Email is run just like this and it has plenty of reach.
The "reach" in gp's context is "audience reach". A measure of "audience size" in a broadcast type of medium. It doesn't mean reach in a technical sense like "reachable ip address or DNS resolution".
Email is more point-to-point and 1-to-1 rather than 1-to-many. Yes, the concept of "mailing lists" could arguably be described as "broadcast" but it requires explicit subscriptions and is not the same "reach" as Twitter tweets acting as global "billboards".
E.g. this HN site often has user-submitted Twitter tweets on the front page. In contrast, it's very rare to have a mailing list mirror post to HN. The reach is not the same. (E.g. last week, the thread on front page about "Adobe removing Pantone colors" was referencing a Twitter post.)
Another example related to HN... the creator of this site, Paul Graham, created a Twitter account in 2010 (https://twitter.com/paulg). It has 1.5 million followers. Clearly, the Twitter platform gives him exposure that his existing email address (and his web blog at http://www.paulgraham.com, and this HN site ycombinator.com) -- does not.
I remember the fluffy days of Facebook, where almost in an instant there was a wildfire of 'you might know' suggestions, from harvested addressbooks and cross referencing.
I found, and find discoverability hard on Twitter. My follow list rarely gets bumped and I am pretty clueless as to who and what is in it. I have to make an effort to grow the list.
Tweet reach may be massaged by favs, comments and retweets. Not totally sure how this is different in Mastodon.
Are we talking timeline/search promotion?
I remember email round robins pre Facebook and you did add people if you recognised them.
Directory services just got pulled as they were harvested by spammers. Not that I think this would help. You need some nudging.
If I read an article and the author had an easy lookupable and addable feed, I might note it.
Email on the other hand is a great example of how decentralized things become centralized over time due to the network effect. There's even a weak network effect with email and it still happened.
Same as democracy in the US. Even if for the most part policy is set by centralized interests, just having the populace believe that it's a democracy and they have the agency to change things makes the system function in a fundamentally different way. The ruling class can't just cart blanche do whatever they want like in China, even if they are the ones holding most of the levers.
Except it isn't because centralization-decentralization is not a binary discrete variable but a continuous one. There's a reason the trilemma exists and you score on each axis. No coin is at any extreme.