however if everything was just core.. expect another 2013 event.
diversity and more then one codebase is GOOD for the network
bu, xt classic, and even some independant core nodes have tweaks to all work together using the same native keys and will not throw out the ban hammer...
there wont be "multiple coins" if dynamics went forward.
core on the other hand are ban hammer heavy and will be the ones causing a split by banning nodes/pools/blocks they dont like.
even bip9 and UASF and even gmaxwell himself have admitted such
where as the other implementations have said they want consensus and wont throw the ban hammer.
wake up and research beyond the reddit scripts.
also core that already has the CEO adam back, CTO gmaxwell etc.
but to add to that they dont want diverse distributed decentralisation. the want just distributed centralisation.
lol. learn segwit atleast realise the soft fork is just the activation... but to actually see any segwit features, users need to move funds to segwit keys.
EG if no one moves funds out of native keys to segwit keys.. no tx count boost, no malleation of sigop disarming.
atleast learn segwit.. or are you only on this forum to defend blockstream CEO and not so much any code that could run..
however if everything was just core.. expect another 2013 event.
diversity and more then one codebase is GOOD for the network
relying on one team that has one king is no better than fiat. and that is bad.
BU, xt, classic, and many others (including some core tweaked version, will all run happily together.
but blockstream(core) only want blockstream approved code running which if one goes down. they all go down.
oh and if there was a bug in dynamics requiring a downgrade. blocks just have to drop to below 1mb again and everything is back to as it is now.
if there was a bug in sgwit requiring a downgrade. anyone with funds on segwit keys cant spend them.. locked
Thats a lie, that rebound from BU nodes comming online again happend in a very short time, thats not diversity those nodes are in control by a few people! Claiming Core nodes comming down when 1 goes bust is another major lie.
Core nodes are more wide-spread over the globe.
why are you even mentioning geolocation (wide spread over the globe)
im talking about.. implementation diversity .. not distribution
imagine core had the assert(0) bug.. if everyone was all running the same exact codebase. all nodes go offline due to exploit.
no protection from classic. no protection from xt. no protection from knots no protection from a dozen differing implementations that would run while core sort out a release without the assert(0) bug..