"old chain"... dies off??
When I say "old chain" I am implying "old protocol that no longer is majority supported".
its never a 2 chain event. its a BRANCH that gets trimmed off depending on which branch is the strongest where the weakest branch gets cut off
Yeah, that was the theory before Ethereum hardforked.
Post ETH hardfork it was discovered that hardforks have different outcomes based on different circumstances.
The term chain is not a constant. It ebbs and flows based on its future or current state and this is why I disagree with the
"branch" term, since all branches by their design, such as in nature, must end at some point. In theory, both chains can survive,
but with the term branch, it is whether one is longer than the other. Using the term "branch" misconstrues what a hardfork
can and can not do. By saying "branch", it implies that there can be no surviving "old protocol" chain.
the old protocol is still acceptable to the branch.. because they are connected.
EG lets say old protocol block 0-450000 new consensus 450001..
by saying old protocol not supported. thats like saying it will reject blocks below 450k
killing off the old chain.
wrong on so many levels.
with 2mb, blocks of 0.0002mb->1.99mb are supported. it also does not mean all blocks have to be over 1mb but under 2mb
again blocks of 0.0002mb->1.99mb are supported. the old protocol IS supported.
Yes, I agree with what you are saying mostly.
The issue here is that I do not speak as technically as I should, since this is not my field.
When I refer to the old protocol chain, it assumes two surviving chains each building different
blocks based upon their separate rules. When only one chain exists, the old protocol chain is the
past of the new protocol chain, since the new protocol won majority.
as for your ethereum mention. please please please research ethereum. rather than just the media hype
learn how they AVOID and drop connection to.. rather than using consensus.
and stop comparing ethereums intention avoidance to a proposed bitcoin consensus. they are not the same thing
My understanding is that there were no changes made and the two chains was a direct failure of consensus.
15% of the Ethereum network chose to not update before the hardfork and chose to intentionally continue
on the "original" chain. I am not aware that those who chose not to update took any specific actions to ignore
the "new chain". If anything, the many weeks of replay attacks after the hardfork showed that they didn't
anticipate that two chains could co-exist.
by claiming old chain. thats a imagery of claiming everything before the branch right to the genesis block dies.
its doesnt
blocks are still mined ontop of the previous blocks. the CHAIN continues on. just that there are not 2 persistent branches
One "branch" will effectively "die off" in an intentional consensual hardfork.
One "chain" will effectively be "abandoned" in an intentional consensual hardfork.
One "fork" will effectively "not be built upon" in an intentional consensual hardfork.
In my understanding, the three above statements have the same meaning.
no new genesis block is created, the mining doesnt start at block 1 again. it continues on..
Yes. There is no new Genesis block, but that is irrelevant to a hardfork normally.
if you want the imagery of a chain. then it would be
some double links of a chain are abandoned and replaced with new single links..
NOT the chain is abandoned
the reason i use trees/branches is 2 fold. one.. devs love their merkle tree
and also trees grow naturally they naturally make branches and can have their branches cut off. devs's also love to "prune" .. not wirecut.. so branches instead of metal chain seems to fit their narative as a way of explaining things
Well the reason why I say "two chains" is because when a split occurs, there is literally two separate blockchains.
So the single blockchain has become two, yet prior to the hardfork point, there was only one past.
/============O
O==========|============O
So when I say the "old chain/protocol is abandoned" I mean the following:
/============O
O==========|==X
You have to remember I am only talking in the sense that a hardfork has occurred.
When that happens, there will either be two chains, or a single chain, but both will always share the same past.
The issue here is that I do not use the proper terms for certain technically recognized things.
A slightly offtopic point: Your assumption is that when 95% agree to run one protocol over another, that
they will stick to that. That 95% is effectively a pledge to perform a future act, it does not guarantee that
post-hardfork. So with that understanding in mind, within a system that no one should trust anyone
else, including miners, we choose to take a risk at a 95% pledge. Because there is a risk due to whether
individuals will keep their word, there can always be an altcoin possibility, IMO. The act of reversing the
pledge after a hardfork, would result in an intentional malicious hardfork, IMO.
you do know that if 95% have nodes that accept 2mb.. they accept ANYTHING below 2mb.
meaning
when nodes get to 95% ... 2 weeks grace pass to make sure it isnt a fluke.
miners then do a upto 95% period too plus a grace period.
by which time the node count should exceed 95%.
if it deminishes, pools wont risk it if the orphan rate risk jumped.
if it stays stable, pools would try it out
where pools only risk making blocks of 10000250byte block and see what the orphan rate is like
if happy they naturally test the waters with slightly more.
in short. back in 2009-2016 miners were not jumping to 1mb. they slowly progressed to 1mb.
dont think/imagine that pools are suddenly going to make 1.99mb the day of activation. (rationally/logically/logistically they wont)
if the orphan rate is too high we stick to 1mb.
this does not mean nodes have to downgrade. because 2mb accepts any data below 2mb
whether its 250bytes 1kb, 100kb 0.99mb whatever.
its more then a pledge because the nodes would already have the 2mb rule inplace BEFORE activation. meaning nodes are fully accepting even before pools push the envelope
right now many nodes are HAPPILY running with 2mb active. just like old nodes were happily running 1mb rule when blocks were only bing made at 0.2mb in 2009-2011 or 0.5mb upto 2013..
yep the rule was 1mb but blocks were under 1mb
future rule of 2mb while blocks are under 2mb.. same thing
Yes, I understand this. My off topic statement was just to point out that when nodes signal for a change,
that does not guarantee they will follow through, and that is also a risk that we take. I agree with everything
you say in theory, I'm just saying that if we hardforked and a large % of hash and nodes decided to immediately
downgrade back to their prior protocol version that rejected higher than 1MB block, that couldn't be stopped. So
when we do hardfork, we take the risk by taking the signalers at their word that they will follow through and not
backtrack after the hardfork. If this did occur, it would be an attack vector on the economy IMO. That's my only
point. So I don't think it is more than a pledge, since there is no guarantee they will stick with their pledge until
well after the hardfork. The pledge is only a signalling prior to the hardfork.
but here is the funny part.. segwit IS a pledge. if you did not realise it 0.8->0.13.1 NEEDS to be upgraded to a new release (not yet available) AFTER activation for users to actually use segwit transaction features. yes only avaiable AFTER.
but things like BU, and other blocksize consensus nodes are ACTIVE AND RUNNING now. they are just waiting for a majority to give miners confidence to push the envelope and test the water.. with worse case, them getting thier attempt orphaned.
Yes, in theory it is a pledge as well, but there is no hardfork since it uses the softfork method.
But yes, if enough pledge for SegWit and that softfork is activated, only after will the SegWit
addresses and transactions be functional. The only difference between the two pledges discussed
is that one is a hardfork pledge and the other is a softfork pledge and personally I think a hardforked
pledge that is reneged upon is bad for the economy. But a reneged softfork pledge shouldn't effect the
economy, that I can currently think of. But yes, they both are activated by % pledges at first, then they
back it up with work/numbers later after the fork.
