EPA Web Work Group 2007

March, 2007

I attended this year’s Environmental Protection Agency Web Work Group conference in Crystal City. As always, there were a few stand-out sessions. This year, the MVP was the Office of Water’s case study on their recent web conversion. You can read more about it near the end of the post.

  • Susan Fagan - Taxonomy

    • faceted
    • BRM is major component
    • some facets may be required
    • planned metadata-driven IA (!)
      • they do have a mockup

  • Susan Fagan - Search Folders

    • training sets - 7 docs to train, 3 to test each topic
    • UI in progress
    • search algorithm done

  • John Shirey - Search

    • 99.7% converted to Northern Light
    • john’s mantra: accurate, timely and complete

  • John Shirey - Web CMS

    • pilots
      • harvesting content and metadata
        • product: Battelle Universal Parsing Agent
      • business processes identified
    • use cases in development by Susan Fagan
    • looking to push to production
    • switching from TSMS to CoreID
    • excellent user interface mockup
    • mentioned Synaptica, API expected
      • cv is ready in staging in Documentum
      • content intelligence services (CIS) demo
    • next steps
      • harvest OSW pages
      • demo content entry process
      • demo integration with MS Word
    • honestly pessimistic about money to do all this

  • Annetta Cheek - Plain Language

    • same presentation as last time…

  • Don Barrett - Accessibility

    • 504 make accommodation for people (wheelchair)
    • 508 make accessible environment (ramp)
    • not a lot of take-aways, but a really excellent talk!

  • Jakita Bass - Strategic Planning for Web - Case Study

    • global warming case study
    • ranked #1 on google’s results for “global warming”
    • but content outdated; not linked in to new content; usability and IA bad
    • content
      • linked to stable ‘concensus documents’
        • esp. those published by cross-org groups
        • not all federal gov’t
        • all already vetted
      • used adminisration-friendly language
      • linked to energy star
    • conceptual framework
      • change name of site
        • global warming
        • climate change
      • feature administration initiatives
      • archiving older material, prominently
      • preference for link to resources rather than text on the site
    • formal review process
      • OAR comms
      • OPA comms and web comms
    • release process
      • memo to linking sites on what’s happening, what’s changed, where to link
      • background doc and Q&A talking points
      • press release via OPA
    • the site
      • site map is the homepage
      • ONE sentence of content on the homepage
      • five sections
      • separate, small, kids site
      • archive of old site
      • * personal emissions calculator
        • ICF did it
        • why epa.gov and not on energy star?
      • they have a cool horizontal tool bar, but there is a LOT of nav

  • John Shirey - Web CMS (part II)

  • Susan Fagan - IA (part II)

  • Susan Fagan - Search Folders

    • some redundancy between browse and folders
    • interested in usability testing
    • near production, but flexible

  • Cy Kidd - OW Web Modernization

    • selling it
      • spent more than six months before launch getting buy in and money
      • senior management approval early
        • process
          • business case to senior mgmt
            • DAA hated their website
          • then sold to programs
            • 24 meetings feb-july
          • then back to snr mgmt
            • memo to staff available in presentation (?)
          • ended with senior execs approving, funding, and sending out a letter to everyone
      • sold change at program level
        • user-centered
        • topical and organizational IA
    • project scope
      • OW web taxonomy
      • OW Information Architecture
      • HTML and CF templates
        • they added their own dynamic components to manage IA centrally
          • navigation elements
          • metadata from database
      • convert content to new template(s)
      • consolidate dynamic pages, apps, and back-end work into a single database
      • content aggregation tool proposal
        • using portal for it
        • can search a topical IA database
        • can drive topical nav, like browse EPA topics, but automated
      • revise SOP for web ops
        • contractor control
        • statement of work boilerplate
        • design principles
        • reduce duplication of data (biz case)
    • lessons learned
      • senior management support
      • alignment with Agency and OW initiatives
        • helps with approval and funding
      • demonstrable improvements
        • cost savings
        • customer satisfaction (user experience)
      • work closely w OEI, esp on IA and taxonomy
        • input into taxonomy dev
        • alignment with taxonomy
    • information architecture
      • structure
        • topical
          • semi-automatic metadata generation based placement in taxonomy
          • pure inheritance
          • predictable useful search
        • organizational, programmatic
          • improved, standardized web navigation
          • aligned with content production
      • development
        • centralized (tried distributed, and it failed)
        • tagging centrally
          • done by a contractor
          • consistent
          • expedited
          • Excel tagging worksheet
            • semi-automatic tagging based on a crawl
            • helps identify ‘mega-pages’ that need to be cut up
        • process
          • originally directory level, and totally IA-based
          • then item-level
          • huge amoung of data generated
            • new synonyms (lots of them)
            • errors (directories with no index.html)
          • draft IA ready for staff review
      • benefits
        • consistent navigation
        • navigation automated by taxonomy structure
        • user experience improved
      • conceptual wireframes
        • quickfinder automated
        • topical nav automated
        • clean design
        • NO PROGRAMMATIC NAVIGATION
          • (wow!)

  • Karen Litwin - Web Stats

    • sources
      • search stats
      • file structure (tssms+directories)
      • web stats
    • services
      • complete inventory of TSSMS area
        • shocking number of non-standard types
          • visio
          • db
          • bak
          • doc
      • apache web log analysis
        • she only gave about half the info available…
        • no mention of cookies
        • talks about PDF request issue
          • they can produce a more accurate report
          • de-aliased
        • 1GB per day of logs at EPA
      • custom reports
        • identification of hosts doing specific things (like data downloads)
        • alias-level reporting
        • no-hit or low-hit reporting
        • file system vs. monthly web stats
          • what’s getting hit
          • what isn’t
      • search logs
        • http://nlquery.epa.gov/webmast/query/querylogreports/
        • EPA search engine only - no external search
        • generated monthly
        • top 100 query terms
        • multiple hits against a result page will get counted
        • there is stemming
      • error reports
        • dead links
        • malformed pages
        • orphans (tricky, given aliases)
    • contacts
      • Mark Fisher
        • fisher.mark@epa.gov
      • Steve Gant
      • Karen Litwin
    • * ask about JS+cookie based tracking
      • legal, but not implemented
      • NO - no current plans
    • * ask about application logs
      • particularly abandonment and failure rates
Filed under Content Management, Information Architecture |

Comments are closed.

Recent Bookmarks

All Bookmarks