You are viewing a single comment's thread from:

RE: What I think makes Bitcoin Bitcoin, and why I think Bitcoin Cash is Bitcoin

in #bitcoin7 years ago (edited)

In Bitcoin Cash no one really waits for confirmations for small amounts outside of exchanges.

How is that any different than BTC?

At least on BTC if a miner does an RBF on a non-RBF flagged transaction he's totally going againt the intent of the user.

If you don't pay enough fee on BCH you're just STUCK, because apparently miners won't do a replace by fee no matter what.

This is so ridiculous to complain about RBF and think Zero-Conf are any more secure on BCH than BTC, They are less!

Sort:  

The difference is that on the SL chain you'll have to consider the risk that a transaction gets stuck because of full blocks from time to time. This also increases the risk of a bidding war and the potential allowing of replacing or even maliciously double spending transactions.

On Bitcoin Cash your transaction is highly unlikely to ever get stuck since the blocks are never full, nor will they be made so artificially to promote changing the scaling roadmap.

With encouraged RBF (under full load or not) the rules of the game change. It becomes more likely that transactions become replaced and this makes zero-conf transactions much more risky for all parties.

Look, I know it may all seem ridiculous on the surface. I've been there too and it's not like information has been readily available so far. Ask a stupid question on the wrong subreddit and you can get instantly downvoted or ridiculed etc. But the information is out there.

You talk about SL chain, If SL meant "Second Layer" I'm letting you know that 2nd layer like lightning network are not chains.

Why are the blocks never full on BCH? Because nobody uses it or because the miners don't mind the ultra large server and extra centralization? If it cost 0.5 BCH to overload the BCH chain with sub 1 sat per byte transactions you'd have the same problem as on the BTC chain with just larger amount. People involved in a bidding war don't care about the fee paid on the rest of the block it only care about the fee of the double spend attempt transaction. There isn't even a way to have a bidding war on BCH, you get fucked or you don't (apparently because the miners on this chain are not profit driven).

The rule of the game have always been the same, First seen first included had always relied on the good behavior of the miners over profit. When you receive an RBF transaction the wallets notify you and tell you not to trust it a zero-conf .

I have no need to go to a reddit to learn about bitcoin, why do you bring this up?

No, SL is merely the chain itself that allows "SegWit+Legacy" adresses and transactions. Before, counting 2nd layers.

Sure BCH relies on the "good behaviour" of miners, but they want to profit and that's why they do it. It's not supposed to be normal for it to be the other way around.

The other way around is how the SL network, also called "Bitcoin BTC" or just "Bitcoin", works during rush hours as of the last years. But I can (almost certainly) promise it won't for BCH, because the theory is sound again and people are working hard on making onchain scaling a reality this time.

You don't "need" to do anything. The choice is yours.

I'm asking you what SL layer meant you use same acronym again.

Sure BCH relies on the "good behaviour" of miners, but they want to profit and that's why they do it. It's not supposed to be normal for it to be the other way around.

So does the miner on the BTC chain (even more), The difference is that there aren't technically illiterate idiots claiming the transaction is safe on the BTC side.

why do you bring this up?

The choice is yours.

Yeah, Whatever!

I'm asking you what SL layer meant you use same acronym again.

Sure BCH relies on the "good behaviour" of miners, but they want to profit and that's why they do it. It's not supposed to be normal for it to be the other way around.

So does the miner on the BTC chain (even more), The difference is that there aren't technically illiterate idiots who claiming the transaction is safe on the BTC side.

why do you bring this up?

The choice is yours.

Yeah, Whatever!

So does the miner on the BTC chain (even more), The difference is that there aren't technically illiterate idiots who claiming the transaction is safe on the BTC side.

Not quite. Not during rush hour, when the upwards bidding war ensues. And don't get me wrong, a "fee market" already existed. It's good that there is one. But it was never intended to push the minimum fees up.

It's interesting that you would note that "technically illiterate idiots" don't exist on the other side. How was this accomplished if you don't mind?

The Bitcoin Cash developer teams know full well that a confirmed transaction is safer. That's not the point. The point is that even transactions without confirmations can be made safe enough, so that ordinary merchants or users can send transactions and not have to worry about long waiting times.

These "technically illiterate idiots" as you call them, such as Peter Rizun, Andrew Stone and Gavin Andresen, understand the difference. There are also suggestions to improve on 0-conf transactions further, making them both safer and also simplifying risk assessment by adding "mini" blocks for quicker high probability confirmations.

Bidding war aren't done to double spend others transactions but to prioritize one's transaction block inclusion.

You said "improve zero conf further" my claim is that RBF does not decrease btc zero conf security but indirectly increases it and that bigger block increase zero conf security to the same extent that a transaction with above average fee on btc is more secure than a lower fee one.

You realize that similar bidding war should occur on BCH with barely more transactions (10x)?

Quicker mini block? Like reducing block time?

Bidding war aren't done to double spend others transactions but to prioritize one's transaction block inclusion.

Please. I never ever suggested this. You need to consider what I say a little more carefully. One of the effects would be that the overall double spending risk would increase, just as the risk of highly delayed blocks. Sabotage becomes easier on all fronts and even if no sabotage takes place, the payees and recipients may start to lose trust anyways.

You said "improve zero conf further" my claim is that RBF does not decrease btc zero conf security but indirectly increases it and that bigger block increase zero conf security to the same extent that a transaction with above average fee on btc is more secure than a lower fee one.

In my opinion RBF is mostly useless, but when it has a use it also makes the situation worse overall, even if you can then personally pay extra and bump yourself up in the line. This can of course theoretically always happen, so RBF on its own is a marginal issue. But if anything it makes the particular problem with spam attacks worse.

You realize that similar bidding war should occur on BCH with barely more transactions (10x)?

Yes, similar. A bare 1000% more. For starters. But it's not like people don't understand that.

Quicker mini block? Like reducing block time?

No, it would establish a lesser worth "weak" block in-between the blocks. Enabling the merchant and payee to easier and more accurately quantify the risk involved. This would also increase the likelyhood that the transaction did get into the actual block. More about this from Peter Rizun if I remember, but there may be others with similar competing concepts developing for BCH.

Now I wouldn't have minded a long conversation on this if it had started out a bit milder, but for now I will need a break. There is more information available elsewhere if you decide you want it.

To most BCASH supporter Replace-By-Fee is a licence to double-spend in reality it's a way for miners to differentiate between legitimate fee pumping over first seen first valid zero-conf security.

If RBF wasn't allowed miners would have had no way of knowing if a 2nd transaction is a malicious double spend or a fee increase. (repeating the same thing sorry)

How does RBF make spam attack worst? RBF lower the total fee paid by allowing everyone to low-ball while still being able to set higher priority fee afterward.

Coin Marketplace

STEEM 0.09
TRX 0.31
JST 0.031
BTC 106980.72
ETH 3879.27
USDT 1.00
SBD 0.59