Caption photos easily. On the left, a road goes through a tunnel. Middle, leaves artistically fall in a hipster photoshoot. Right, in another hipster photoshoot, a lumberjack grasps a handful of pine needles.

Archives Unleashed 2.0, The Library of Congress

Collaborators:

The group coalesced around a project that was primarily concerned with big, abstract questions surrounding web archives and the ethics of wide-spread practices that surround capturing web information for posterity. We therefore co-designed a collaborative project that examined ‘publicness’ online, accountability and ethics of researching deleted web content by looking explicitly at deleted tweets. Inspired by popular services like PolitWhoops, we wanted to know more about the ephemerality of tweets and the content that is linked from within social media (and by extension, the coverage of this content in web archives aiming to capture the tweets). We extensively debated the ethics of both web archiving practices, but also the validity of designing a study around deleted web content - is it ethical to capture and then investigate tweets that users have subsequently deleted from online? Who decides the boundaries of these investigations? As such we did a pilot study on UK MPs, as a mechanism for discussing ‘publicness’ online by public officials, or people elected to public life that have some reasonable expectation of accountability online.

Descriptive Findings
  • Linking as pervasive practice by UK MPs on Twitter
  • Often linking back to party page, party Twitter account, page supporting agenda
  • ~10% of linked content no longer available (in line with existing research)
  • Can see the page 404 in WayBack
Future Work Potential
  • Look at the capture presence/absence over time (outside the scope of the election season)
  • Investigate (qualitatively) the change in URLs over time - are they substantive changes or reflect errors in crawling/capture behaviour?
  • Who archived the links/captures at the IA - were they crowdsourced? Comparison between British Library UK General Election Archive 2015?
  • How do metrics compare across parties/other thematic identifiers within the election dataset?