Process versions

Certified Associate Developer

I have 3 doubts in Appian environment. Can you help me, please?

1. In Appian, can you work with branches associated with process versions, similar to GitHub?

2. Is it possible to have two versions of the same process running in production and choose at any time which one the application should use?

3. If a new version of a process is deployed to production, can migrations be performed for the processes that are active at that moment?

Thanks in advance.

  Discussion posts and replies are publicly visible

Parents
  • Hi Daniel

    Thanks for the questions. Appian does not have anything like branching and merging as you might find in other languages or platforms. It is a 'trunk-based' development environment so there's only ever one active version of any object type. 

    If you want two 'versions' of a process running in an environment you'd have to create a 'clone' so that the two process models were effectively independently managed objects.

    The last question suggests that you're conceptualizing the implementation of Appian applications as 'long-running process' paradigms. This is absolutely not the way to think of Appian apps. There are some very good reasons to design and implement Appian as 'short running procedures' that manage the state of the data, where the data is conceptualized as 'records'. Records are the key entities that make up your data model that underpins your application. If there's a 'process' anywhere in Appian it's a conceptual one where the state of the records governs, via rules, what procedure(s) are enabled/disabled at any point in time. Think more about a state-engine paradigm than a process paradigm.

Reply
  • Hi Daniel

    Thanks for the questions. Appian does not have anything like branching and merging as you might find in other languages or platforms. It is a 'trunk-based' development environment so there's only ever one active version of any object type. 

    If you want two 'versions' of a process running in an environment you'd have to create a 'clone' so that the two process models were effectively independently managed objects.

    The last question suggests that you're conceptualizing the implementation of Appian applications as 'long-running process' paradigms. This is absolutely not the way to think of Appian apps. There are some very good reasons to design and implement Appian as 'short running procedures' that manage the state of the data, where the data is conceptualized as 'records'. Records are the key entities that make up your data model that underpins your application. If there's a 'process' anywhere in Appian it's a conceptual one where the state of the records governs, via rules, what procedure(s) are enabled/disabled at any point in time. Think more about a state-engine paradigm than a process paradigm.

Children
No Data