All of those “features” come with with a heavy price—responsibility. Running your own instance is increadibly time consuming and difficult to understand how to secure and scale. It also can be quite expensive if it grows to any meaningful amount of use.
1) Ruby on Rails is, well, bad. It was in the past, but I hoped things had changed. I tried running Mastodon, and in the two weeks I was playing with it late last year there were two dependency updates which broke Mastodon, until Mastodon was updated. It's untenable, and it certainly adds to the idea that running your own is "increadibly time consuming", if you have to worry about breaking things any time any dependency gets updated.
2) The failure of Ruby on Rails leads to insecurity. Most people don't update because of #1 and just let things run. This is all fine and dandy until there are real security issues, at which time people sometimes panic, update and break things, or there's a compromise.
3) Mastodon (and Ruby on Rails) isn't performant. A decent modern server, particularly one that could be bought for $5,000, should easily be able to handle 30,000 users, if one were to extrapolate from some of the articles written about Fediverse hosting... But the articles showing very good performance aren't about Mastodon. Some specifically talk about migrating to other software because of Mastodon performance.
So is it time consuming? Yes, because it's slow, breaks easily and doesn't scale well. I think it's demonstrably true that other software (Pleroma, for example, or Akkoma for those of us who aren't fans of the Pleroma admins) is faster (and therefore scales much better), isn't remotely as fragile (and therefore can be kept up to date and secured with ease), and less time consuming (most updates take literally a minute or two).
Perhaps the author should consider the idea that the failure of Mastodon is not the same thing as the failure of the Fediverse precisely because other options exist.