Technology Infusion Working Group

1
Technology Infusion Working Group

• Karen Moe, NASA/ESTO
• Rob Raskin, NASA/JPL
• Earth Science Data Systems Working Group Meeting
• College Park, MD
• November 14 - 16, 2006

2
Agenda

• Mission Scope
• Activities Accomplishments
• Process Strategies
• Web Services
• Semantic Web
• Capability Vision
• Breakout Session Agenda
• Technology Showcase
• Web Services and Semantic Web Demonstrations

3
Tech Infusion Working Group

• Mission
• Enable NASAs Earth Science community to reach
  its research, application, and education goals
  more quickly and cost effectively through
  widespread adoption of key emerging information
  technologies
• Scope
• Information technologies that...
• Provide capabilities critical to the ESD mission
  vision
• Have been substantially developed (TRL6-9) but
  have not been widely deployed
• Cannot be obtained simply through reuse of mature
  subsystems or software
• May be slow to be adopted because of the unique
  characteristics of Earth science (e.g., high data
  volumes)
• Approach
• Improve community understanding of the technology
  infusion process
• Identify barriers and solutions to technology
  adoption
• Use case studies to evaluate effectiveness of
  infusion processes
• Identify and evaluate new and emerging
  technologies
• Develop roadmaps for adoption of key technologies

4
TIWG 2006 Activities

• Maintained 3 active subgroups
• Infusion Process and Strategies
• Subgroup lead Steve Olding
• Web Services
• Subgroup lead Ken Keiser (UAH)
• Semantic Web
• Subgroup leads Rob Raskin (JPL) and Peter Fox
  (NCAR)
• Conducted weekly telecons
• 1st Thursday Full working group
• 2nd Thursday Process and Strategies
• 3rd Thursday Web Services
• 4th Thursday Semantic Web
• Presented posters at January and July ESIP
  Federation meetings, Geoinformatics 2006, AGU
  Joint Assembly, and ESTC 2006
• Held TIWG breakout session at July ESIP
  Federation meeting

5
Subgroup Activities

• Process and Strategies
• Completed infusion process document
• Conducted infusion readiness case study (REACT)
• Developed infusion readiness assessment
• Web Services
• Reviewed OGC Web Coverage Service
• Developed web services demonstrations in
  collaboration with ESIP Federation Web Services
  Cluster
• Presented web services demos at the July ESIP
  Federation meeting
• Semantic Web
• Reviewed current maturity of semantic web
  technologies
• Developed semantic web roadmap
• Capability Vision
• Continued dissemination of the capability vision
• Presented the vision at Geoinformatics 2006, AGU
  Joint Assembly, ESTC 2006, and July ESIP
  Federation meetings

6
Technology InfusionProcess and Strategies
Sub-Group
7
Technology Infusion Process Overview

• Summarizes 2005 findings
• 3 perspectives on technology infusion
• The innovation development process
• The innovation-decision process
• TIWG infusion process
• Conclusion and Recommendations
• Infusion Education
• New Technology Awareness
• Technical Support
• Funding for Infusion
• Technology Assessment Process

8
Innovation Decision Process Model
- the means by which a message gets from the
source to the receiver -different channels may
play different roles in creating awareness and
persuading individuals to change their attitude
towards the innovation
Previous practice Felt needs / problems Innovative
ness Norms of the social systems
Prior Conditions
Knowledge
Decision
Persuasion
Confirmation
Implementation
Continued Adoption
Characteristics of the 'Decision-Making Unit'
Adoption
Perceived Characteristics of the Innovation
Socioeconomic characteristics Personality
variables Communication behavior
Later Adoption
Relative advantage Compatibility Complexity Triala
bility Observability
Discontinuance
Continued Rejection
Rejection
Adapted from Rogers (1995),Diffusion of
Innovations
9
REACT Technology Infusion Case Study

• Rapid Environmental Assessment Composition Tools
  (REACT)
• Client application - part of the decision support
  tools being implemented by ACT under a
  cooperative agreement with NASA
• Provides network centric services
  (client-to-server and peer-to-peer) for reading,
  reprojecting, and subsetting geo-spatial
  satellite imagery

• Infusion case study
• Conducted infusion readiness case study using
  REACT as an example of a new technology preparing
  for infusion
• Evaluated infusion readiness using the infusion
  decision process model
• Developed a self-assessment questionnaire
• Questions designed to elicit preparedness of the
  technology for infusion
• Based on infusion decision process model

REACT Schedule and Key Deliverables Year 1
Application Component a) Req. Gathering, b)
Install high end HW/SW to ingest/process/serve
data, c) enhance current Fusion server
Decision Support Tool Technology. Year 2
Application Component a) Demonstrate widely
available internet access to all data levels, b)
develop initial Hazardous Algal Bloom product
based on satellite/model data in GIS and other
compatible formats, c) add new data streams (e.g.
NPP VIIRS) Year 3-5 Application Component a)
Expand to other application areas, geographic
areas, and data sources, b) identify and
customize architecture to provide additional
decision support tools, assist in transition of
application to operation.
10
Technology Infusion Readiness Assessment

• Questions
• Target users and target user characteristics
• Identifying the prospective users.
• Innovation characteristics
• Factors affecting prospective users perceptions
  of the technology.
• Financial
• What are the financial implications of adopting
  the technology for the prospective user?
• Productization
• How much effort is required to take a technology
  that is proven in the lab and turn it into a
  'product'
• Communication channels
• How will we communicate with the different
  prospective users?
• Characteristics of the innovation decision
• How will the decision to adopt the innovation be
  made?

11
Technology InfusionWeb Services Sub-Group
12
Web Service Chaining Demo

• Goal
• Demonstrate collaborative service chains by
  connecting together services from many ESIP
  Federation members
• Example Scenario
• Analyze an Air Quality event
• Demo Participants
• GENESIS SciFlo Workflows Brian Wilson
• DataFed Rudy Husar
• BPELPower Workflows Liping Di, Peisheng Zhao
• WIPE/REACT engine Eric Malaret
• OpenDAP / ECHO Peter Cornillion Michael
  Burnett

13
Web Service Chaining Demo

• Success Story
• Developed as a collaboration between TIWG and
  ESIP Federation Web Services Cluster
• Activity organized using the Federation Wiki, WS
  Cluster telecons, and TIWG telecons (participants
  are geographically dispersed)
• Nine groups published callable services on the
  Wiki
• Six groups participated in service choreography
  demos
• Demonstrated service chains for data query
  access, visualization, image overlay, and simple
  data fusion
• Demonstrated interoperability between different
  types of services OGC WMS/WCS, SOAP, OpenDAP
• A service chain can combine all three
• Identical Air Quality analysis chains executed by
  both BPELPower SciFlo workflow engines
• Two different environments for authoring XML
  workflow documents

Carbon Cycle
14
Geographically DistributedParticipants Services
Join us at http//wiki.esipfed.org/index.php/Avai
lable_Services
Carbon Cycle
GMU LAITS WMS/WCS, Reprojection, Reformatting
services BPELPower workflow
DataFed WCS, Render, Aggregate services
WIPE from ACT Query, coverage, mosaics MSHELL
analysis pipelines
GENESIS SciFlo GeoRegionQuery services SciFlo
workflow engine
UAH WMS, HDFEOS Subsetting ADaM Data Mining
SURA SCOOP SOAP Catalog Inventory services for
ocean data
15
Air Quality Scenario
AIRNOW.pmfine WCS
Executable by BPELPower or SciFlo workflow
engines
Carbon Cycle
Image
Render SOAP
SURF_MET.temp WCS
Fused Image
Overlay Images SOAP
Image
Render SOAP
MODIS Image via WMS

• Dataflow Combines
• 5 remote services
• OGC WCS calls (2)
• OGC WMS call
• SOAP services to
• render overlay
• images

16
Lessons Learned

• Challenges / Lessons Learned
• Daunting semantic variation in service interfaces
• Many ways to describe a lat/lon bounding box,
  etc.
• We leveraged OGC WMS/WCS popularity by creating
  SOAP services with similar semantics
• Ultimately need standards, taxonomies, semantic
  mediation
• Simple, modular XML standards (Microformats) are
  best
• GML is too large complex for easy use
• Simple, flat service interfaces are more reusable
• Complex, deeply hierarchical XML input/output
  docs. are difficult to generate correctly
  extract information from
• Should hide complexity whenever possible
• Demonstrated service chains were simple, but a
  good start
• More complex analysis data fusion scenarios
  will be challenging

Carbon Cycle
17
Web Service Chaining

• On-Going Work
• Add domain scenarios from all Federation
  Application Clusters
• Science Analysis
• Real-Time Decision Support
• Publish call more kinds of services
• Subsetting, Aggregation, Geo-Transformations,
  Re-projection, Data Fusion, Custom Analysis, Data
  Mining, Decision Support
• Develop exploit standard XML Microformats
• De facto geo-location standards from Google
  Earth, Microsoft Virtual Earth, GIS world and
  science domain standards
• Simple microformats agreed upon on-the-fly during
  collaborations
• Join the growing Chain Gang, please . . .
• Publish your services on the Wiki (we will call
  them)
• Develop or participate collaboration scenarios

Carbon Cycle
18
Web Services Roadmap

• On-Going Work
• Update Web Services Roadmap to reflect experience
  with the demonstrations

Results
? Accelerated Research System Cost Savings
? Increased Collaboration Interdisciplinary
Science
? Increased PI Participation in Information
Production
? Increased Data Utilization
? Improved Information Sharing
Outcome
? Open geospatial services proliferate
? Production quality geospatial services
? Intelligent Services
? Geospatial services established
Output
Capability
? Parameter-based product searches and access
? Automatic service mediation
? Semantic geospatial search access
? Full geospatial logical searches and access
Assisted Discovery Mediation
? Local processing data exchange
? Basic data tailoring services (data as service)

• Interoperable geospatial services(analysis as
  service)

? Metadata-driven data fusion (semantic service
chaining)
Interoperable Information Infrastructure
Technology
? Geospatial service catalog established (WSDL,
UDDI)
? Common geospatial schema adopted (GML, ESML)
? Open geospatial ontology converges (OWL)
? Open data access established (OpenDAP, OGC)
Data
? Standard workflow language infused (BPEL)
? Common service protocol, description adopted
(SOAP, WSDL)
? Open service protocols established (HTTP, REST)
? Unified security identity management
(WS-Security, SAML)
Messaging
Current Near Term
Mid Term Long Term
19
Technology InfusionSemantic Web Sub-Group
20
Semantic Web Activities

• Reviewed current maturity of tools for the
  Semantic Web
• Ontology languages RDF, OWL (OWL-Lite, OWL-DL,
  OWL-Full)
• Ontology editing and visualization Protege,
  SWOOP, Medius, Cerebra Construct, SweDE
• Reasoners Pellet, Racer, Cerebra Server, Medius
  Knowledge Brokering Suite.
• Knowledge Inference Thetus
• Web Services OWL-S
• Ontology storage KOWARI, Sesame
• Search Tools SWOOGLE
• Ontologies SWEET. Upper level ontology for Earth
  system science.

• Updated Capability Vision to reflect semantic
  technologies
• Developed Semantic Web Roadmap
• Technologies
• Capabilities
• Results

21
Semantic Web Roadmap
Results
? Improved Information Sharing
? Increased Collaboration Interdisciplinary
Science
? Acceleration of Knowledge Production
? Revolutionizing how science is done
Outcome
? Geospatial semantic services established
? Geospatial semantic services proliferate
? Scientific semantic assisted services
? Autonomous scientific inference
Output
Capability
? Common vocabulary based product search and
access
? Semantic agent-based integration
? Semantic agent-based searches
? Semantic geospatial search inference, access
Assisted Discovery Mediation
? Local processing data exchange
? Basic data tailoring services (data as
service), verification/ validation

• Interoperable geospatial services(analysis as
  service), explanation

? Metadata-driven data fusion (semantic service
chaining), trust
Interoperable Information Infrastructure
Technology
? SWEET 3.0 with semantic callable interfaces via
standard programming languages
? SWEET core 2.0 based on best practices decided
from community
? Reasoners able to utilize SWEET 4.0
? SWEET core 1.0 based on GCMD/CF
Vocabulary
? Scientific reasoning
? Geospatial reasoning, OWL-Time
? RDF, OWL, OWL-S
? Numerical reasoning
Languages/ Reasoning
Current Near
Term Mid Term Long Term
22
Semantic Web Roadmap Details
Competing catalog schemas
Common semantic service catalog established
Enhanced semantic search into search engines
Automatic knowledge discovery and mining
Discovery
Semantic service chaining, SWSL
Intelligent algorithm programming chaining
Semantic framework for Web Services, WSMO
Standard workflow language (BPEL)
Workflow
Built into code logic and in the head of the user
Basic semantics (DL, FOL)
High degree of semantic understanding
Intelligent message routing (SOL)
Inference
SWEET Core 1.0 VSTO, MMI, others
SWEET core 2.0 domain and math plug-in
SWEET 3.0 science applications plug-in
Earth Science Standards
GCMD, CF, ESML, GML, etc.
Languages
PML
XML, RDF
OWL-DL, OWL-Full WSML
OWL-S, SWRL
Current Near Term Mid
Term Long Term
Proof/Trust
Syntax
Explanation/Rules
Semantics
23
Technology InfusionCapability Vision
24
Capability Vision Outreach

• Continue to raise awareness of the Capability
  Vision
• Improve online availability
• PowerPoint and PDF versions
• New Flash version created
• Rudy Husar and Erin Robinson
• NASA web sites
• DSWG, ESTO, Tech Infusion

25
Capab
Capability Vision Posters Geoinformatics 2006 AGU
Joint Assembly ESTC 2006 ESIP Federation summer
meeting
26
Extending the Vision

• Leverage the Capability Vision to stimulate
  interest within the Earth science community in
  new and emerging technologies
• Provide assessment of current state of
  technologies to support the top 10 capabilities
• Identify current research and available
  technologies
• Assess technology maturity
• Identify gaps and opportunities
• Verify alignment with similar technology vision
  activities (e.g. EOSDIS Evolution)

TIWG Capability Vision
EOSDIS Evolution
27
Technology Infusion Breakout Sessions
28
Breakout Session Agenda
Review web services demo lesson learned,
recommendations etc. Updating the roadmap
Extending the capability vision. Assessment of
technology maturity.
What are the hot new and emerging technologies?
How do we identify and monitor them?
Review of the Semantic Web roadmap.
29
Technology Showcase

• Technology demonstrations on Wednesday afternoon
• 230 330 Web services chaining choreography
  update
• Brian Wilson, Rudy Husar, Liping Di, Eric Malaret
• 330 315 Web service based data mining
• Deployment orchestration of automated workflows
  at DAAC. Ken Keiser Chris Lynnes
• 400 500 Semantic Web
• Ontology-supported knowledge discovery in
  geospatial semantic web. Liping Di.
• NOESIS an ontology-based semantic search tool
  resource aggregator. Rahul Ramachandran Ken
  Keiser.
• Ontology collaboration web site. Peter Fox Rob
  Raskin.