
Today we released the first Flowable engine version (5.22.0). This release is a fork of Activiti 5.21.0. There have been no package renaming or config file name changes. The easiest way to start using it is through Maven:
The version 5 code is on the flowable5 branch on Github https://github.com/flowable/flowable-engine/tree/flowable5.The version 6 code is on master https://github.com/flowable/flowable-engine.
The 5.22.0 release has the following highlights:
First Flowable release. With a new Maven group (org.flowable) and artefact id (flowable-engine etc), there are no changes to package names or config file names in the fork of Activiti.
Introduction of transient variables. Transient variables won’t be persisted in the process variable tables, but are still available on the execution for the duration of a single transaction. Transient variables can for example be used to store a REST response that is only needed to create a request to another REST service, or to use a Java service task result in the next Java service task without it getting persisted. You can read more about it here
Several bug fixes and smaller enhancements
A lot of people will also be wondering what will happen with the version 6 engine. It’s our main focus to get a new beta release out as soon as possible and we want to get the final 6.0 release out shortly after.
We understand that there will be a bit of confusion around starting the Flowable project. We don’t plan any package renaming or config filename changes for example. For now, please get in contact with us through the forum http://forum.flowable.org, and by creating issues on the Github issues page https://github.com/flowable/flowable-engine/issues.The developers of the Flowable project are keen to help you answering your questions and helping out with possible issues.The 5.22.0 release is the starting point for the Flowable project, a lot more releases will follow.

Tools like ChatGPT can handle a variety of business tasks, automating nearly everything. And it’s true, GenAI really can do a wide range of tasks that humans do currently. Why not let business users work directly with AI then? And what about Agentic AI?

In this post, we continue our exploration of workflow complexity - learn how key metrics like activity count and control flow reveal natural groupings of models, making it easier to identify and improve overly complex designs.

In this post, we continue our exploration of workflow complexity - learn how key metrics like activity count and control flow reveal natural groupings of models, making it easier to identify and improve overly complex designs.