You can't expect us - the general public - to know the detailed discussions of Perl6's genesis.
I don't expect that at all. I expect you to be able to understand my argument:
If feature parity was a milestone (and I expect it was) and speed is a goal (as we have heard), that's at least four and a half years to focus on speed, assuming that the new architecture of Moar only makes speed possible and doesn't confer any measurable speedups on its own. So where are the dramatic speed improvements? What feature or capability that exists in September 2019 justifies all of the churn and drama of the past 8 - 9 years by giving me something usable I didn't have in 2011?
It's especially interesting to read the comment on Patrick's post that "Rakudo needed NFG support and Parrot refused to provide it", because that's utter nonsense. But I suppose I'm fairly sensitive to opportunity costs and the community costs of not delivering something that's been long promised.