2008-11-16: Community Catalog

From Datafedwiki

Back to DataFed Development Events

• Search for WCS to ISO 19115 Mapping
• Look for Stefano Nativi ISO 19115 Mapping
• What does it mean to have a WCS/ISO 19115 compliant server.

Publish Services with ISO, Find with Catalog Services, Bind with Data Access Service

• ISO 19115:2003 defines the schema required for describing geographic information and services. It provides information about the identification, the

extent, the quality, the spatial and temporal schema, spatial reference, and distribution of digital geographic data.

• ISO 19119:2005 identifies and defines the architecture patterns for service interfaces used for geographic information, defines its relationship to the Open Systems Environment model, presents a geographic services taxonomy and a list of example geographic services placed in the services taxonomy.

• ISO19119 Catalog Service that operates on Get Capabilities
• What are mandatory fields to register data access service?
  • Metadata for Services
  • INSPIRES registers spatial services:
    • Scope Code = Service, PoC, Date Stamp, MD Std. Name, MD Std. Version, Service ID: Citation (Title, Creation, Publication, Revision dates) Identifier, Abstract, PoC, Role, Service Type, extent (temporal), Coupling Type, Contains OPerations, operates on Distribution Keywords, constraints.

---

• What to Include T. Habberman
• New ISO Features, T. Habberman
• Metadata for Services
• INSPIRES ISO 19115 Metadata Editor

• Break down of ISO Fields by Function

• Sample ISO Metdata

Fields to include:

• MD_Metadata is main object and encompasses MD_Identification which can include
• MD_Service ID
• MD_Data ID

MD Service operates on MD_Data - could use this to describe data access.

• aggregationInfo/aggregateDataSetIdentifier
• platform/identifier
• platform/instrument/identifier
• Could add more people: + user
• MD_Usage- Could track problems users have
  • specificUsage : CharacterString
  • usageDateTime [0..1] : DateTime
  • userDeterminedLimitations [0..1] : CharacterString
  • userContactInfo [1..*] : CI_ResponsibleParty
• Lineage can encode multiple sources - A view could be included?
• Vertical Extent
• CI_OnlineResource - Download, information, offline access, order, search
• MD_AggregateInformation
• QE_CoverageResult - vector, grid, textTable,...

• Rework framing with full ISO 19115 metadata.
• What makes FGDC a better search/discovery metadata, CF Conventions/Opendap good for usage - ISO for usage, search and understanding?
  • Search/Find metadata covers who? what? when? where? why? how?

1. Identify key fields for FGDC/ISO/KML/ATOM translation
2. What fields match what we already have in DataFed Catalog?

Image:Community Catalog Fields Spreadsheet.xls

End Users would see DataSpaces v2.0 where the content is both structured from a SQL table, harvested through usage (goolge analytics), search of the literature and user-generated feedback/ratings.

Data Access:
Data Access needs to be upfront and center. Metadata hides or doesn't include data access links. End Users want data, not metadata. Give multiple data formats, viewer options and metadata types. Link to other versions of data, other catalogs - i.e. used/new versions of the books. Also, when you buy online it's added to your "shopping cart" catalog should keep track of datasets you want to access, but allow you to continue to browse. Additionally, this shopping carts could be monitored to help end users with other commonly associated datasets.

Data Quality:
Metadata includes section on data quality, lineage logical consistency, completeness - provider-oriented, provided content. Data quality is also given by the user through reviews/ratings, capturing number o times particular dataset is used (accessed or saved in page file). End users want to know what other end users are doing in order to trust the quality of data, who else used this data? what for? Are there multiple versions? Using analytics tools in order to monitor and harvest to improve system what end users are really doing.

Multimedia Sample:
SERF metadata includes multimedia sample, a picture of what the data looks like with no information. Multimedia sample could transform into the Preview Book concept at Amazon, where you can practically touch the book. Users should be able to manipulate scales, rendering, see a temporal view and look at a particular date. Also at amazon in this view you can purchase the book - data version should have very clear data access button as well. End users may also want to compare datasets (i.e. transparency slider).

Keywords:
Keywords are already included in metadata, so to leverage the keywords let search and then refine with keywords as they go. Save the search and similar to shopping cart harvesting of information, saving popular searches would provide benefit to end user. Additional free tags by users

Ability to Pivot:
All online profiles/catalogs allow user to pivot on multiple dimensions - by person/agency: publisher, originator, custodian, serf author, End User; other Views dataset is used in (View registration with title, tags, date and bbox, as well as datasets, ability to find same datasets different location or time?); keyword; other datasets viewed; similar searches; similar datasets.

Search:
Search by bbox, time, best match word search, category. Save searches and allow RSS of search. Allow sorting of search by criteria, most views, most popular, publication date. Show related searches (Allow to pivot on search?). Allow users to improve search

Allow update of fixed metadata. share tools, methods for working with data.