Topic: BIP100 updated - By Jeff Garzik and Tom Harding (Read 3187 times)

Do You like spreading FUD?

Add in a 'BIP100' flag to your coinbase string, and I will join with my 625Th/s! I'm aware the Kano pool is not a large percentage of total and this will likely not change much, but I'm highly motivated to initiate any traction for BIP100.

Bitcoin price is crashing coz core is centralised and people are losing trust in Bitcoin.

Core is promoting a minority agenda (segwit) since August last year (core 0.13.0)
It hasn't done a run on acceptance, it's basically a failure and it has been that way for a while now.

If core was decentralised it would include independent options for each of the block options and probably also a version of BIP100 ...

Kicking out miners who don't understand their role just creates a gap in the market where the uncooperative miners used to be.


New miners would be more than happy to take on the risk of securing Bitcoin if people like you carry on like this. Can you wave like The Queen, kano? Can you say "bye-bye"?

no the emboldened parts are exaggerated propaganda of taking softs best case scenario and hards worse case scenario.
hiding the fact that even going soft can split the network(core actually admit that it exists in their bip9, to orphan and ban the opposer once active).
hiding the fact that even going hard can keep the network together(thats how true consensus works node and pool agreement).

for reference:

Lulz "attacked" - I guess if anyone disagrees with core (and you) they're attacking you?
Narcissism much?

P.S. I'm not BU - my pool is mining "No Vote" - which is the current "running 2nd option" (BU is running 1st, SegWit is running last)
P.P.S. with no miners you have NO bitcoin. With no exchanges, you still have bitcoin.

Oh, you've solved the problems of malleability and quadratic sighash scaling a different way? What is your proposal?

This is a good solution IMO , to make bitcoin one again...
Its simple , core coukd build it in , and if segwit would never activate I could live with that.
Other small changes in future could be adopted by everyone.
f.e fixing Malleability only , without all changes of signing in a outside block.

Ok, I think I see what's leading you to say this.


To begin with, remember that BU is explictly defined as a hard fork. Naming no names, but there are some excessively over-complicated descriptions of the difference between the two. But there are nuanced sub-categories of each type, so there is some grounding in reality to presenting every subcategory at once (but it's more confusing presented that way).

It's clearer like this:


The 2 overall categories: hard and soft

Hard fork: the blockchain splits in 2

1 Rule or more are either changed, or removed. The implications are that those running the previous versions of the Bitcoin software detect the new and different version, but refuse to accept the new version of the blockchain, because the new blocks allow actions that were not in the previously rules. That's why it's described as "hard", it's an all or nothing change, as a user, you're forced to choose one or the other, because you cannot choose both.

Soft fork: the blockchain remains as 1

All the previous rules are maintained and respected by the newer Bitcoin software. The difference is that extra rules (that don't contradict the old in any way) are added to the new software. Bitcoin was designed specifically to let this happen smoothly, so that when previous versions encounter blocks that observe new rules they do not have the code with which to understand or apply, there is a predefined way for those older version of Bitcoin to safely ignore the new rules (how could the old software interpret newer code that was written in the future?). So, the implications here are the opposite to hard forks: people running the old version don't have to choose between 2 blockchains, they can stay on the same chain as the newer Bitcoin software, ignoring the new rules while still observing the old rules.


With that out of the way, it should be clearer why it's not that likely that the 3 chain forks you suggested would happen, or, at least the way things are now. For your scenario to happen, either Bitcoin "As is" must do a hard fork from Bitcoin "Segwit", or the other way round. Because there is no proposal like that right now (or at least not with any meaningful support), it's far more likely that only one out of 1) and 2) would take place, not both.

And your final question, I hope, should be inherently easy to answer for yourself now: because BU makes changes to pre-existing rules, it is a hard fork by definition. BU must fork into a separate change to allow BU's changes to be expressed in blocks, because older nodes will reject BU blocks as they break the previous rules in Bitcoin.

Love that word "nonplussed".

I can't remember exactly where (somewhere in this forum for sure; getting too old) but I saw something about neither SW nor BU gaining enough "votes" and so Bitcoin would just plod along as it is (I could easily have misunderstood).  From there I speculated that neither camp would be content and so they would both hard fork into new altcoins.

It probably isn't safe/reasonable for me to speculated thusly; it's kinda like letting kids play with matches.

Why does Core automatically get to keep the legacy?  Why does BU have to fork off?

I agree, but the word "control" was referring to the specific specs/implementations a group of developers are working on. Code maintainers can have complete control over a code repository, but the users decide which protocol parameters and which implementation to use, as you say. The model outlined above is obviously very simplified.

@David Rabahy: I agree in some ways. But how do you know which is the "official" implementation then? It's all OK if a developer group achieves universal acceptance for their work. But a developer group focused too much on one particular parameter set or solution can lead to fragmentation if there isn't wide acceptance for this solution.

It would be interesting to see a working BTU network "live" and "without interference to Core" as an altcoin. But as the Bitcoin protocol gives the option to do a hard fork from the existing blockchain, then this way to fork can be more attractive for the disagreeing developer groups, for obvious reasons (value attached to the chain).

Now you can say: Let 'em fork and end that theater. But I think such a fork would do harm to Bitcoin, whatever the outcome may be. Very probably it wouldn't be "over" that fast: For example, the "BU shilling" miners could continue to block Segwit with a small part of their hashrate until October when the possible UASF Segwit is scheduled. Until October is a long time, where both sides would probably continue with their mutual accusations, flamewars, short sellings of the "enemy token", etc. I wouldn't be surprised with a sub-500 USD BTC in this case.

A way better way would be to achieve more acceptance for a slightly modified "official" implementation. And that's where solutions like BIP 100/102/106 could help.

Or "could have helped".  Perhaps it's too late. Or not?

Will you explain why you think 1 will happen? I am non-plussed
 


Ethereum is not a good crypto to my mind. PoS is just not viable incentive wise, and the overall design of Ethereum is flawed in many other ways.



I wouldn't recommend that, cryptocurrency is too significant an innovation to abandon outright. Keep some small amount of whatever makes you feel most comfortable, or better yet, do as you're doing and continue to learn through reading and asking questions, anyone can become more confident in what to invest in with enough study.



You don't need to do anything, that's the safest option. As Angry Dwarf says, you can hedge with very simple set of steps, bear in mind you'll need double the hard disk space to run that configuration (for both the BTC chain and the BTU chain)


A recent announcement by 12 or so major exchanges has taken alot of the uncertainty out of a potential BU hard fork.

They have agreed between them to refuse to list BTU if the BU development team don't change their code to prevent replay attacks. This is highly disruptive attack, enabled by unwillingness to alter BU sufficiently from Bitcoin that the attack is impossible. The attack allows the malicious party to -

• monitor the 2 blockchains
• copy transactions from one of either forked chain (which are of course free for all to see)
• send the same transaction on the chain the original was not sent on

All the attacker gains is annoying those who get attacked, they can't steal for themselves. They're just moving the coins against your will, simply because they have everything they need (a signed transaction) to playback the same transaction on the other blockchain. But if you were sending the money to an address not in your node's wallet (an exchange address being the obvious example), there is a risk that you wouldn't be able to get your coins either. So it's pretty damn serious overall.


Because of this statement from the (major) exchanges, BU now basically have no choice but to fix the attack vector, as the exchanges couldn't list BTU in good conscience knowing the serious losses or uncertainty this could otherwise have caused. So they've done us all a favour, and the ball is in BU's court. I expect they'll fix it.

In the case of a bilateral split:

Make a copy of the blockchain/wallet directory
Install both Bitcoin Core and Bitcoin Unlimited (Might need to make a Program Files directory copy, or change the name of the binaries if they conflict)
Change the ports in one of the blockchain directories in bitcoin.conf to prevent conflict.
Now you can run both Bitcoin Core and Bitcoin Unlimited using the -datadir option. You are automatically hedged at this point by doing nothing.
You do not need to move Bitcoin Core UTXO's to SW-UTXO's unless there is an announcement to retire native UTXO's by Bitcoin Core.

I think, if it is even functionally possible, we might end up with three coins;

1) Bitcoin as it is running right now with no significant development team to love it at first
2) Bitcoin-SW as envisioned by the "Core" development team
3) Bitcoin-BU ...

One thing I know for sure; I will continue to run my own full node (which variant I need to think about).  If SW does indeed soft fork then I will need to remain vigilant and either move my non-SW-UXTO to SW-UXTO (it's too early for that, right?) or move it to the -BU coin (how is that done or am I confused?).  I don't want to lose my investment.  All this debating/fighting is undermining my confidence.  Maybe I should punt it all over to ETH (although I may have missed the majority of the run up already but at least maybe I could hold value there).  Otherwise maybe I should just take my profits back to the fiat world (yuk).

Indeed.

The diversity you're seeking is to be found in the other cryptocurrencies on the market.

This observation lies behind my objection to all the competing Bitcoin dev teams; the cryptocurrency marketplace is where the competition of ideas can legitimately and productively take place, not within Bitcoin's development itself, where it's obvious that only one development team can be in charge at any one time. Hence the fierce debate.

If a modification of the Bitcoin client is not accepted by the team of coders doing Bitcoin development, it can compete on it's own merits as a separate cryptocurrency. All competing Bitcoin development ideas (XT, Classic etc) always took the line that Bitcoin must be their vision, and that they must depose the "negligent" or "corrupt" etc incumbent team with a hard fork.

It seems to me we are grappling with a multi-dimensional problem and there might not be any one "perfect" set of parameters and certainly what might have been good enough at one time could be less than good at a later time.  I shall attempt to list the dimensions;

• strategic intent
• functional goal
• durability
• stability
• robustness
• participants & their roles
• consensus
• PoW vs. PoS vs. TaPoS vs., etc. {what is the general name of this?}
• block size
• transaction fee
• block reward
• target time between blocks

I just bet there are others.  It seems unreasonable to expect everyone to agree on a common set of parameters across all of the dimensions and across time.  Many/most/all of these dimensions are interrelated.  Someone/anyone espousing a parameter choice for a single dimension risks being misunderstood/abused.

Given that no one set of parameters could possibly satisfy everyone, it is completely natural to have multiple solutions.  To resist this natural process and attempt to "force" everyone to agree is a disagreeable thing to watch.  Moreover, I think it will lead to disillusionment both within and outside the community.

I agree with you. I don't expect you to continue to be reasonable, but I suppose I will try.

Do not exaggerate my position,  sidestep, divert, answer questions with unrelated questions, or use non pertinent slander. Not that it should need saying


So, I agree. But you're starting by implying that I want 100% offchain transaction, with only channel opening tx's for fees. I've never said any such thing, and so your dangerously close to behaving in bad faith already.



You're right. Again, you're implicitly suggesting that's what I want; it isn't. Careful.


There is a balance to be struck between on-chain and off-chain, this is correct. That's what we're really arguing about, so it's good that someone has outlined it (and no, not you, me. This is the first time you've even gone near talking about the overall long term of late, and it's me that's summed that up in simple terms here)




You're still having this problem. I've never once said that on-chain should be ruled out.

I have said that on-chain scaling and blocksize increases do not correspond to one another, at all.


Read my lips: increasing the blocksize does not change the scale, it adds more resource usage at precisely the same scale. Stop arguing that blocksize increases are on-chain scaling, it's very very simple logic that blocksize changes are resource increases, not scaling increases.


Do not make me repeat this to you, I am fully expecting you to exploit the subtlety between adding resources and increasing scaling


And so my argument is simple: there are ways to scale on-chain. Let's do those, and achieve a sustainable balance between on-chain and off-chain transactions that has provable longevity. Any problems?

People are able to discuss issues directly just fine, only not with you.  Because your first port of call is to insult their character, call them dishonest and then refuse to actually have a discussion at all unless they agree to join your little circlejerk echochamber.  You don't make technical arguments, you just shout about it being dangerous to even utter the word blocksize and then throw more insults around.

If the "ideal" scaling solution involves significant off-chain activity and smaller percentage of transactions on-chain, where are the network fees coming from as the block reward diminishes?  The general notion is that fees have to increase over time to compensate for a reduction in mining rewards.  If the cost for an individual to settle their payment channel is prohibitive or excessively time consuming because they have to wait for a block, not only does that mean off-chain doesn't work, it also potentially diminishes network fees because all it serves to do is prompt users to explore other options.

The biggest argument, which for as long as the scaling debate has been raging, no one has ever overcome, is that users will not support a system that does not support them.  If you turn Bitcoin into a gold 2.0 settlement layer, it could either fail to generate sufficient tx fees to support the network, or the fees will become so excessive that only corporations and banks can afford to use it and its utility is vastly diminished.  At the end of the day Bitcoin provides a service.  Services that don't meet customer demand die in favour of those that do in an open market.

If Bitcoin's main chain achieves a greater capacity, the cost can be spread over a greater number of participants and fees shouldn't rise to the point where most users are unable to afford it.  Bitcoin retains its competitive edge.  This is a perfectly valid argument as to why we shouldn't rule out on-chain scaling.

Tread carefully with your "I know best" condescension, because you have no way of foreseeing the outcome of your beliefs and direction any more than the rest of us.  Depending on how the system develops or evolves, your supposed "technical superiority" hardline totalitarianism could prove just as short-sighted and dangerous as the "gigablocks by midnight" crowd.

You know you're getting the balance wrong when you're unable to discuss the issues directly, and need to dog-pile your way to a back-slapping false victory.


You've got no proper technical arguments, just like the rest of your buddies. I don't need hype men on the other mics to back up my choruses, because the lines are right

You know you've got the balance right when you're getting it in the neck from both sides.  



That's pretty close to my stance, although I'd probably phrase it more that developers guide the protocol rather than control it.  The shift in emphasis is subtle but still significant.  Any developer can make any changes they want, but it's ultimately down to the users if they elect to run that code or not.  And obviously most devs would have a pretty sound idea of which changes the users would likely accept and which ideas would never get off the ground, so things won't get too anarchic.