Hi dev@streams, One of our project objectives that I feel strongly about and want to make progress on in 2018 is to reduce disparities between the configuration objects that initialize similar providers from various data sources.
TL;DR - The way a provider is launched should be based on the category of data expected and the constraints that are expected to hold on the normalized version of that data. Not based on the internals of how the provider goes about obtaining that data as is currently the case. I’ve created a wiki page to get into the design of this fairly significant change, and would appreciate your feedback and participation in the process. https://cwiki.apache.org/confluence/display/STREAMS/Common+Provider+Config+Format Thanks, Steve
