What I mean is that it's inevitable for a closed-source, proprietary solution like Discord. The service does matter, and there are better alternatives to Discord for online OSS communities.
Sure, just like how there were better alternatives for Facebook, and Twitter. Even when they were at their peaks. I think we've seen often enough in history that the best, most technically impressive tool is rarely the most popular one.
If you have the power to move the people you can help try to fight that aspect of society. But at large we can't even lead the horses to those theoretical oases (let alone make them drink)
I think you're missing the context of the thread. An open-source project doesn't need to choose a tool that competes with Discord in popularity -- popularity is not important here. The aim is not to attract millions of users, the aim is to provide a valuable and sustainable resource to its community that it can link in the project's README file. A tool like Discourse or Zulip is more sustainable than Discord because they are open source, history is exportable, and the software is trivially self-hostable in case the managed hosting solutions don't pan out.
