Based on my understanding of the Polymer strategy, I wonder if any name that speaks to directly to web components will prove too limiting in the future.
On a
separate thread, I presented the view that the current polyfill strategy is suitable for much more than web components, and could (and should) be adopted for future specs. The fact that, for example, Pointer Events and Web Animations are grouped under the "web components" umbrella seems more a historical accident than intention. In the case of Pointer Events, that shipped independently in IE, and both of those technologies could be used in a page that has no component nature at all. I hear Google referring to these as web component technologies, when to me they're just new web technologies.
The same thing will surely happen going forward. The next time someone proposes a new, polyfillable spec for a feature that has nothing to do with components. Having to create a new polyfill library just for that one spec seems overly complex. Won't the most natural thing to do be to include that in platform.js? If that library is renamed webcomponents.js, it seems likely that name will eventually become increasingly inappropriate. (Cf. jQuery)
My two cents: Polymer is a great name. It's catchy, memorable, and has an etymology that can be explained. If it were up to me, I'd rename platform.js to polymer.js. Polymer would be the "brand" by which the world refers to a vendor-independent browser compatibility library.
Otherwise, I'd recommend selecting a name that suggests the role the polyfill library fills: standards.js, or compat.js.
The thing I'd rename is actually polymer.html. In my understanding, polymer.html has a more focused responsibility: support the creation of custom elements. That role seems unlikely to evolve much over time. If that's correct, then a descriptive name like customelements.html would be attractive. (Of course, I could easily be misunderstanding the boundary between platform.js and polymer.html. Please correct me if I'm wrong!)