[Alex Humphreys]

Recently, JSTOR Labs released our Reimagining the Monograph project. This project (http://labs.jstor.org/monograph) includes a quick video overview, a prototype called Topicgraph (http://labs.jstor.org/topicgraph) that uses topic modeling and data visualization to help researchers better evaluate books before they download them, and a white paper which has now been added to Humanities CORE: https://hcommons.org/deposits/item/hc:10367/.  The JSTOR Labs team and I are eager to share this work with the Humanities community.

We would be thrilled for any and all feedback you might have (the white paper is released as a draft for comment until Jan 31).  I do have a few specific questions I am especially interested in the community’s guidance on:

1. At the core of the white paper are twelve principles for “the reimagined monograph” that emerged from a workshop held at Columbia University Libraries that brought together scholars, publishers and librarians from many disciplines (including a handful of very very smart people who belong to Humanities and MLA Commons).  These principles deal with how to make digital long-form scholarship better for researchers.  Do these principles resonate?   What concepts should be added?
2. The methodology we employed for this project was inspired by the “design thinking” and “lean startup” communities, in which projects are developed very rapidly and incorporating a great deal of user feedback.  (The bulk of this project took two months – from the workshop in October to release in December).  The white paper includes a description of this process — is this process of interest and applicable to the DH / scholarly communities?  Or is it of more specialized interest and so might be better separated into a secondary paper?
3. The Topicgraph prototype makes use of a topic model that’s been derived using the JSTOR Corpus and a controlled vocabulary that describes that corpus.  It also includes a visualization of topic model data that exposes not just the named topics but the terms that make up that topic.  Are there other applications of this method that could be helpful to the DH community?  In addition to the open source front end code and the create-your-own-topicgraph functionality, what else might we provide to make this useful?
4. As the white paper states, the “reimagined monograph” will not be built by a single organization or in a single project.  Topicgraph solves — or purports to solve — just one small use case of many.  Our hope is that more projects will emerge that address other principles of the reimagined monograph.  What should that next project be, and who are the people and organizations that should contribute to it?

I look forward to discussing these and other questions with you all.

-Alex

[Author]

Posts