This is the disconnect between the modern commercial web and the real web. Personal sites aren't usually about gettting views, making money, or grabbing attention. Those are profit motives.
Personal sites are about putting something out there you care about. It doesn't matter who, if anyone at all, sees it. It doesn't matter if your site only has 90% uptime. Or if it looks weird and only works in some browsers. The only thing that matters is if you like it.
On the "old web", content freshness is what the creator wants it to be. Old articles can still be as relevant five years down the line, given that the author updates it as needed. Cool and weird stuff can still be cool and weird after the fact, doesn't depend on word-of-mouth or algorithms to "live", doesn't have to be "released" on a schedule, and doesn't have to compete against other creators. It just exists for people to stumble upon eventually: the when doesn't matter.
But most people will still need some kind of validation to make it feel like they're not just putting information out into a silent, gaping void. You don't have to stoop to Twitter style hot takes to do this, but some acknowledgement from the outside world is still desirable.
Back in the day, there were "guestbooks" you could sign and webrings you could join, linking to other related sites. Blog software had user comments and blogrolls built in from Day 1. Unfortunately spambots and useless pingback notifications killed the discussion sections.
Pingback is old and well known so it gets a lot of spam. But webmention is still new enough it's still useful. Sure, it'll eventually get spammed out to but this is the lifecycle of web-objects. Gotta get use of out them while they're young and not terrible yet.
Following a 90/9/1 rule [1], where most of the users are consumers, it becomes difficult for "average" people to contribute content when you've raised the bar so high that they need to have a personal site. That level of overhead will hinder many people. It's not just the technical aspect either -- just the problem of choosing a color scheme or layout is a roadblock to many people. When you can just type a post into a short input box or in a webform and just let Twitter and Facebook worry about the "publishing", that's a much lower bar to cross. And when you lower the cost of entry, you'll have significantly more participation. I think that's the main draw of the current centralized services -- the bar of entry is low enough that the proverbial mom or grandmother can now use Facebook (or whatever) to post their own content.
Unfortunately, that leads to one of the downsides of the centralized services, which was your point. Now that 9% of people who want to contribute content and learn about how to do it (design their site, wire up the services, etc...) now don't have as big of a potential audience.
Which I guess we have the same point, just from different angles (consumers vs producers).
[1] https://en.wikipedia.org/wiki/1%25_rule_(Internet_culture)
Their crude design has led to the fact that every iteration of the past 30 years is pretty much trying to solve the same problems: discoverability, interactivity, easy-of-use, scalability, security, performance and so on. We try to fix things without breaking the world.
More to the point, HTML, HTTP or URI don't give you the tools to actually promote your content. They just enable you to put a node in a large network of hypertext. That's it. HTML is meant to structure content by adding cues / tags in such a way that machines can grok that content. But it doesn't tell you how to interpret the content. HTTP just allows you to connect a client to a server and exchange HTTP messages. It doesn't define what's in those messages. And a URI is simply a spec that defines an identifier, not what it defines exactly nor how things could or should be linked.
It's only when the Web became this pile of hypertext, too large to index by hand, that the current denizens of the Web started to pop up. Google is ultimately nothing more then a fancy telephone book. And social media are at their core just a single big shared wall where everyone can tack on messages.
Arguably, building a website is the least efficient way of reaching a large audience, simply because the core tools of the Web weren't inherently designed to do that.
Hence why we've seen how SEO and marketing hackers are at odds to game search indexes and their algorithms. It used to be that their influence was limited to the space telephone book publishers reserved for advertising. Isn't it that then strange how SEO marketeers aren't effectively not doing things that different from what their forebearers did in the 1950's?
The big power of the Web isn't the fact that your page can be found just by Google or shared via a social network. That's a reductionist form of looking at what you can accomplish.
The true power is that you can connect to others without needing advertising companies or moloch telephone book publishers.
See, the same crude robustness provided by those open standards allows you to create your own network of hypertext between you and many friends. The Web is not a single glob of content; it can be many globs of content that are interconnected and can transform as human relationships and interaction transforms. It's not a static thing that needs to be held down by incumbents merely because mom and pop don't know how to write HTML.
You can perfectly have your own web presence and be part of a centralized service as well. Try looking at principles such as POSSE (https://indieweb.org/POSSE) You can have your cake and eat it at the same time. That's totally possible.
See, if Google and big tech try to slay URL's; they effectively disconnect themselves from the Web. They isolate themselves. And that's a bigger risk to themselves then it is to the survival of the Web. Because although the Web is huge, it's resilience is found in it's ability to adapt and evolve as provided by the core design principles that underpin the entire tech stack.
Everything else is mainly sugar and sprinkles.
And as rightly said by superkuh > "Personal sites are about putting something out there you care about."; someone somewhere will share it and someone somewhere will find it useful.