WIP: Add strict binding parser note

[twin-tigon]

Hi @TimvdLippe,

This is what I got so far from issue #2508.
I'm missing the info for Polymer 3, I wanted to check with you before adding that.
Also, I'm not sure if Polymer 1 should be in scope for this.

Anyways, looking forward to your comments

[TimvdLippe]

Thanks for writing this up! This in general looks good. 1 improvement would be to add a problematic case in this example, taken from one of the issues linked in the original PR. I will let the doc writers do the full text review, as I am not the expert in that.

[twin-tigon]

@TimvdLippe I modified the example using issue Polymer/polymer#4723

[TimvdLippe]

Thanks! This all looks fine to me, but I will leave the rest of the review to our doc writers.

[arthurevans]

Thanks for this! I'd like to suggest alternate text to focus more on the use cases (when you need it, what to do) and less on the implementation. Here's my first take:

The default binding parser emphasizes performance over completeness. Some bindings can't be parsed correctly by the default parser—for example, bindings with JavaScript identifiers that include non-ASCII characters.

For a more exact binding parser, use the StrictBindingParser mixin. This mixin provides an alternate binding parser that handles all possible cases, although it's slightly less performant than the default parser.

I don't love the phrase, "handles all possible cases," but I can't think of anything better at the moment, either.

[twin-tigon]

Thanks for the feedback. Regarding the last sentence, what about something like this:

This mixin provides an alternate binding parser that supports Unicode characters, although it's slightly less performant than the default parser.