The problem with the initial social media response is every possible critique is loud, regardless of how valid it is, so you're left in the same place you started: trying to find what, out of anything, is actually a problem.
It's like trying to get the hour of the day from something which just responds back a list of all 12 and then claims it's valuable because 6 (or whatever) was actually one of the things brought up in the list.
That's not true, following your analogy, many posts had the actual clock photo with an added red arrow pointing to the hand! And you don't need much expertise to evaluate readability once your attention is focused on it, so no, you're not left in the same place even if you see a hundred more posts saying the opposite and posting all the 1-12 numbers.
Of course, you also have plenty of assessments like "bold" and "fresh" where your view is as fluid as liquid glass because, well, there is little substance behind them. But then this is no different from non-social media articles.
Whether correct posts appear or not is besides the point when the problem is posts for any hand position also appear.
If you already know how to filter these for things that make sense and don't make sense then going through the posts doesn't help. If you don't already know then it doesn't help to do so either.
> the problem is posts for any hand position also appear.
This is not a problem, I've specifically described how it's not (you can read off a photo, so other posts with numbers can't confuse you), so you need to address that instead of repeating the same claim.
> If you already know how to filter these for things that make sense and don't make sense then going through the posts doesn't help.
It does, the posts supply you with information, that's a net increase in your knowledge after you've applied the filter! Someone says it's not readable, you see the screenshot, apply your own filter and come out with more knowledge about the new design. Then you see posts "i can read just fine, it's great", apply your filter and ignore this
You couldn't even come up with an analogy where it doesn't work, and it also worked in this specific case of liquid design.
I think the likely outcome is you and I disagree what the problem with social media is in this context. You are sold there is no problem so long as some valid posts exist (i.e. so long as you can eventually extract something) and me/the author seem to be sold the problem is the loudness of invalid posts (i.e. if you can filter the social media posts accurately then you already have the tools to evaluate the design more directly for less work). No amount of showing how select posts can be useful is going to impact my perception and no amount of talking about invalid posts is going to impact your perception.
It's like trying to get the hour of the day from something which just responds back a list of all 12 and then claims it's valuable because 6 (or whatever) was actually one of the things brought up in the list.