On Thu, Jun 9, 2016 at 5:02 AM, Mike Shaver wrote:
> If engines pass parsing through to system libraries, it may be difficult
> for them to fully specify their quirks.
True; I wouldn't be surprised to see experiments assessing the impact of
specific proposals turning out to be influential.
Having things that currently fail become valid could also change the
> behaviour of existing programs that do speculative parsing; something to
> keep in mind.
The current behavior varies from one browser to another anyway. Assuming
the new grammar only codifies existing practice, won't any such programs
already be behaving inconsistently across browsers?