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

This is version 21. It is not the current version, and thus it cannot be edited.
[Back to current version]   [Restore this version]


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

Wednesday, Nov. 3, 2004

Thursday, Nov. 4, 2004

Friday, Nov. 5, 2004



Go to top   More info...   Attach file...
This particular version was published on 02-Nov-2004 10:09:26 PST by SDSC.bowers.