Last week, Blizzard announced major changes upcoming to Roadhog, specifically his hook, dubbing it "Hook 2.0." When those changes hit the Overwatch PTR, my suspicions were confirmed: Blizzard went too far, making it incredibly easy for anyone to break a hook even after being hooked legitimately.
After people pointed out how easy it is to break a hook now in the PTR build, Geoff Goodman, principal designer for Overwatch (the same person who announced Hook 2.0), posted on the Overwatch forums that they are "playing around" with some other changes that will limit the more "extreme cases," such as the video linked above.
Goodman said they are trying to make it so Roadhog's hook slows down a player when it impacts them, making it harder for people to dart behind cover after being hooked. While I don't think this goes far enough, since, as Goodman said, it will really only help the "extreme cases," I'll take it, but don't be surprised if this still severely impacts Roadhog's effectiveness and subsequent appearances in competitive play.
Those complaining on the forum about the recent D.Va nerfs in the PTR don't seem to be getting the same reprieve. In fact, Game Director Jeff Kaplan recently posted in the forums in response to complaints about D.Va's nerf (more health, but less armor and less damage), saying that only ".26% of Overwatch players have tested these changes on the PTR," meaning that "99.74 percent of the community hasn't actually played with any of these changes."
So, for now at least, it doesn't seem that Blizzard is all that impressed with complaints about D.Va, and this may lend itself to the importance of community involvement in testing more than anything else.