This is a very simple scenario to implement the basics.
NOAA has a service that provides weather forecasts. This service is freely available on the web, there is no access control on the weather forecasts. It is a REST oriented service: the URL that is used to access the weather map include a parameter to specify the region that the weather is desired for. There is, presumably, a web page that lists all the possible weather maps that it can generate.
NGIT has a plume service which will generate a model of a plume. You must pass the address of a geographical feature map file, and another data file giving specifics of a fire underway. These files must be accessible without any access control. The Plume service will retrieve the files from the web, calculate a plume, and generate an output file, returning the web address of that output file. That file has not access control and can be retrieved by anyone who knows its address.
NGIT is not configured to know about NOAA, otherwise we would have no need for workflow. It seems like we could just take the address of NOAA weather service, and pass the address of that weather map into the plume service, and thus a single invocation of the plume service is all we need.
GeoBPMS has a workflow that will ... what?
Last edited by kswenson myopenid com 12/18/2008(Effective date 12/18/2008)
Simplified OGC Scenario
This is a very simple scenario to implement the basics.
NOAA has a service that provides weather forecasts. This service is freely available on the web, there is no access control on the weather forecasts. It is a REST oriented service: the URL that is used to access the weather map include a parameter to specify the region that the weather is desired for. There is, presumably, a web page that lists all the possible weather maps that it can generate.
NGIT has a plume service which will generate a model of a plume. You must pass the address of a geographical feature map file, and another data file giving specifics of a fire underway. These files must be accessible without any access control. The Plume service will retrieve the files from the web, calculate a plume, and generate an output file, returning the web address of that output file. That file has not access control and can be retrieved by anyone who knows its address.
NGIT is not configured to know about NOAA, otherwise we would have no need for workflow. It seems like we could just take the address of NOAA weather service, and pass the address of that weather map into the plume service, and thus a single invocation of the plume service is all we need.
GeoBPMS has a workflow that will ... what?