with connections between them so that a user can follow their own interests when exploring a digital collection.
For instance, if you were interested in seeing all the images in our Barnard Stockbridge Photography Collection pertaining to the 1910 “Big Burn” forest fire, you could browse the collection by searching for “fire” on the browse page, then, when looking at individual item pages, you could link out to the year 1910 on the timeline page and see the wide variety of images collected during that year by the studio.
And perhaps that would lead you to investigate some other subject terms or to focus on a particular town via the map. The intention is to connect users and to uncover the accretive magic of cultural heritage collections.
As one examines these collections more closely, we feel that one can get a larger sense of the collection and the era, as well as its connection to one’s own area and time frame.
We call the methodology we use to build CollectionBuilder Lib-STATIC. We’ve created a little website to document the ideas that inform this approach.
Basically, we believe that the systems libraries have been locked into, particularly concerning digital collections, serve neither the librarians/GLAM professionals that maintain them and prepare their content nor the collections we are trying to promote using them.
CollectionBuilder prioritizes pragmatic, sustainable, and simplified approaches to infrastructure to ensure the tool is ‘do-able’ and approachable for digital knowledge workers in libraries and museums, empowering them to take control of their web systems.
How It Works for Us
We use CollectionBuilder to build our digital collections on top of our CONTENTdm instance. To do this, we use the tool together with a Git-Based Workflow. Within our own CollectionBuilder-CONTENTdm repository, there is a “main” branch of CollectionBuilder from which we create specific branches to build our individual collections. The trajectory looks like this:
- Create a new branch for a collection from our main CollectionBuilder-contentdm repository
- Export collection metadata from CONTENTdm as a .tsv file
- Import the file into a Google Sheet (or other non-excel spreadsheet program)
- Add a metadata field for “objectid” based on CONTENTdm item number and collection name
- Save as a CSV file
- Move the file into the “_data” directory of our branch
- Adjust the theme and config files
- Test and view the collection on Jekyll-generated local development server
- When finished, build the site (using command that also adds analytics scripts)
- Move the finished files to our web server
To see some of what we’ve done, go to https://www.lib.uidaho.edu/digital/.
Most of our collections have been built using this method in the past six months.
We are currently being funded by the Institute for Museum and Library Services (IMLS) to build, document, and promote CollectionBuilder. We hope in doing so to lay out a model for building other tools in a similar manner.
This year’s grant will let us fine-tune CollectionBuilder through our own development and through the hiring of a developer. It will also help us to develop the type of promotional materials, documentation, and other products that should help us spread the word about the tool and make it easy to adopt and use.
The grant also provides funds for us to embed ourselves into several academic libraries and/or similar institutions to set up the technological tools and basic workflows necessary for using the tool to build digital collections more pervasively.
Opportunities for Collaboration:
- Partner! - If you, or some institution you know, might be interested in adopting this tool more seriously, we can help. Please get in touch for current opportunities and help us guide the future of CollectionBuilder.
- Feedback! - If you have thoughts on the documentation, tool, visualizations, user interactions, or any other features in the tool, please please please provide us with feedback. Our intention in doing workshops and presentations is to gather feedback on the tool so that CollectionBuilder can continue to grow, becoming more useable and useful.
- Contribute - If you take to the static web development model and would like to provide contributions of your own to our tool, we would LOVE to work with you and incorporate any contributions you might make. We have a code of conduct for contributors.
Interested in becoming a collaborator? We’d love to talk to you more about it! Contact Olivia Wikle for more information.
Our highest hope for this project is to enable a small army of librarians to develop the type of tools and sites that keep the GLAM professionals in control and not subservient to bloated infrastructures and/or third-party contracts. We hope you will feel drawn to help us build that community.
CollectionBuilder in the Wild
Select digital collections from users outside of University of Idaho.
- She Changed the World, North Carolina Digital Collections
- NCF Special Formats Collection, New College of Florida
- Founding Stories, University of Washington Tacoma
- Annotated Alvin Josephy, Josephy Library Collections
- Colors of Ozu, Dave Rodriguez (Florida State University)
- Beaverhead Mine Documents, Montana Bureau of Mines and Geology
- W.K. Kellogg Arabian Horse Library Postcard and Notecard Collection, California State Polytechnic University, Pomona
- Namibia Heritage Week 2020, Namibia University of Science and Technology
- Workshop: Devin Becker, Evan Williamson, and Olivia Wikle, “CollectionBuilder: A Static Web Approach to Digital Projects in Libraries”, ALA Core Classroom, 2020, https://osf.io/ku67f/
- Article: Devin Becker, Evan Williamson, and Olivia Wikle, “CollectionBuilder-CONTENTdm: Developing a Static Web ‘Skin’ for CONTENTdm-based Digital Collections”, Code4Lib Journal 49, 2020, https://journal.code4lib.org/articles/15326
- Article: Olivia Wikle, Evan Williamson, and Devin Becker, “What is Static Web and What’s it Doing in the Digital Humanities Classroom?”, dh+lib, Special Issue: Literacies in a Digital Humanities Context, 2020, https://dhandlib.org/2020/06/22/what-is-static-web-and-whats-it-doing-in-the-digital-humanities-classroom/
- Presentation: Olivia Wikle, Evan Williamson, and Devin Becker, “CollectionBuilder: Developing an Agile Approach to Digital Collections”, Info2Go!, Idaho Commission for Libraries, 2020, https://youtu.be/PhXhY5G--p4
- Workshop: Olivia Wikle, Evan Williamson, and Devin Becker, “CollectionBuilder: Using Static Web Technologies to Create Digital Collections and Teach DH Principles”, ADHO DH2020, 2020, https://collectionbuilder.github.io/dh2020/
- Workshop: Evan Williamson, Olivia Wikle, and Devin Becker, “A Static Web Approach to Digital Collections: Hands-On with CollectionBuilder”, Code4Lib, 2020
- Presentation: Olivia Wikle, Evan Williamson, and Devin Becker, “Metadata + APIs → Beauty: Building Engaging Digital Repository Skins With Static Web Tools”, Digital Library Federation Forum, 2019, https://osf.io/hteq3/
- Workshop: Olivia Wikle, Evan Williamson, and Devin Becker, “Digital Collection Magic with Static Web Technologies: Improving CONTENTdm Front Ends using CollectionBuilder”, Learn@DLF, 2019
- Presentation: Olivia Wikle and Evan Williamson, “CollectionBuilder: Creating & Teaching Digital Collections with Minimal Infrastructure”, ASIS&T Webinar Series, 2019, https://osf.io/pfhcy/
- Presentation: Evan Williamson and Devin Becker, “GitHub-Pages CollectionBuilder: Teaching & Building Digital Library Collections using Minimal Computing Techniques”, DLF Forum, 2018, https://osf.io/d8e6b/
- Olivia Wikle, Digital Initiatives Librarian, Communications/Assessment Director for Grant
- Evan Williamson, Digital Infrastructure Librarian, Technical Director
- Devin Becker, Head of Data/Digital Services, Project Director
- Jylisa Doney, Social Sciences Librarian, Documentation Director
- Chelsea Codling, Graduate Assistant
- Michael Decker, Graduate Assistant