NDIIPP Web-at-Risk Curators' Meeting

Oakland, CA

05.30.2007

• LOC: article about our collection development plans on NDIIPP website
• talk about recent NDIIPP "funding kerfuffle"
• Dec 2007: WAS usable release date
• Jan/June 2008: end-user phase

Next Step

• be promoters of the service
• communicate to our institutions the value of this service

WAS Demo of Release 4 (Tracy Seneca)

• tips @ right of screen (some open to detailed guides)
• host site + linked pages
  • grabs external *pages*, not entire site
  • Question: does it grad the images/elements/CSS that make up that page, as well?
• descriptive data *will* include auto-captured data (examples in later release)
• "delete: doesn't delete captured content; it only deletes the capture action for that site
  • hence, the option to limit the site list under "Manage Sites" to "inactive sites"
• "view job history" = capture dates and various results
• "View Results" = view successful captures
  • will display most recent job at top
• must choose capture date; will make site title non-clickable to avoid confusion
• timestamps currently in PST (working to change this)
• results screen (first time) will always have some lag (alot of data to load), but will load quickly thereafter (cached)
• "start/finish time" slightly different than "duration:
  • s/f includes time spent in queue, etc.
  • duration only counts active crawler time
• Robots Exclusion
  • Heritrix will capture the *rules* even if it can't capture the site (there is help to decipher these files)
  • common exclusion = CSS files
• "response code" reports
  • may be more meaningful for long-term captures (health of a site)
• "Hosts" report
  • lists subdomains
  • lists all hosts
• will also be developing FAQ
• will provide search tips
  • abstract = auto-generated (portion of site included; search terms?)
  • full-text search capability
  • can add individual documents to a collection from the search tab
• display top bar will show navigation to various (dated) versions
• can add *comment* to user-defined metadata (for curator view)
• will add sorting to file list under "results"
• included/excluded under "collection"
  • good way to exclude a few non-public domain files from a site
• time delay between collection building and viewing/searching contents
  • currently trying to predict this amount of time
  • or may appear to be a mismatch between "contents" and "files"

note: CC: way to manage list of captures

Usability Testing (Kathleen Murray)