Science Environment for Ecological Knowledge
Ecoinformatics site parent site of Partnership for Biodiversity Informatics site parent site of SEEK - Home
Science Environment for Ecological Knowledge









 

 

 



All Hands Meeting SMS Notes Nov 04

Notes created by Shawn Bowers

Tuesday, Nov. 2, 2004

  • 8:30 - 9:00
    • Jim Beach on today's schedule; politics in Kansas primer
    • Guest from NatureServe (Gwynn ...)
    • Mike Frame from USGS
    • Liz Blood from NSF (not here yet ...)
    • NCSA ...
    • New Folks
      • Laura Downey (unm, usability engineer)
      • Jianting Zhang (postdoc, unm)
      • Steve Teckell (seek web-developer)
      • Mark Serilla (LTER, network information system)
  • 9:00 - 12:30
    • Overviews, etc.
      • Matt Jones
        • Kepler
          • Goal is to: create powerful analytical tools that is useful across disciplines; ecology, biology, engineering, geology, physics, ...
          • Model the way scientists work w/ their data now: mentally coordinate export and import of data among software systems
          • Workflows emphasize data flow
          • metadata-driven data ingestion
          • output generation includes creating appropriate metadata
        • Kepler Today
          • EcoGrid support: EML data support; DarwinCore data support; EcoGrid registry to discover data sources
          • Ontology-based browsing
          • Demonstration workflows
            • Ecological niche models
            • Biodiversity analysis tools
        • EcoGrid queries ...
        • EcoGrid: EML Data Access
        • Right-click data box to get EML metadata display
        • Query builder
          • QBE style query
          • Query execution: EML Metadata + Data -> generate DDL to create temp dbs -> HSQL executes data -> result
        • EcoGrid DarwinCore Access
        • Actor Search and Browse: support multiple views (tuned to ecologist) ... switchable at runtime
        • Kepler for Ecological Modeling
        • Biodiversity indices (from Dan)
        • R in Kepler
        • ROADNet Realtime (Revelle ship)
        • Kepler This Year:
          • User Interface ... need a full eval and reworking of all UI for our target group
            • Semantic annotation interface / ontology browsing interfaces
          • Distributed computing / grid computing
            • large jobs; lots of machines; detaiched execution
          • "Smart" data discovery (as for actors)
            • support for data sources that are distributed
          • Actor repository / downloadable actors
          • Integration and transformation using semantics
            • generate required transformation for: data-actor integration; actor-actor integration
            • handle "semantics" for R scripts, etc.
              • file based buffering reduces our ability to reason about the data flow
              • SMB: not exactly sure where the precise problem is here ...
            • Finish EcoGrid Access
              • full EML support
              • support for "large" data and 3rd party transfer
              • more data sources and types of data sources (e.g., JDBC, GEON data)
            • Provenance and metadata propagation
      • Shawn Bowers (me)
      • Jessie Kennedy
        • SEEK Taxon Scenario
        • Goal is to develop the taxon name server
        • High-level approach
          • extract information from taxonomic concept providers; and convert to taxonomy transfer schema (TML)
          • "Semantic Mediation System" used to do concept matching/expansion/weighted concepts; leveraged from EML
          • Most focus on Taxonomy Transfer schema
        • Names versus concepts problems
          • use concepts instead of names .... decouple concepts from names
          • taxon name: label/word/string
            • used for communicating ideas about organisms or groups of organisms ... meaningless without a definition, but it is used as if it has a clear meaning
            • full scientific name: sci name + author abbreviation +date?
              • implies a concept: as intended and documented by the original author of then ame
            • vernacular name: scrub history
          • What is a concept?
            • taxon concept: unique name plus definition
              • full sci. name + "according to" (author + date)
            • Taxon goals
              • define use case; acceess current taxon working practice; design schema for object server; implement prototype; eval globally unique ids and their applicability
            • investigate interface between SMS and taxon group
            • end user tools
          • progress:
            • use cases finished, but still under review
            • taxon working practice: developed a langauge definition, etc ...
          • version 1.01 ... aimee and robert will demo
          • globally unique ids: UUIDs, handle system, DOIs, LSIDs
          • prototype ...
          • interface with sms: session during the week to discuss
          • case study, anticipating north american mammals example
      • Deana Pennington
        • Original goals
          • case studies in ecology and biodiversity
          • assemble a framework that interconnects ...
        • First project: eco niche modeling
          • mammal project
          • 2-3 dispersal scenarios; 2 major evolutionary-computing algorithms (GA and NN); 21 GCM scenarios, including all IPCC scenarios
        • eco niche modeling
        • 2nd beam project, focusing on biodiversity
          • use biodiversity and productivity field data
          • link with enm project
          • use biodiversity indices
          • expand group of scientists
        • nceas group: Waid, Scheiner, Willig, Cox, Chalcraft; biodiv/productivity scaling rels in semi-arid grasses
        • pdtnet group: Suding, Weiher, Collins, Cleland, Clark; biodiv/productivity functional rels in semi-arid grasses
        • ENM/biodiv integrated analysis
          • remotely sensed imagery -> from Jim Quinn at Davis
      • Samantha Romanello
        • EOT
          • Current activities: on-site trainings; conf. workshops and related activities; new website
          • future activities: on-line modules; distributed graduate seminar
        • on-site trainings
        • new fac. and postdoc workshop
          • how to incorporate into their teaching
          • results of the workshop from last january
      • Raja
        • GOAL: EcoGrid as a standardized interface (using web and grid services)
        • Grid-standardized interfaces: uniform interface to metacat, srb, digir, xanthoria; anyone can implement these interfaces; hides complexity of underlying systems
        • metadata-mediated data access
          • supports multiple metadata standard, EML, Darwin Core as foci
          • computational services: pre-defined analytical services; on-the-fly analytical services
        • all based on gwsdl service
        • metacat and srb services implemented
        • metadata: EML and darwin core
        • authentication
        • query sevice level 1: get/query services; web-based available
        • registry services: implemented in 2 phases -- reimplemented -- list of avail services on web
        • Next steps:
          • design and impl. other ecogrid services
          • extend existing services
          • more ams integration - kepler
          • more sms integration - growl (?)
          • add new services: onto servers, taxon servers, LSID server
          • user interface development
          • documentation and training
          • enable more apps and data ingestions
          • increase user collaborations
      • Laura Downey
        • usability
          • "the extent to which a product can be used by specified users to achieve specified goals with effictiveness, efficiency, and satisfaction in a specified context of users"
          • "an approach that incorporates direct user feedback througout the development cycle to reduce costs and create products and tools that meet user needs."
        • benefits o fusability: increased productivity, sales and revenues, customer satisfaction; decrease traning and support costs, etc.
        • real world example: after move.com completed the redesign of two problem features, users ability to find a home increases from 62 to 98 percent, sales lead generation to real estate agents increased 150%.
        • user interface + functionality = usability
          • elegant or sophisticated internal algorithms don't make any diff inf the interface is bad ...
        • software versus usability engineering: ui code 30 percent, now almost 50-60; maintainence, etc.
          • typical software program is released with 40 usability design problems; over 70% of CIO's state on of biggest problems is communicating with users
        • incorporate usability in seek
          • collect demographic data
          • task analysis (interviews and observations)
          • apply good human factors and design principles
          • iterative design
          • gather user feedback and input: ind. interviews; group discussions; usability evals and testing; categorization exercises; self-reporting; satisfaction surveys
        • general design principles
          • use minimalist approach but provide details on demand
          • apply consisten visual and information design (readability, coding, color, icons, metaphors, data organization)
          • apply appropriate interaction design (tasks, and user's understanding)
          • minimize memory and attention load (memory and processing information limitations) http://www.laoairlines.com
          • speak users' language (user terminology not geek-speak)
        • providing flexibility (shortcuts, RCMs, wizards); provide status and feedback; provide meaningful error and feedback messages
        • upcoming usability activities
          • seek annual meeting: general usability discussion
          • kepler - december: needs dicsussion, observation during training, group executioin of two simple scenarios
          • esa data registry -- february: individual or group usabilitiy testing on registering data
          • issues or ideas on the usability of SEEK tools then contact me: Laura L. Downey

Wednesday, Nov. 3, 2004

Thursday, Nov. 4, 2004

Friday, Nov. 5, 2004



Go to top   Edit this page   More info...   Attach file...
This page last changed on 02-Nov-2004 10:26:25 PST by SDSC.bowers.