This is a very good list by @davidslifka. I'd propose that the goal isn't to scale -- we see what mass media's and venture capital's expectations have done to the internet -- but to succeed, with many definitions of success. David explores a number of needs, includilng onboarding, hosting, moderation, and media coverage of the fediverse.

https://medium.com/@davidslifka_86286/what-the-mastodon-ecosystem-needs-to-scale-cb845f30cfe8

What the Mastodon ecosystem needs to scale - David Slifka - Medium

This is our collective chance to fix social social media. To make that happen I think the items below are necessary, in very roughly descending order. Mastodon publishes the number of servers…

Medium

@jeffjarvis @davidslifka

I can think of a few things I’d like to see that he omitted (from a great list) but the big one is easier instance switching. It should be more complete and the default would be to provide name and credentials for new instance then click switch. #mastodonwishes

@jeffjarvis @davidslifka

One click instance switch enables alternative revenue models. Like interest based instances. Trek could do a bike community for example based on marketing opportunities. They would outsource instance management of course.

The risk to users in this model is instance loss. So easy switching is key.

@jeffjarvis @davidslifka

Another part of enabling sponsored instances (Washington Post, etc) is some kind of federated tag directory (remember LDAP?) made up of billions of tag-ID, time stamp, post-ID triples. This makes small instances less disadvantaged by tag scope limits.