Pages:
Author

Topic: Why not treat Core/Blockstream Lightning/Segwit like an Alt? - page 2. (Read 2255 times)

sr. member
Activity: 490
Merit: 251
Make winning bets on sports with Sportsbet.io!
Cause in the end they want to end up being Bitcoin as a whole, they don't want ALTernate coin status based on the simple fact that they want to be the premier cryptocurrency which is bitcoin. It doesn't matter on what name the scaling solution has now (Segwit or BTU) they want to end up being Bitcoin in the end and to just stay as Bitcoin.

If they were an alternate currency, they wouldn't be taken as seriously nor would anyone see a real reason to use them unless everyone is going to attempt to move over from one chain to another without any sort of fork.

Alternate coins are always going to take a backseat to bitcoin, so that's just not possible to have a scaling solution be an alternate if they want to improve / save the actual bitcoin we have and use today.
legendary
Activity: 1596
Merit: 1026
We can have both. BU wants to attack any chain which isn't BU though.

Obviously supporters of each side are going to call the other side the "altcoin".

BU is the original Bitcoin without changes.

SegWit has many radical changes.  SegWit is the alt.  BU is Bitcoin.
hero member
Activity: 896
Merit: 500
The way you call bitcoin is a really stupid alt, that's a misconception. It seems like you have not considered Bitcoin, it is the only legitimate and legitimate currency in the world, you are not allowed to smear its reputation, your calling has made it no longer clean. . BTC branching is a bad thing, BTU can be alt coin, because its nature is not limited, it is like other coin, it only shows the greed of the miners, but risk Possible cause, BTU can carry a low value, and they can pay the price. ETH and ETC are one example, while ETH grows and reaches a terrible value (currently 0.04), the ETC is only 0.01.
legendary
Activity: 4396
Merit: 4755
im just addressing your comment about you and other peoples bad maths of block times

EG
imagine a room
4 tables and groups of people sat at each table.

they are all given a set of  6dice
they have to roll them until someone gets 1,1,1,1,1,1

the red table could get it in 10 minutes.
it does not mean that because red get it in 10 minutes that the brown table would take 20 minutes or the purple table would take 30 minutes or the blue table would take 40 minutes

they could all be seconds apart but because red won, that round only red counts so no one cares about the other 3 tables times.
next round, starts again, this time the brown and blue table both get it but based on how many shakes to get it was just one extra shake in favour of brown. they are shown as doing the slightly miniscule extra amount of work so they win.. other 3 table are not counted.

now then.. if you evict the red and blue from the room.. this does not suddenly make the groups any slower. infact it helps them as there is less competition to beat them by seconds. so now brown and purple can get more blocks more often because there are less competitors to lose against by seconds

..
also a note
the difficulty is not set by "network hashrate" nor is the speed of say red table or blue tables shakes/dice helping brown or purple.
they all have different dice and doing their own work
infact the faster red and blue shake the less chance brown and purple get because red and blue could beat them by seconds.
its not the faster red and blue shake the more thy help brown and purple. (increase of network does not help the work of a single team)

in short
when antpool wins a block. its because of the work done by antpool and antpool only... the other hashrates of the other pools did not jump into their pool to help antpool. antpool alone made a bloke with thir own 600peta. not the networks many exa..

when red wins a diceround. its because of the work done by red and red only the other shakes of the other tables did not jump into the red table to help the red table. the red table alone shook the correct dice combination first with their own teams hands not all the hands of the room.

its a competition of different pools/tables doing their OWN work. competing against each other.. not a collusion of a single pool/room all shaking/hashing one thing
legendary
Activity: 1092
Merit: 1001

...
Quote
Very high consensus can't make two surviving chains.

actually it can.
an altcoin can be made by just 1 node deciding it wll do its own thing if it wanted to.
an altcoin can be made by majority banning and splitting that last 5% minority.

but then that minority have to ask themselves is there a good enough reason to try keeping it going.
(so many scenarios running through my mind.. kinda hard to summarise it all)

a few things we are sure of
1. BU wont trigger unless safe majority of node and pool (no threats, no time bombs, no blackmails. just plod along and let nature handle it)
2, core will have the triggers, time bombs and threats and could trigger while things are contentious, forcing the issue
still too early to tell

If a handful of non-mining nodes are still active, that does not count toward chain status.

In the event we are talking about, there must be some miners. If there is 1 miner/pool with
a "reasonable" amount of hash power, where that single miner has the ability to mine till a
difficultly adjustment within the next year or so, then it is still alive. The question then comes
down to whether they can afford to do so. The answer is very likely that they can not and as
such, the minority chain will die. If it takes that single miner 500 years to get to the difficultly
adjustment, even though the whole community likely left them behind, that chain survives and
will work its way out of the "difficulty snare".

Bitcoin versus Altcoin status can not be determined on "longest chain/most work" because one
day that could be used against the community in an obviously malicious way, IMO.


i think im going to drive your mind crazy now.. Cheesy sorry in advance

5% of pool consensus of blocks does not mean 5% hashpower.
5% of pool consensus of blocks does not mean 20x longer to make a block once going at it with 95% less competition...

No, I know that. This is "luck at finding the nonce" versus "raw hash computation".



the maths does not work that linearly.
EG if there are 4 pools of equal hash. it does not mean take one away and the time moves to 20 minutes.

No, I know that. The time would in theory be an average 12.5 minutes per block for the last three.


because the competition may have only been only seconds behind getting their own solution.
bitcoins are not mined based on the combined hashpower of the network.

Now I am losing you. The difficulty that exists is based to an average 10 minutes, from prior 2016
blocks. This can help estimate the networks hash rate. The nonce is based on this "combined
hashpower of the network". This is my understanding. Are you talking about "luck"?



each pool makes their own effort. and the "75%" [others] mentioned [as threshold] is not 1 pools.

No, I know that, but the 75% is still trying to find the nonce with the difficultly that assumes the
missing 25% would filling in the time gaps and balance the blocks back to 10 minutes on average.
In theory, over a 24 hour day with 144 blocks each around 10 minutes, if 25% "dropped off", then
over the same amount of 24 hours, the 75% should only be able to mine about 115 blocks. Until the
difficultly readjustment, that loss in block work will compound, which leads to less supply of new coins,
regular txs, and fees would be slowed and restricted.



here this image will make it a bit clearer

as you can see 75% of blocks is just HALF the network hashrate in this example,(top half of image) but the block
timings still are reasonable whichever way you play it(bottom half when they have split)
...

I'm not following now. are you saying that if 75% of the hash power leaves the competition,
that does not affect the remaining 25% to find blocks within the average 10 minutes over time?

What is the significance of this in relation to my comment that this was a response to?
I'm not following.
legendary
Activity: 4396
Merit: 4755
in the post from like an hour ago which u revised
I posted and ran to the store, returned, and revised.
When it posted, I noticed you answered even though before I started
your comment was not there.


Quote
Very high consensus can't make two surviving chains.

actually it can.
an altcoin can be made by just 1 node deciding it wll do its own thing if it wanted to.
an altcoin can be made by majority banning and splitting that last 5% minority.

but then that minority have to ask themselves is there a good enough reason to try keeping it going.
(so many scenarios running through my mind.. kinda hard to summarise it all)

a few things we are sure of
1. BU wont trigger unless safe majority of node and pool (no threats, no time bombs, no blackmails. just plod along and let nature handle it)
2, core will have the triggers, time bombs and threats and could trigger while things are contentious, forcing the issue
still too early to tell

If a handful of non-mining nodes are still active, that does not count toward chain status.

In the event we are talking about, there must be some miners. If there is 1 miner/pool with
a "reasonable" amount of hash power, where that single miner has the ability to mine till a
difficultly adjustment within the next year or so, then it is still alive. The question then comes
down to whether they can afford to do so. The answer is very likely that they can not and as
such, the minority chain will die. If it takes that single miner 500 years to get to the difficultly
adjustment, even though the whole community likely left them behind, that chain survives and
will work its way out of the "difficulty snare".

Bitcoin versus Altcoin status can not be determined on "longest chain/most work" because one
day that could be used against the community in an obviously malicious way, IMO.


i think im going to drive your mind crazy now.. Cheesy sorry in advance

5% of pool consensus of blocks does not mean 5% hashpower.
5% of pool consensus of blocks does not mean 20x longer to make a block once going at it with 95% less competition...

save re-writing it
the maths does not work that linearly.

EG if there are 4 pools of equal hash. it does not mean take one away and the time moves to 20 minutes.
because the competition may have only been only seconds behind getting their own solution.

bitcoins are not mined based on the combined hashpower of the network.
each pool makes their own effort. and the "75%" [others] mentioned [as threshold] is not 1 pools.

here this image will make it a bit clearer

as you can see 75% of blocks is just HALF the network hashrate in this example,(top half of image) but the block
timings still are reasonable whichever way you play it(bottom half when they have split)


ok im gonna take a break.. (brain switched off in 3.2.1)
legendary
Activity: 1092
Merit: 1001
in the post from like an hour ago which u revised
I posted and ran to the store, returned, and revised.
When it posted, I noticed you answered even though before I started
your comment was not there.


Quote
Very high consensus can't make two surviving chains.

actually it can.
an altcoin can be made by just 1 node deciding it wll do its own thing if it wanted to.
an altcoin can be made by majority banning and splitting that last 5% minority.

but then that minority have to ask themselves is there a good enough reason to try keeping it going.
(so many scenarios running through my mind.. kinda hard to summarise it all)

a few things we are sure of
1. BU wont trigger unless safe majority of node and pool (no threats, no time bombs, no blackmails. just plod along and let nature handle it)
2, core will have the triggers, time bombs and threats and could trigger while things are contentious, forcing the issue
still too early to tell

If a handful of non-mining nodes are still active, that does not count toward chain status.

In the event we are talking about, there must be some miners. If there is 1 miner/pool with
a "reasonable" amount of hash power, where that single miner has the ability to mine till a
difficultly adjustment within the next year or so, then it is still alive. The question then comes
down to whether they can afford to do so. The answer is very likely that they can not and as
such, the minority chain will die. If it takes that single miner 500 years to get to the difficultly
adjustment, even though the whole community likely left them behind, that chain survives and
will work its way out of the "difficulty snare".

Bitcoin versus Altcoin status can not be determined on "longest chain/most work" because one
day that could be used against the community in an obviously malicious way, IMO.
legendary
Activity: 4396
Merit: 4755
in the post from like an hour ago which u revised
I do not agree. I think an altcoin is always an altcoin until there is one chain.
If there is a bilateral split, the spiting chain will always be an altcoin since that
breaks the Consensus Mechanisms.

So, if both chains survive in a non-bilateral situation, the determination of altcoin
status can not be determined. I think one must die to truly determine who is
the "Bitcoin". If an implementation wishes to be the "new protocol" the old must
die, it can not be "the most hash currently". It must be finalized in general and
for the economies.
its not a simpl "hash wins"

it is about nodes, network confidence of majority nodes and also which merchants allow deposits of it.

funny part is..
by the exchanges saying they will accept bu (and if controversially call it BTU/BCU or consensus call it bitcoin and segwit BCC/SWcoin)
they are actually helping BU and other dynamics to be accepted as "the bitcoin" in a majority event
remember BU wont trigger itself unless it it has the 3.. hash, nodes and merchants.

other matter is.
BU is just one implementation. if a base blocksize limit movement trigger does occur. many implementations(classlic xt) would work with it.
core is just one implementation. if a segwit block trigger does occur. many implementations(btcc, knots) would work with it.

so its not strictly a BU event or a BU chain.. its more of a

dynamic peer vs segwit tier
rather than
bu vs core

but thats just confusing the matters more.
legendary
Activity: 1092
Merit: 1001
So back to the OP, I should have stated this:

Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins two surviving chains.

SegWit would only be an "altcoin" if there was a hardfork with two surviving chains. and segwit was minority
whereby BU and a flock of other implementations of consensus (classic, xt, and other adjustable blocksize PEER imps) become bitcoin
due to having majority of hash, nodes and merchant acceptance

core could kill off its minority chain and then just add a few lines to be dynamic and join the PEER network of many implementation



BU would only be an "altcoin" if there was a hardfork with two surviving chains. and BU was minority
whereby segwit and a flock of other implementations of consensus (classic, xt, and other imps running as downstream second TIER) become bitcoin
due to having majority of hash, nodes and merchant acceptance

BU could kill off its minority chain without any code changes can just join segwit as a downstream second tier of many implementations


i removed this line:
If BU survives and the minority chain dies off, BU transforms from altcoin into "Bitcoin".
because if BU had majority. its already bitcoin along with the other implementations (where segwit is the altcoin or just dead)

theres still some irregularities even with the FTFY summary. due to when/if core triggers their bip9/UASF/PoW change, and which markets treat it as such.
but its close enough

I just revised as you commented, see my revision.

I do not agree. I think an altcoin is always an altcoin until there is one chain.
If there is a bilateral split, the spiting chain will always be an altcoin since that
breaks the Consensus Mechanisms.

So, if both chains survive in a non-bilateral situation, the determination of altcoin
status can not be determined. I think one must die to truly determine who is
the "Bitcoin". If an implementation wishes to be the "new protocol" the old must
die, it can not be "the most hash currently". It must be finalized in general and
for the economies.
legendary
Activity: 4396
Merit: 4755
So back to the OP, I should have stated this:

Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins two surviving chains.

SegWit would only be an "altcoin" if there was a hardfork with two surviving chains. and segwit was minority
whereby BU and a flock of other implementations of consensus (classic, xt, and other adjustable blocksize PEER imps) become bitcoin
due to having majority of hash, nodes and merchant acceptance

core could kill off its minority chain and then just add a few lines to be dynamic and join the PEER network of many implementation



BU would only be an "altcoin" if there was a hardfork with two surviving chains. and BU was minority
whereby segwit and a flock of other implementations of consensus (classic, xt, and other imps running as downstream second TIER) become bitcoin
due to having majority of hash, nodes and merchant acceptance

BU could kill off its minority chain without any code changes can just join segwit as a downstream second tier of many implementations
BU could kill off its minority chain and can with segwit code additions join segwit as a upstream first tier with core

i removed this line:
If BU survives and the minority chain dies off, BU transforms from altcoin into "Bitcoin".
because if BU had majority. its already bitcoin along with the other implementations (where segwit is the altcoin or just dead)

theres still some irregularities even with the FTFY summary. due to when/if core triggers their bip9/UASF/PoW change, and which markets treat it as such. and as highlighted if there remained 2 chains or not. or the opposing implementations give in to join the peer or tier network
but its close enough
legendary
Activity: 1092
Merit: 1001
a fork is a different path.
it does not mean it has to survive more than on block. it does not mean it has to live on forever. it just means a different path has opened up

orphans are where forks are killed

Yes. I assumed that a softfork was named as such not because of a literal chain split,
but because its effect was likened to a chain split in that it allowed protocol changes.

For example, if you could make certain aspects more "strict" within the preexisting
protocol, in my mind, you wouldn't need to create a new chain to enforce this, but a
"layer" within the chain.



So back to the OP, I should have stated this:

SegWit would only be an "altcoin" if there was a hardfork with two surviving chains.
Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins two surviving chains.
If SegWit survives and the minority chain dies off, SegWit transforms from altcoin into "Bitcoin".
If BU survives and the minority chain dies off, BU transforms from altcoin into "Bitcoin".
The only time a "protocol change" skips altcoins status entirely, is with very high consensus.
Very high consensus can't make two surviving chains.
legendary
Activity: 4396
Merit: 4755
a fork is a different path.
it does not mean it has to survive more than on block. it does not mean it has to live on forever. it just means a different path has opened up

orphans are where forks are killed
legendary
Activity: 1092
Merit: 1001
Ok, I understand all that.
Are you saying that when Miners/Pools "activate" the softfork, there is literally a new chain?
no.. im saying
there are 3 possibilities.. well 6 in total of what happens.
depending if its consensual (one chain all agree, opposer's just left dead in the water unsynced)
or controversial they dont agree but they fight it out with orphans and swapping chains drama and headaches and double spend risks etc
or bilateral, by ignoring and thats when there is 2 linear nonfighting chains

as i said too many people just take soft and only mention softs best case.. then take hard and only mention hards worse case

So then in the most simplest form, a "fork" within the softfork termonology is
indeed a new chain separate from the pre-softfork chain. Correct?

If a softfork chain split is possible, why do old nodes still read the same blockchain?

This is what I don't understand. If it was a new chain, they are lost since they didn't upgrade.
A sfoftfork "tricks" the nodes with new rules without an upgrade or a chain split.

Where as I wrong here?
dependant on whats been changed. yes while the POOLS(soft) are either endlessly fighting or going separate ways
the non-mining nodes could also be in the orphan drama or just not getting relayed anything.. leaving them stuck and unsynced.

This above statement can only occur if there is indeed a new chain.


...
but based on BU refusing to bilateral split and only want consensus.

if BU got consensus.. nodes such as BU,classic, xt and other including some core nodes that did tweak their blocklimit will carry on.
the blockstream(core) that refuses dynamics or shifting to a higher base block would be left stuck and not syncing. dead in th water

Yes. I already understood that aspect.
The confusion was that I thought you are not splitting the chain but performing a "insert"
within the same chain. That is how I understood it (a restriction level applied to the same chain).


however core are threatening soft bilateral. which then can allow corefanboy pools(soft) to build on by ignoring dynamic pools.
and then that leads the nodes(hard) are in controversy because of orphan drama.. because there are 2 chains(but core would orphan the dynamic one) eventually leading to nodes(hard) doing a bilateral, to avoid seeing the dynamic block just to stop auto orphaning them by just being blind to them.. meaning it becomes a hard bilateral split.

excuse the pun... its not a trigger.. its a CHAIN of events


I don't care about bilateral splits since that is outside of Consensus in relation to my question.
I have no idea if CORE will or will not perform a second hardfork that is bilateral.



So, basically a softfork is a chain split in the same way a harfork is a chain split.
I understood a softfork as a restriction layer added to the current (only) chain.
legendary
Activity: 4396
Merit: 4755
Edit: You added more to your post: So you are saying there is a chain split and the nodes
are bouncing between the two chains with two different rules depending on "most work"?


your trying to ask a question about chain of events,

but its not a 2 answer question.
also we can speculate all day long about will core actually pull the trigger to their bip9 early(possible) if they would UASF or even PoW banish.

without knowing if core will trigger it at say dynamic vote of under 50%... or wait for things to get very threatening for core by waiting for majority on dynamics side of 75%-95%

the chain of events can alter
legendary
Activity: 4396
Merit: 4755
Ok, I understand all that.
Are you saying that when Miners/Pools "activate" the softfork, there is literally a new chain?

no.. im saying
there are 3 possibilities.. well 6 in total of what happens.
depending if its consensual (one chain all agree, opposer's just left dead in the water unsynced)
or controversial they dont agree but they fight it out with orphans and swapping chains drama and headaches and double spend risks etc
or bilateral, by ignoring and thats when there is 2 linear nonfighting chains

as i said too many people just take soft and only mention softs best case.. then take hard and only mention hards worse case

From what you provided, if there was a new chain, the nodes wouldn't be able to see the miners
anymore since they are not participating, they are literally blind and still on the same chain.

it all depends on circumstance there are 6 possible results.. not 2

If a softfork chain split is possible, why do old nodes still read the same blockchain?

This is what I don't understand. If it was a new chain, they are lost since they didn't upgrade.
A sfoftfork "tricks" the nodes with new rules without an upgrade or a chain split.

Where as I wrong here?

dependant on whats been changed. yes while the POOLS(soft) are either endlessly fighting or going separate ways
the non-mining nodes could also be in the orphan drama or just not getting relayed anything.. leaving them stuck and unsynced.

like i said for months.. its not a simple yes no answer multiple things can occur.

but based on BU refusing to bilateral split and only want consensus.

if BU got consensus.. nodes such as BU,classic, xt and other including some core nodes that did tweak their blocklimit will carry on.
the blockstream(core) that refuses dynamics or shifting to a higher base block would be left stuck and not syncing. dead in th water

however core are threatening soft bilateral. which then can allow corefanboy pools(soft) to build on by ignoring dynamic pools.
and then that leads the nodes(hard) are in controversy because of orphan drama.. because there are 2 chains(but core would orphan the dynamic one) eventually leading to nodes(hard) doing a bilateral, to avoid seeing the dynamic block just to stop auto orphaning them by just being blind to them.. meaning it becomes a hard bilateral split.

excuse the pun... its not a trigger.. its a CHAIN of events
legendary
Activity: 1092
Merit: 1001
well you did add the "non-contentious".. rather than just ask about soft..
but here goes.

What you are describing is what I and others call a bilaterial hardfork-- where both sides reject the other.

I tried to convince the authors of BIP101 to make their proposal bilateral by requiring the sign bit be set in the version in their blocks (existing nodes require it to be unset). Sadly, the proposals authors were aggressively against this.

The ethereum hardfork was bilateral, probably the only thing they did right--

Ok, I understand all that.
Are you saying that when Miners/Pools "activate" the softfork, there is literally a new chain?
From what you provided, if there was a new chain, the nodes wouldn't be able to see the miners
anymore since they are not participating, they are literally blind and still on the same chain.

If a softfork chain split is possible, why do old nodes still read the same blockchain?

This is what I don't understand. If it was a new chain, they are lost since they didn't upgrade.
A softfork "tricks" the nodes with new rules without an upgrade or a chain split.

Where am I wrong here?

Edit: You added more to your post: So you are saying there is a chain split and the nodes
are bouncing between the two chains with two different rules depending on "most work"?

legendary
Activity: 4396
Merit: 4755
well you did add the "non-contentious".. rather than just ask about soft..
but here goes.

What you are describing is what I and others call a bilaterial hardfork-- where both sides reject the other.

I tried to convince the authors of BIP101 to make their proposal bilateral by requiring the sign bit be set in the version in their blocks (existing nodes require it to be unset). Sadly, the proposals authors were aggressively against this.

The ethereum hardfork was bilateral, probably the only thing they did right--

by pools doing this. they can keep building their own without halting just by looking at the version and declining it. without even validating its tx contents

what you will find it that there are 2 chains. growing forever.. but because its soft. the non-minin nodes(hard) will get confused be and swapping between the two dependant on height (non-mining node mega orphan drama(causing hard controversy) which then forces the nodes to pick a side just so the nodes dont see the orphan drama causing a hard bilatral split.. or remain with the orphan controversy mega orphan drama of endlessly swapping

but with or without non-mining nodes the pools are building 2 chains and not fighting, just ignoring each other
legendary
Activity: 1092
Merit: 1001
SegWit would only be an "altcoin" if there was a hardfork with two surviving chains.
Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins.

yes they can
and thats where the fake sales pitch of the reddit script writers have fooled you

by only talking about the best case scenario of soft
and worse case of hard.. but not mentioning all the options

I don't go on reddit.

Can you explain to me how two surviving chains could exist in a non-contentious softfork?

se now ur twisting things

soft can be consensus... meaning no split.. or contentious possibly split or bilateral guaranteed

by you intentionally saying non-contentious.. your baiting...

try
"Can you explain to me how two surviving chains could exist in a non-consenus softfork?

and your answer is bip9 has code in it to trigger banning and orphaning..
oh and UASF does too.. i think you can guess what the S stands for

I don't bait.

My understanding was that softforks do not create a chain split since the miners are the
leaders and nothing is lost, and my understanding was that hardforks do create a chain split
because that is the mechanism that enacts the "upgrade".

So, if I am incorrect, please explain how a softfork split occurs.
Do not cite BIPs or UASF which from day one I disagreed with.

Please explain simply.
legendary
Activity: 4396
Merit: 4755
SegWit would only be an "altcoin" if there was a hardfork with two surviving chains.
Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins.

yes they can
and thats where the fake sales pitch of the reddit script writers have fooled you

by only talking about the best case scenario of soft
and worse case of hard.. but not mentioning all the options

I don't go on reddit.

Can you explain to me how two surviving chains could exist in a non-contentious softfork?

see now your twisting things

soft can be consensus meaning no split.. or contentious possibly split or bilateral guaranteed

by you intentionally saying non-contentious.. your baiting...
try
"Can you explain to me how two surviving chains could exist in a non-consenus softfork?

and your answer is bip9 has code in it to trigger banning and orphaning..
oh and UASF does too.. i think you can guess what the S stands for
legendary
Activity: 1092
Merit: 1001
SegWit would only be an "altcoin" if there was a hardfork with two surviving chains.
Altcoins, in the OPs intention, are only created by hardforks. Softforks can't make altcoins.

yes they can
and thats where the fake sales pitch of the reddit script writers have fooled you

by only talking about the best case scenario of soft
and worse case of hard.. but not mentioning all the options

I don't go on reddit.

Can you explain to me how two surviving chains could exist in a non-contentious softfork?
Pages:
Jump to: