MIT Libraries logo MIT Libraries

MIT Logo Search Contact

Experiments graveyard

Wondering what happened with an experiment? This page lists previous experiments that we decided to end and why.

Text Data Mining (TDM)/Fedora – MIT Theses

Fedora is an open source repository system especially suited for digital libraries and archives, both for access and preservation. It can provide specialized access to very large and complex digital collections of historic and cultural materials as well as scientific data. The MIT Libraries are experimenting with this platform to gain knowledge around text data mining (using the digitized MIT thesis collection in MIT’s DSpace) and Fedora4.

Why this experiment?

Text and data mining are important areas to understand in order to offer supporting services to our research community. Currently, this is not a standard feature within institutional repositories, which leaves publishers as the primary entities offering these services. While much was learned during this 4-month experiment that will speed future work in this area, there are existing technical challenges and a lack of demand so it is not a key priority for the Libraries to continue with at this time. If you are interested in seeing the API and other technical details, visit the project’s technical repository.

Experiment runtime: Fall 2016 – Fall 2017


preview of Yewno
Yewno is a new type of discovery tool developed in Silicon Valley, which uses full text analysis and machine learning to create a visual, interactive map of connected concepts. The MIT Libraries participated in a beta project along with several other institutions, including Harvard, Stanford, Oxford, Stonehill College, Bavarian State University, the University of California-Berkeley, and the University of Michigan.

Why this experiment?

The goals of this experiment were to encourage experimentation and creative thinking in the discovery space, to evaluate the Yewno tool for possible inclusion in our discovery environment, and to serve as a template for future “beta” projects within the Libraries. After a 9 month beta period, we determined that while the Libraries are still interested and engaged in the tool that Yewno is developing, the timing to incorporate it into our discovery environment is not right. MIT Press will continue with Yewno Unearth; the Libraries will follow Yewno’s development and may include the tool in the future.

Experiment runtime: 2016

TryIIIF, an IIIF viewer and generator

preview of TryIIIF
The International Image Interoperability Framework (IIIF) aims to define a uniform method to access image-based resources, which are hosted at institutions around the world. IIIF wants  to develop shared technologies that allow scholars to view, compare, manipulate and annotate images. TryIIIF demonstrated two IIIF viewers and the JSON configuration file.

Why this experiment?

As a contributor to the IIIF open source effort, the MIT Libraries developed TryIIIF to demonstrate the power of IIIF, the viewer options, and the code needed to create an image view.

Experiment runtime: 2016


preview of BrowZine
BrowZine is a tool for reading e-journal issues on tablets, phones or on the web. It consolidates journals from several publishers, allowing you to read and browse issues through a single interface rather than looking up each journal individually.

Why this experiment?

The MIT Libraries tried BrowZine to explore ways to make online browsing more visually interesting and personalized. After a year-long trial, we decided not to renew our BrowZine subscription due to low adoption rate among our user community.

Experiment runtime: 2015 – 2016