For an example in politics, larger expanding unions have typically dominated the world.
Roman Empire, Ottoman Empire, British Empire, arguable an "American Empire" spring to mind.
Isolationism typically causes the failure of these unions, cf. modern history "Brexit", USSR, North Korea etc.
The same holds for open source projects or any social endeavour.
Sadly it seems people are unwilling to put aside base tribalism in pursuit of a greater gain because of the fear that Johnny Foreigner may somehow better you by taking your work, isolating it and build on it, so you do it first. This fear is not without justification, isolation can be the start of power transfer and how empires start (USA splitting from the British Empire for example, or the EU/NATO becoming larger, arguably Russia could be attempting this now by changing world order). This policy now extends to our tech, including open source.
The only way to stop this that I can see is if the fear of loss from isolationism is greater (cf. the international space station, for Russia not having the ability to do cheap space research, for the US not be able to bring astronauts home or take up cargo cheap).
So that begs the question, how do you stop politics interfering in open source projects, how do you make the fear of loss of partnership outweigh the risk from Johnny Foreigner of stealing it?
That is required is a global leadership perspective. One thing I fought for heavily in the PostgreSQL community was to try to have the Code of Conduct Committee be geopolitically and culturally diverse. I would say that any time you can have a geopolitically diverse leadership team, you can figure out how to avoid interference by building a positive leadership culture.
> then we would risk the possibility of a fork forming around geopolitical lines.
The world's forked.
In the short term, it'll be bad for open source, but over the long term, I think it'll be a net positive as it'll lead to more diversity ( linguistically, culturally, etc ) in the open source ecosystem. The more forks the better.
If a Russian fork of a project reviews changes from an American fork and uses whatever they want but the reverse is not true, then the Russian fork gets ahead. So the incentives stack up for interdependence.
For a weak example, old and potential crypto laws in the US and probably other Countries could be a concern for a project. see:
https://www.openbsd.org/crypto.html
> If any non-American cryptographer...
It seems to be hard to draw a fair line with these things.
