There's plenty of precedent that states forks die, however to describe it as entirely futile is:
A) Not exactly constructive.
B) Possibly untrue.
The move might indeed die, if the rust foundations leadership gets it's act together. Which is a strong push for them to sort things out rather than assuming everything will be fine with their current heavy handed and opaque approach.
Failing this, we could be in a situation with another Rocky linux, FlatCar linux or LibreOffice ;)
In the examples you gave, either the original project shifted course significantly, or there are major technical disagreements. Here, pointless trademark drama is all there is, and the fork AFAICT replaced some strings and nothing else.
Not sure what you are trying to convey. There are a ton of IRC hosts out there and they are all interchangeable.
A real language fork is not interchangeable, even a different implementation of the same language is going to have its subtle compatibility problems. People pay the price for tangible benefits like performance or portability. You don’t need to use a fork and pay the technical price for it if the only benefit is that Rust Foundation hypothetically won’t sue you for using the word “rust”; just frigging use “crab” as the moniker for “rust”.
Its an example of a wildly successful fork due to poor leadership, without any other technical reasons.
You can argue that “its different because its a community” but thats dishonest, because ultimately migrating an entire community is much harder than having an alternative development stream like MariaDB and MySQL.
IRC is practically stateless, migrating is about as easy as someone signing up at the new host and changing your readme/contact page. If you have some bots you change some URLs. It may not be an easy decision, but the migration is a straightforward one time cost. Guess why a lot of projects moved off Freenode overnight? Because it’s very easy.
Supporting a different language implementation, on the other hand, is an indefinite ongoing burden until one dies. Not to mention the trademark bullshit affects a tiny slice of the user base at best.
Everything you said is just as true for an alternative language. You can maintain compat (like rocky linux) or diverge (like mariadb), so it remains to be seen.
However the switching cost of moving your entire community is large and IRC is not stateless, pedantically it is a stateful connection, but you really mean that theres nothing you need to carry with you; but thats not true, permissions, identities, hostmasks etc. all of these tiny things spread across thousands is quite an effort and undertaking it is only done with a heavy heart.
If you maintain compatibility then it could be as easy as changing a reference or just calling another compiler (like the Java implementations). Afaik all CentOS and RHEL scripts work without modification on Rocky.
Either-way, I think I made my point, I am sure you are convinced of the inverse, so I leave this as my final word to other readers of the thread.
A) Not exactly constructive.
B) Possibly untrue.
The move might indeed die, if the rust foundations leadership gets it's act together. Which is a strong push for them to sort things out rather than assuming everything will be fine with their current heavy handed and opaque approach.
Failing this, we could be in a situation with another Rocky linux, FlatCar linux or LibreOffice ;)
(examples of successful forks)