"Feature Flags" or segregating new elements from existing implementations


#1

We are working on replacing an existing element with a new (refactored) element.

We’d like to be able to have both activities available during the development period so we can fall back on the existing implementation until we’re confident the new implementation is ready to replace it.

In other systems, I’d use feature flags to isolate the features from one another and read the flag at runtime to determine which path to take.

I’ve experimented using gateways in the modeler, but this has produced a highly-cluttered diagram as we try to route around the new and old elements.

Any suggestion on how to segregate new development while keeping an existing implementation around would be highly appreciated.


#2

Can you explain what you mean by “element”?


#3

I mean a task/subtask in the modeler. Sorry, still coming up to speed on the jargon. :slight_smile:


#4

If you deploy 2 processes with the same key then you’ll get the 2 versions that will be running in parallel.

You can then use Version Tags if you want to find specific versions of a definition.


#5

Thanks! I’ll see if we can make that work for our case.