Allow me to be off topic and rant about git here for a hot second. It is not as off topic as you'd think.
It has been proven again and again, over thousands of year but especially recently that the real valuable data is often in the raw notes of a research and not the published material. git allows you, very easily, to destroy history just to make the log prettier later. Instead, it should allow you to construct a hierarchy of logs and hide the nonimportant details but allow it to be shown later. Similarly, git reset --hard should not throw away work, that's criminal https://gist.github.com/chx/85db0ebed1e02ab14b1a65b6024dea29 fixes it but bah.
Back to topic: when there are hundreds of thousands of public posts, the mind boggles you are asking whether there's any value to it. There's no question some of it is valueable. I do not know how to impress this mindset where losing information created by people is henious, unthinkable.
I was part of a decently sized reverse engineering group there. A lot of valuable knowledge and interesting discussion is stored on G+ for some bizarre reason. It was never a great idea, but it's there and I suspect that there are plenty of similar groups that used G+ that simply aren't hellholes. Your anecdotal experience doesn't mean anything when there are clearly people who believe there's something there worth saving.
Because there is a huge amount of important public posting on there. It didn't get billions of users but it had enough. Keeping up a read-only blob, or even better letting someone else do it, should be in the shutdown considerations of any major product.
G+ had a nerdier audience and many interesting communities and posts. I don't get this ghost town line. For example HN is a niche site, with almost no audience, with almost all accounts dead. Right?
HN has niche appeal, a strongly focused discussion (there's one story feed, and effectively about 30-40 stories that really register on the front page, though many more are submitted), and pretty dedicated professional moderators. As of 2015:
Roughly 2.6M views a day, 300K daily uniques, 3 to 3.5M monthly uniques. It depends on how you count, of course.
Which ... actually probably compares favourably with Google+, which had a core of about 50-100k highly active users (posting 50-100x monthly), and maybe an extended set of as much as 100 million who'd interacted with the site at one time or another significantly.
I've done a fair bit of measurement (limited by available resources and indicators), and one conclusion I'm coming to is that raw numbers do a pathetic job of indicating media or forum vitality. Most especially raw census numbers.
Looking at G+ communities, and running grid plots of engagements, it turned out that posts drove other engagement, not members, and in fact it seems as if there's some kind of fall-off (at least on a per-member basis) when a given forum gets above about 5,000 members (though I need to check this).
Google had more users. But they were spread out over a vastly larger set of forums and discussion, there was no central "square" (as with HN's "new" or "news" pages), moderation was exceedingly uneven, and often entirely absent, and there were (and remain) huge barriers for like minds to come together.
HN's overall focus is fairly (but not excessively) narrow, and much of the conversation takes itself too seriously (and certainly myself), but relative to the rest of the Net it's an exemplar. Good conversation remains exceedingly hard to find.
> it turned out that posts drove other engagement, not members
Yes, this was something particular to G+. Incidentally see these in Linus interview [0], HN post [1]
" The whole "liking" and "sharing" model is just garbage. There is no effort and no quality control. In fact, it's all geared to the reverse of quality control"
"I'm not on any social media (I tried G+ for a while, because the people on it weren't the mindless usual stuff, but it obviously never went anywhere)"
