Every blog I've tried to follow for years has had an RSS feed. Are microformats really as reliable? Like, picking a blog at random from my feed, I don't see any microdata on the page.
The biggest challenge with this strategy is discovering which links to index.
The microdata is much more data.
You can see this metadata when you post to Twitter or Facebook and it shows a preview.
We have the internal stats to show that this is MUCH more reliable than RSS but I don't really have it in a format that I can readily share.
Discovering the links can be done by just spidering every post on the front page.
The biggest challenge is the timestamp. We handle that by just keeping a fingerprint of all the URLs and we only index URLs that have either never been seen before or have timestamps.
I honestly with timestamps were required by Facebook and Twitter or at least given special treatment.
The special treatment strategy by Google and others has REALLY done a great job at upgrading the quality of the Internet by not breaking it.
Basically saying that your site will get some tangible financial benefit if you run X spec.
So if you enable IPv6 or HTTPS you get an SEO boost.
Companies actually pay attention to this and implement these changes.
Out of everything you've said, my only takeaway is that og:image is useful for Facebook and Twitter. I'll happily admit that's true, and I use it on my own sites.
Otherwise... What does "MUCH more reliable" even mean?
The rest of it seems to be you trying to catch up to RSS...
"Every blog I've tried to follow for years has had an RSS feed. "
I follow many blogs (Rule of thumb: the fewer posts per time, the higher the content of the blog likely is) and you would be surprised how many feeds fail over time. Sometimes new domain, sometimes new URL (e.g. blog.domain.com to domain.com/blog very common). And often an RSS feed contains errors that never go away.
