I think Polymer isn't as discussed as much as many of the rest is exactly because we don't try to take over the entire browser and abstract away the DOM. By working with the DOM, and being entirely based on web components, Polymer is a more incremental evolution (though a revolutionary evolution!) of the browser. That's not as exciting to some people as something entirely new and pure.
We believe that Polymer will have a very big impact though as we focus on interoperability and easy incremental adoption. Polymer's ultimate goal is to play nice with all other component libraries via the web components standards. As that cooper-etition plays out we are going to see a huge varetiy of web components, and styles and methods of implementing them. People will eventually realize that they can use their preferred programming paradigms even within web components, and still interop with other web components.
Basically, our goals and strategy are so different, it's not surprising to me that what we're doing hasn't fully caught on yet with everyone :)
A lot of React proponents say that the defining feature of React is the cheer amount of community modules and components available for it.
After having used Polymer, I have not been able to bring myself to use React. React is more complex but it ultimately does the same thing. In these situations, I always pick the simpler solution.
I suspect (but have no evidence) that this caused lots of people to dismiss it because of that, and they might not have taken another look at it since 1.0 released.