My initial reaction was, "Well, I don't know if it's a major issue." About a second later, I remembered that scrolling is pretty fundamental to the user experience. That's especially true today, with single-page layouts and feeds everywhere.

Most of us have an intuitive sense of what scrolling should feel like, so when that get's tampered with, we're likely to feel confused or frustrated rather than pleasantly surprised or delighted. If your UX is frustrating, I'd call that a major issue. Scroll hijacking can be fun, but I don't think you should trade function for fun.

With that said, there might be a right way to do scroll hijacking. I don't want to say that scroll hijacking itself is inherently a UX issue. I'd say that the way most people use it is a major UX issue. It's like parallax: It was annoying until people started figuring out how to use it effectively.

Yes, on at least two counts. Forcing a user to do something typically only works if they agree to it - wacky, I know; and why would you want to exclude pretty much every kind of assistive technology?

Even the normal cases make me furious, but on mobile you are additionally at risk of trapping yourself in a position where the whole screen is hijacked and the only way out is closing the browser. So yes, it's definitely a major issue.