On the other hand, he wanted his work published in the community registry where they got exposure and were made into dependencies in lots of projects.
When the author offered their modules and then suddenly walked away with them, lots of innocent devs who built their projects with his modules got hurt. He did more bad than good overall, especially that he unpublished hundreds of modules at once without warning, not just one. He should carry some responsibility for that.
The author can publish and unpublish all he wants from his personal site where there are zero expectations that it will continue to exist, but when he's doing it from a public repository where he received a lot of confidence from the community, he should at least make sure his users don't suddenly fall flat. Now people will start wondering if a module with millions of installs in the last month is still going to exist tomorrow.
That's a smart thing for people to wonder when there is a very real possibility that it won't. I'm not going to applaud the author's action here, which I consider reckless, but it did bring attention to how fragile this "essential" infrastructure really is.
Should really just be a 'publish is forever' mentality
You of course need to audit and improve your local store, but you need to do that with your dependencies anyway
Blame NPM. They were the ones shutting down his module and giving it to someone else. That's the first of the crazy parts.
That's a good thing. People should be aware of it - and be aware of the fact that all awesomeness of the open source world is maintained by continuous - and often unseen and unpraised - effort of thousands of people. Who are mostly known when something breaks but not much otherwise. It's ok not to think about it every minute, but it's also useful not to forget about it completely - and that we shouldn't take it for granted as it is in no way law of the universe, it is a consequence of continued effort and continued good will.