Here are two recent questions that I would consider low in quality:
Both refer to videos. The first one is a single sentence with a link to a very lengthy video. The second says he saw the video somewhere but can't find the link.
It seems to me that there are multiple problems that apply to almost all questions that are asked about videos:
They usually don't take the trouble to write up what they think is going on in the video, so in order to answer the question, people would have to watch the video, which is very time-consuming. The asker usually doesn't even give a time index that can be skipped to for the relevant part.
The other problem with people failing to write up anything is that the question no longer stands on its own. The video will at some point stop being available on youtube, or youtube will be superseded by something else, or whatever...
Physics.SE is supposed to be for "active researchers, academics and students of physics and astronomy," i.e., for people who are somewhat serious about the subject. People who show up here saying, "I saw this video..." are typically not at all serious about the subject.
The video medium itself is just not well suited to presenting complex ideas. This also tends to drag down the level of the questions to something below what I perceive as the intended level of physics.SE.
Of course there are exceptions to all of the above. For example, some people watch a video of a physics lecture online and show up here with a useful question, e.g., "At 3:47, Feynman writes the equation $E=-\nabla\phi+\partial A/\partial t$, but isn't the sign of the second term wrong? It disagrees with the sign in Jackson, but is this just a different sign convention?"
But for the more common case where the questions are of low quality, would it be helpful to set some kind of standard policy and/or consensus for how to respond to the question?