Taxonomy of data services

In the DCAT revision we have introduced a small taxonomy of data services, including DataService, DataDistributionService, and DiscoveryService.
Annette has recently raised the question of whether the sub-classing hierarchy is correct - https://github.com/w3c/dxwg/issues/432

As a possible additional data point, see below a summary of NASA’s data services – much more detailed but might help an analysis.

From: Lewis John McGibbney [mailto:notifications@github.com]
Sent: Thursday, 4 October, 2018 20:08
To: ESIPFed/sweet <sweet@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Subject: [ESIPFed/sweet] Shortcoming in reprDataService's knowledge (#99)


NASA has been working on a basic Data Services Taxonomy which can be seen below

  1.  Subsetting by...
     *   Variable
     *   Geographic Region
        *   Bounding Box
        *   Shapefile
     *   Time
     *   Values
        *   Ancillary Values: quality filtering
        *   Target Values: extrema filtering
  2.  Regridding
  3.  Reprojection
  4.  Reformatting
  5.  Aggregation
     *   Variable
     *   Spatial: Stitching and Mosaicking
     *   Time
  6.  Subscription
     *   Notification
     *   Data Delivery

SWEET maintains a very nice collection of Classes and Individuals covering Services. Examples include

reprDataService.ttl - SWEET Ontology Representation Data Service (parent)
reprDataServiceAnalysis.ttl - SWEET Ontology Representation Data Servcie Analysis
reprDataServiceGeospatial.ttl - SWEET Ontology Representation Data Service Geospatial
reprDataServiceReduction.ttl - SWEET Ontology Representation Data Service Reduction
reprDataServiceValidation.ttl- SWEET Ontology Representation Data Service Validation

There are a few Classes missing and some other discrepancies e.g.

  1.  Subsetting by Variable - in SWEET it is referred to as 'by Parameter' e.g. http://sweetontology.net/reprDataServiceReduction/ParameterSubset

  2.  Subsetting by Value (neither
     *   Ancillary Values: quality filtering
     *   Target Values: extrema filtering) is covered. This needs to be added.
  3.  Reproject is absent
  4.  Aggregation is missing entirely
  5.  Subscription is missing entirely

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub<https://github.com/ESIPFed/sweet/issues/99>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AAlIL7b7LMoJRnDBOZEbZw-MLVwSanWDks5uhk5xgaJpZM4XIoVw>.