opengeospatial / ideas

Public repository for Innovation Program Ideas
20 stars 3 forks source link

Additional implementations & experiments of modular OGC API workflows #115

Open jerstlouis opened 3 years ago

jerstlouis commented 3 years ago

In the Modular OGC API Workflows project, we are developing specifications and prototype implementations of a simple and natural way to define client-driven processing chains connecting OGC API services.

Financial support for this project is being provided by GeoConnections, a national collaborative initiative led by Natural Resources Canada. GeoConnections supports the integration and use of the Canadian Geospatial Data Infrastructure (CGDI), an on-line resource that improves the sharing, access and use of open geospatial information.

We are developing this in active collaboration with the OGC API SWGs, OGC API - Processes in particular, and the main resulting specifications would become an extension to OGC API - Processes.

This extension provides the ability to chain nested processes, refer to external processes and collections accessible via OGC API standards, and trigger execution of processes through OGC API data delivery specifications such as OGC API - Features, Tiles, Maps and Coverages.

Additional information:

Early implementations developed within the project include client-side support in GDAL and QGIS in a single unified driver, which in addition to OGC API - Features, Coverages, Tiles and Maps, also supports execution of modular workflows.

Participants in the project are also currently investigating the integration of ADES, JupyterLab, WCPS, DGGS, and looking at practical applications in the domains of Smart City/3D visualization and Earth Observation.

It would be great for further implementations to be developed and additional experiments conducted under the umbrella of the OGC Innovation Program, e.g. in the upcoming Testbed 17.

ingosimonis commented 3 years ago

thanks a lot for you input, @jerstlouis. I wonder if this topic is best located in a long-term project such as Testbed-17. I have the impression that - given you have already developed a draft spec - it would better into a Sprint, where we could experiment with the draft spec the first day and decide the second day how to move forward. Let's see what @ghobona and @ogcscotts think about it.

jerstlouis commented 3 years ago

@ingosimonis Thanks a lot Ingo. A Sprint might be the way to go. However my concern is that for most implementing this capability might require a lot more than a day. The vast number of different ways / combinations how different components can work together is also something that would justify a long term project, and it may also provide the glue enabling to bring together different threads and/or work packages.

Maybe @pvretano and @bpross-52n could also chime in.

bpross-52n commented 3 years ago

Having worked on this for a while now, I got the feeling that there is enough work / experimenting left to do for a long term project. I also see potential for cross thread / SWG activities. Just my two cents.

ghobona commented 3 years ago

The spec could benefit from both a sprint and a testbed thread. It takes sometime to find a sponsor for a testbed requirement, so perhaps while a sponsor is being sought, the SWG should:

jerstlouis commented 2 years ago

March 2022 Update

jerstlouis commented 1 year ago

November 2022 Update