It seems a bit much that we ask for companies to constantly grow and simultaneously ensure they never pass 70% market share.
To your point: if you're the CEO of a company, and the shareholders are pushing you to capture market share, you're quite inclined to do it. Especially since capturing market share makes it more likely your company will stay profitable or become profitable, even absent shareholder wishes. Which works until/unless you reach the point where you've gotten the combination of skill and luck and circumstance that means you end up with a monopoly market share, at which point "we" (where we == the non-shareholder general public) start posting about how your company has taken over the market and the government should declare you some combination of "common carrier" or "utility" or "out of business".
Not to mention that in this capitalist society, messaging services are incentivized to grow at all cost because user-base is the best fuel for monetization. Holding onto XMPP federation like Google did in XMPP's final years of relevancy would be suicide; you would be helping your competitors while gaining nothing.
Of course everyone (on HN) says "Oh, well not me." Everyone who complains about users think that everyone needs to be Richard Stallman and only connect with the internet by faxing HTTPS requests and then reading forum threads printed off of a dot-matrix printer.
In reality anyone who is using a smartphone or modern software is pushing AGFAM to be bigger and more consolidated and anyone who doesn't totally blow open one of those segments is a wannabe, trash or should just join one of those.
How can a new search engine - or email service, for that matter - compete with Google's servers? How can a new social network compete with Facebook having all your friends already on it? How can a new encyclopedia compete with Wikipedia's vast store of volunteer-checked knowledge? How can a new operating system compete with the fact that everyone already knows hw to use Windows?
It's not impossible, but an incremental improvement won't do it; you need to be massively better than the default to overcome the sheer advantage that comes from being the default.
I mean Apple is the richest company in the world but they choose to compete in very niche markets, and hasn't actually completely dominated a market since the iPod even though they easily could with their massive stockpiles of money.
Instead of trying to bend the Internet to the point of breaking, Google could just... not do that. They could focus on putting the industry best ads in the industry best search results, running the largest video streaming service, sell a wildly successful mobile OS, and have the best online office suite without all of the other terrible things they do as a corporation.
They're already the best at basically everything, so they got bored and now need to burn their reputation like it's kindling. It's pointless.
Its often brought up, that corporations and the economy have to keep growing ... but is it really necessary? Why can't both achieve some steady state equilibrium within a system of finite resources?
Who does that? Why do you have to have 'too much market power' to succeed? If google was 1/10 its size, it'd still be successful.