Do you not know how modules and importing work, or why don’t you want to do that? Why would that be harder to manage?
What you have read appears to be misleading or wrong.
There is no operating system mechanism that works at the level of variables between processes.
The only way I could think it may make sense to be different applications is if the intent is to expose an API or contract and then allow other ones to be written in whatever language and follow the contract.
Even then the contract would have to be logging/printing/doing something specific. A far end example of this type of thing could be grpc with protobuf (Google it if it’s a new concept to ya). Then any language could be used with protobuf being the contract. A more simple one is just printing json or specifically formatted text.@
If that isn’t the plan, these should just be modules, imported, used with data returned as needed.
Don’t make things more complicated than they need to be. If you do, you’re just making things harder on yourself and future people dealing with the code.
If we’re all misunderstanding, give an example of a language that does what you’re describing.