If Segwit2x fails to activate by August 1st SW2X supporting miners could follow the UASF chain, activate Segwit on that and then 90 days later the SW2X code would STILL hard fork – despite never activating itself.
This is possible because the Segwit2x code's condition for a Hard Fork is that about 90 days (actually 144 blocks * 90 to be exact) after Segwit activates it will hard fork – regardless of how Segwit activation was achieved.
It doesn't require that Segwit2x itself was ever activated by getting to the 80% bit 4 signalling threshold.
Both Segwit2x and UASF are trying to activate Segwit under BIP 9 by orphaning blocks that don't signal for Segwit – thus forcing a chain where 100% of blocks signal for Segwit activation. Details on this here.
So… if (say) 70% of miners were signalling for Segwit2x on bit 4 and it didn't activate in time to prevent a chain split with UASF they could follow the UASF chain by running compatible code (e.g. BIP 91 or even 148 itself), activate Segwit and then run Segwit2x again… and they would still be on for their hard fork.
I'm not saying this is particularly likely – just saying it's another possibility to consider.
Ref for the HF code itself: http://bit.ly/2vo6mgE
Submitted July 15, 2017 at 03:26PM by wintercooled
via reddit http://bit.ly/2vo6nkI