He isn't saying that Core will unilaterally take the decision to start orphaning blocks. He's saying that the network participants, miners and nodes, have the power, if they really wanted, to force activation of SegWit by orphaning blocks. But it's also equally possible the network participants won't elect to do that. The option is there. That's all he's saying. The users make the decision, not Core.
But in fairness, I suppose if anyone proposed the same tactic in an attempt to active a hard fork, it's pretty clear that the people in the other camp would overreact in equal fashion.
bip9 is a core invention that bypasses users choice.
its purely a pool election. not a full node election.
they went soft to avoid user choice. and if pools dont go for core code those pools loving core can ignore those pools not voting for core.
thus creating an intentionally splitting hardfork to activate a softfork
thus again no user choice and no pool choice. core can blacklist if they got anything above 75% to force it in but prefer to wait for 90% to force it in or hope for 95% to not need to force it in
as for my mindset.
although core do their own spoonfeeding. there needs to be open discussion from those not core.. otherwise its just circle jerking the flock of sheep down a single path.
in an open network there needs to be diversity. i personally think that the only 2 options are core or BU is bad. there needs to be more diversity and that all implementations should have releases of eachothers 'bips' and let the networks have a true free choice related to which bandcamp they prefer
EG BU have a dynamic block release and a segwit+dynamic block release
EG core have a segwit release and a segwit +dynamic block release.
then its true free choice
but like i said only 2 'bandcamps' that dont support open choice is bad