"Seegrid will be due for a migration to confluence on the 1st of August. Any update on or after the 1st of August will NOT be migrated"

Use Cases

The initial use cases for the gazetter web services are simple and these restricted by some internal project timelines.

  • GA will re-engineer its current Place Names Search to use a service sitting over the top of the existing gazetteer infrastructue.

This task is currently scheduled to be completed in Feb 2006, but may be delayed a couple of months due to staffing issues.

  • NT, NSW and AAD will provide services based on the same model as GA but providing as much extended information as is practical or useful. GA will then add an option to its Place Names search page to get the extended data from these services and display it within the Place Names application. The data model will therefore contain some "core" information while an extended model (with the same core) will be much more inclusive of all the potential attributes.

There is currently no time frame for this part of the project and it is expected that the services will be provided by organisations as resources are made available.

-- StuartGirvan - 18 Nov 2005

Of course there are potentially many more use cases for the place names gazetteer.

Visual Use Case Descripion

The following powerpoint presentation gives a visual presentation of the use cases describe above.

* Usecase for National Gazetteer Web Services Project:

Implications from the use cases

Data Models and Query Response

1) For GA to retrieve more detailed information from the state and territory based services the unqiue identifier for places must be identical between GA and the states. This is because the query to the states for more detailed information will be by ID of the place.

2) If the service follows the current data model proposed where does it derive the 1:100k topographic data sheet and name. Should a second service be set up to handle this? Its certainly something that could be re-used.

3) Where is Status dealt with? Another attribute to be added to the model? Or does it fit somehow with temporal extent?

4) What about post code? Ditto Status?

Query Profiles and Vocab Lists

1) If the query on the first page is by placename, feature type and jurisdiction then the service will have to be queryable by these attributes. So one of the canned queries will be by name, feature type and jurisdiction.

2) There'll almost certainly be a query by ID required internally to bring back the data used in step 6 in the use case.

3) There'll have to be a standard lists of feature types (which is agreed -see FeatureTypesAndVocabularies) and jurisdictions to query against.

-- StuartGirvan - 21 Nov 2005
Topic revision: r5 - 15 Oct 2010, UnknownUser

Current license: All material on this collaboration platform is licensed under a Creative Commons Attribution 3.0 Australia Licence (CC BY 3.0).