Difference between revisions of "NASA ACCESS09: Tools and Methods for Finding and Accessing Air Quality Data"
From Earth Science Information Partners (ESIP)
| Line 26: | Line 26: | ||
==Approach== | ==Approach== | ||
| + | |||
| + | * Service Orientation, while accepted has not been widely adapted for serving NASA products | ||
| + | * SOA allows the creation of loosely coupled, agile, data systems | ||
| + | * SOA -> requires ability to Publish, Find, Bind (Register, Discover, Access) | ||
| + | |||
=== Data Service Publication === | === Data Service Publication === | ||
| + | * Data as Service | ||
| + | ** Wrappers, reusable tools and methods ( dataset classes) | ||
| + | ** To illustrate the Network, Coding (faceting) through metadata, WMS to show data | ||
| + | ** WCS next | ||
| + | * Binding to data through standard data access protocols, publishing and finding requires metadata system | ||
| + | |||
| + | |||
=== Data Discovery === | === Data Discovery === | ||
| + | * Metadata system for publishing and finding content has to be jointly developed between data providers and users. | ||
| + | * Generic catalog systems - metadata collection of not only what provider has done but also tracking what users need | ||
| + | ** Collecting and Enhancing Metadata from observing Users | ||
| + | * Communication along the value chain, in both direction; | ||
| + | * Metadata the glue and the message | ||
| + | * Market approach; many providers; many users; may products | ||
| + | |||
| + | * Faceted search | ||
| + | ** user is happy | ||
| + | * Search by usage data | ||
| + | |||
Description on how users will discover and use services provided by NASA, other Agencies, academia.. | Description on how users will discover and use services provided by NASA, other Agencies, academia.. | ||
* Detail on discovery services | * Detail on discovery services | ||
Revision as of 09:14, May 31, 2009
Air Quality Cluster > AQIP Main Page > Proposal | NASA ACCESS Solicitation | Context | Resources | Forum | Participants
Outline
Background
Problem
- Lots of data resources in NASA, elsewhere
- Not available as data as services
- if it is a service, e.g. openDaP, may need rich clients, hard coded, loose coupling not easy
- Even if reusable service is available it cannot be found.
- The Users Dilemma (direct problem)
- There are no data for what the user needs
- If there are needed data, the user can not find them
- If the user can find them, she can not access them
- If the user can access then, she does not know how good they are
- If he knows how good they are, she can not merge them with other data.
- Providers Dilemma (indirect problem)
- There are no users for the data
- If there are users, the provider can not find them
- If she can find them, she does not know how to deliver the data
- If she can deliver them, she does not know how to make it valuable
- If the she can make it more valuable...
Approach
- Service Orientation, while accepted has not been widely adapted for serving NASA products
- SOA allows the creation of loosely coupled, agile, data systems
- SOA -> requires ability to Publish, Find, Bind (Register, Discover, Access)
Data Service Publication
- Data as Service
- Wrappers, reusable tools and methods ( dataset classes)
- To illustrate the Network, Coding (faceting) through metadata, WMS to show data
- WCS next
- Binding to data through standard data access protocols, publishing and finding requires metadata system
Data Discovery
- Metadata system for publishing and finding content has to be jointly developed between data providers and users.
- Generic catalog systems - metadata collection of not only what provider has done but also tracking what users need
- Collecting and Enhancing Metadata from observing Users
- Communication along the value chain, in both direction;
- Metadata the glue and the message
- Market approach; many providers; many users; may products
- Faceted search
- user is happy
- Search by usage data
Description on how users will discover and use services provided by NASA, other Agencies, academia..
- Detail on discovery services
- System components for persistent availability of these services
- machine-to-machine interface
- GUI interface
Classes of Users
- by value chain
- by level of experience
- by ...
Data Assess and Usage
The Network
- Stages: Acquisition, Transformation, Usage
- Acquisition is a stovepipe
- Usage is value chain
- In-between is the 'market'
- Fan-In, Fan-Out
- (so is GCI) not central
- holarchy , data up into the pool though the aggregator network and down the disaggregator/filter network
- Data distributed through Scale-free aggregation network. Metadata contributed along the line of usage. Homogenized and shared.
Performance Measurement and Feedback
Metadata from Providers
Metadata from Users
- Google Analytics - provides feedback and helps market process - creates values to both users and producers. Where do users come from (spatial, internet, i.e. search engine, direct Url), which pages are most popular, where do they go next, how much time do they spend on the site, how many leave after just one page?, new vs. returning visitors, ...
- Amazon - what do customers ultimately buy after viewing this product? popularity in different categories, what else did they buy at the same time? targeted advertising - interested in this product, may be interested in this company
- Ebay - knows recent searches, so on return displays products of interest,
- Google Reader - items posted, items read, which blogs did you read more from, how active are the blogs you read? how obscure are the blogs you read (# of subscribers), time of day blogs are posted, day of week blogs are posted.
- Blogger?/Wordpress - Technorati - where do readers come from?
Metadata from Mediators
Management
Partnership
- Toward an Air Quality Information Partnership (AQIP)
- Agencies: EPA, NASA, NOAA, DOE ....
- Intrastructure first
