Cobweb�Collaborative Collection Development for Web Archives�#cobwebarchive

Stephen Abrams

Kathryn Stine

California Digital Library

Peter Broadwell

Andrew Wallace

UCLA

Janet Taylor

Ann Whiteside

Harvard University

Dodging the Memory Hole, Internet Archive, 15-16 November 2017

Imagine …

Flint, 2016

… a fast-moving event unfolding online as much as on the ground

Carlos Osorio/AP © 2016

How can we respond as a community appropriately and responsibly?

Imagine …

Harnessing the domain knowledge and technical capabilities of the entire community

Enabling local collection development decisions based on global information

Complementary, cooperative, and collaborative collecting

Institutional participation at a level commensurate with local expertise and capacity

Increasing scholarly and public awareness

​

Cobweb

Centralized catalog of collection- and seed-level metadata

Establishment of thematic collecting projects

Open nomination of topical seed URLs by interested stakeholders

Claiming of seeds by archival institutions intending to harvest

Holdings records for seeds actually harvested

Thematic discovery of web archives of interest

Why Cobweb?

The demands of archiving the web in comprehensive breadth or thematic depth exceeds the technical and financial capacity of any one institution

Curators cannot make rational collection development decisions without knowledge of what others have collected or intent to collect

Relevant seed URLs can be meaningfully contributed by various stakeholders: curators, archivists, subject area specialists, scholars, journalists, event participants, and the public

Apportioning collection responsibility into granular pieces encourages participation by smaller institutions and programs

Why Cobweb?

Peter Broadwell at UCLA was well into collecting “fake” news sites before it occurred to him to wonder if anyone else was doing something similar

There was; Mark Graham at IA

Cobweb project

One-year collaborative project between CDL, Harvard University, and UCLA, funded by IMLS #LG-70-16-0093-16

Public online service hosted at CDL

Python/Django stack

MIT license

Targeting initial production release in conjunction with the November 2018 IIPC General Assembly and Web Archiving Conference

https://cdlib.org/cobweb

https://github.com/CobwebOrg/cobweb

​

Demo

Next steps

​

​

Cobweb is a tool for collecting communities…

we need you!

Next steps

​

​

• Keep learning about your collaborative collecting projects and workflows
• Refining and validating use cases
• Engaging testers to walk through prototypes (early 2018)
• Iterative development, responding to user testing findings

​

Q & A

​

​

Questions for us?

​

Questions from us!

Cobweb

California Digital Library https://www.cdlib.org/

Harvard University Library http:// library.harvard.edu/

UCLA Libraries http://www.library.ucla.edu/

​

https://cdlib.org/cobweb

https://github.com/CobwebOrg/cobweb

Kathryn Stine, Cobweb Outreach Manager

Kathryn.Stine@ucop.edu

​

Collaborative Collection Development for Web Archives

#cobwebarchive