This month, the group discussed user-contributed metadata. Resources to be consulted in preparation for the March meeting can be found here.
The meeting began with a question: were any of the survey results or recommendations made in the RLG Partners Social Metadata Working Group’s reports, Social Metadata for Libraries, Archives, and Museums surprising? In general, the consensus was no, there wasn’t anything terribly shocking in the reports. A participant asked those present for examples of how they have used social networking sites for LAMs related work. A participant explained how she uses Twitter to participate in professional conferences–an activity that doesn’t require physically attending the conference. Another attendee noted that Twitter use by professionals has trended from back-channel chatter to a tool that is purposefully utilized by conference planners, for instance, broadcasting a hashtag to be used for a conference.
Sheet Music Consortium is investigating social metadata. Exploring different motivations for tagging, commenting, re-posting content, etc. and how those motivations/rewards are different on a site like Facebook are different than those on a LAMs website.
Create communities around collections–need critical mass first. More focused the community, the more successful. This should be something to be considered in selecting collections to experiment with. Successes: Wheaton College doing transcription of mss and diaries in TEI; NYPL doing menu scanning.
Suggestion how LAMs frame an initiative is important (Nat’l Archives admits that all of the fun stuff to transcribe is already transcribed). However! If you build it, they will come?
Successful projects tend to have not just architecture in place but good workflows. Get an open source software package that serves the infrastructure?
IDEA: use touch screens now in Wells Lobby–allow folks to comment on something
Vetting: Does Cherry confirm info for Hoh? No.
Folks understand that comments are evaluative and subjective; maybe LAMs don’t have to worry about unauthorized info appearing on our sites?
Number of comments tends to drive interest… comments can give more context; use drives more use: see that 15 copies are checked out, the message is: this is a good/useful item
Idea that we’re moving away from mediation; moving towards self-service (the crowd describes)
Provide evaluative info: use notes fields to indicate evaluative statements? What do we include in our info systems?
More self-service built into resources!
Cushman feedback went beyond comments and tagging; people were recreating pictures–how do we minimize barriers to user contributions?
Provide tools for transcription: finding aid folder-level scanning of random documents: build an in-interface system for collecting data and spitting it back to researchers in a format they want (this will require research on our part: what do researchers want, what is useful to them)? We can then capture their data.
Info syncing behavior: it takes a lot of effort to keep info current
Time to experiment! Controlled, user-focused experiments such as turning on finding aids comments; work with faculty; outsource work to students. Figure out workflows: vetting, etc.
Presentation on page is important! Key contextual metadata, comments falling towards bottom of page, etc.
Important cultural change is needed: it’s OK to fail, it’s just an experiment, which will yield lots of good information to move forward with
Does Blacklight support comments? Yes (see email to Group). Problem is that Blacklight doesn’t have local storage… IUCAT and DLP Cross Collections Search both powered by Blacklight
Gathering metrics: Digital Humanities Quarterly asks commenters to identify themselves (undergrad, grad, faculty, etc.); such a metric could generate recommendations lists
Question of MPAA ratings in IUL records: AM says that because LAMs are moving towards evaluative statements in our records, perhaps IUL needs to review this policy