"If it's popular enough to be forked by the community"
Yes.
And if it isn't you need to switch.
If you are tied in in a way to proprietary tools or APIs then you have some major migration in front of you. If you don't you you might be able to use Node.
If you have a tight schedule already, or no development capacity for the migration, you will need to buy a license because you might not want to run your code on a plattform that no longer gets security updates and you no longer get support.
When this happend to me in the past, I've got some major bills for licenses b/c most of them are server based (or core based) and also count development, testing, staging and CI servers (>$100k/y).
"You only need to be concerned if you're heavily invested on their platform and it's not popular enough for community support and you need new features added. In this unlikely scenario, [...]"
s/unlikely/likely/g
I have no data but would assume this is the default ending, because most projects are not being forked (Mongo, ...) and create a successful community around a fork and most companies I work with are heavily invested in a plattform and do not adhere to the standard APIs and tools (e.g. I'd say 10% of my customers use AWS in a standard way, 90% are heavily invested in the plattform, same for GCP.) If you have other numbers and/or studies, I would be interested.
If it is run by an established foundation like Apache with established procedures for how to manage large complex projects, then you can be sure that the project will remain open-source and usable for your purposes. Otherwise, I'm not so sure.
Yes.
And if it isn't you need to switch.
If you are tied in in a way to proprietary tools or APIs then you have some major migration in front of you. If you don't you you might be able to use Node.
If you have a tight schedule already, or no development capacity for the migration, you will need to buy a license because you might not want to run your code on a plattform that no longer gets security updates and you no longer get support.
When this happend to me in the past, I've got some major bills for licenses b/c most of them are server based (or core based) and also count development, testing, staging and CI servers (>$100k/y).
"You only need to be concerned if you're heavily invested on their platform and it's not popular enough for community support and you need new features added. In this unlikely scenario, [...]"
s/unlikely/likely/g
I have no data but would assume this is the default ending, because most projects are not being forked (Mongo, ...) and create a successful community around a fork and most companies I work with are heavily invested in a plattform and do not adhere to the standard APIs and tools (e.g. I'd say 10% of my customers use AWS in a standard way, 90% are heavily invested in the plattform, same for GCP.) If you have other numbers and/or studies, I would be interested.