Hybrid Nodes: Why I'm Glad Bitcoin Cash Lost the Hash War.
Size matters.
Especially when that size is in reference to the size and frequency of the blocks on-chain. For example, Hive has very tiny blocks (current max size 65KB). Heh, seriously look at that "post". I filled up an entire block with "0123456789" until I reached the max limit of ~65k one byte characters. One day people will look back and be like, "Wow, this @edicted guy filled up a block just because he could? That's crazy." If you tried to do something like that on Bitcoin it would cost like $40k. Blockchains are weird and inefficient. Bandwidth is precious.
So why am I glad that Bitcoin didn't increase blocksize?
This is an extremely contentious issue. The debate about whether Bitcoin's blocksize should be increased has been argued in a thousand different ways and even spawned multiple conspiracy theories. For example, some people would tell you that Blockstream controls Bitcoin code updates and keeps the blocksize low to centralize the network so that it can more easily be controlled by centralized exchanges and Lightning Network operators.
Blockstream and much of the Bitcoin community would claim that they want a very low blocksize to increase decentralization of nodes. If running a node takes very little resources, many people all around the world will be able to run one on the cheap. That's a good thing, but is it worth it? Many would say that it's not. After all, computers are at least ten times faster now than when Bitcoin was invented. Shouldn't we be increasing the blocksize if not only to "keep up with inflation"? There is certainly merit to both sides of the debate.
But I am definitively warming up to the idea.
Why is this? Because the Bitcoin network hasn't even come close to unlocking its full potential. This has become painfully obvious to me during my research. For example, the Bitcoin network is creating an unhackable random-number seed every ten minutes on average after minting every single block.
Unhackable RNG.
Imagine someone created a lottery using the Bitcoin block hashes as the seed for random number generation. Let's say the prize for winning this lottery was one million dollars. You'd think that with a one million dollar prize pool it would be worth it to cheat, but mathematically this is simply just not the case.
First of all, cheating is only an option for users who control a massive amount of hash power, meaning they have to control one of the top ten mining pools (also they have to risk their reputation and their delegations in the process if they get caught). Already this is a tiny tiny subset of people compared to the world's population. Already we are on thin ice, but let's keep going.
So this person controls a metric ton of Bitcoin hashpower. How are they going to cheat at the lottery and win one million dollars? Well, that's easy... all they have to do is mine Bitcoin blocks and throw away valid blocks that don't win the lottery. Already we can see that this is an impossible scenario.
Every Bitcoin block has a reward of 6.25 BTC plus all the transaction fees of that block. At these prices, they'd have to discard $200k in block rewards for every single block they needed to throw away before they won.
Let's say tickets cost $1 and each one has a one in a million chance of winning. Even if they bought 1000 tickets they'd still have to throw away 1000 Bitcoin blocks in a row on average (two hundred million dollars) in order to win the one million dollars. And this all assumes that a single mining pool could mine 1000 blocks in a row before another mining entity found the next block and beats them to the punch. It's simply not possible, and if it was possible it certainly isn't profitable.
We can see by this math that it's safe to run even the biggest prize pools a lottery has to offer. The payout could be hundreds of millions of dollars and the chance/profitability of hacking the outcome would still be lower than the chance of cheating lotteries that don't use Bitcoin for security (all of them).
Another example (the real example):
Say someone makes an MMORPG and piggybacks off of Bitcoin for the same reason (superior RNG security). Say the game uses Bitcoin blocks when anyone wants to craft a super rare item with super rare materials and they need to make sure it's impossible to cheat.
Congrats, you just randomed the Sword of a Thousand Truthes
On a technical level, what needs to happen here? Well, if the game is decentralized (we hope so) then multiple nodes are all running the same code. Maybe the game runs on Hive, meaning these nodes need access to a Hive node they can trust. Maybe the game runs on a Hive second layer, meaning they might have to run that second layer code as well. And finally, maybe these nodes also need access to the Bitcoin blockchain as well.
Wow, that's a lot of overhead.
Certainly, there are ways around these things and corners that can be cut. We might not have to run a Hive node; we could trust one of the many trustworthy Hive nodes out there (or redundantly confirm across multiple nodes). We might not NEED our own Bitcoin node, but wouldn't it be nice to have one? Surely it's harder to hack a network that has direct and trustworthy access to the data they are using.
This is why I like the small blocksize on Bitcoin.
Because I believe that Bitcoin nodes will start being a requirement for other networks that use Bitcoin security in unintended ways like this. If running a Bitcoin node was expensive, upkeep on one of these hybrid platforms would be really really expensive. Luckily, this isn't something we have to worry about, as the Bitcoin network is going to just keep chugging along like it has been all these years. The unchanging nature of Bitcoin code is a feature, not a bug.
Conclusion
Secure random number generation is one of the hardest things to do when it comes to blockchain gaming. Luckily, for certain functions (ones that can wait 10-20 minutes in between batches) Bitcoin has already solved this problem... and no one is taking advantage of this epic solution. It's wild. Truly, we are on the cutting edge of crypto. It's still very early in the game.
Not only that, but it's only going to get better. Bitcoin will continue going up in price. Hash power will continue to rise, and the security of the Bitcoin network is going to reach godlike levels of adoption. Most people don't realize this, but it's a great thing that Bitcoin nodes are so cheap to run. The most secure network is also going to be the least expensive to track. That's powerful.
I underestimated Bitcoin in 2017.
I don't plan on making that mistake again.
Posted Using LeoFinance Beta
They literally have attempeted my murder and are trying to kill me with V2K and RNM. Five years this has been happening to me, it started here, around people that are still here. Homeland security has done nothing at all, they are not here to protect us. Dont we pay them to stop shit like this? The NSA, CIA, FBI, Police and our Government has done nothing. Just like they did with the Havana Syndrome, nothing. Patriot Act my ass. The American government is completely incompetent. The NSA should be taken over by the military and contained Immediately for investigation. I bet we can get to the sources of V2K and RNM then. https://ecency.com/fyrstikken/@fairandbalanced/i-am-the-only-motherfucker-on-the-internet-pointing-to-a-direct-source-for-voice-to-skull-electronic-terrorism ..... https://ecency.com/gangstalking/@acousticpulses/electronic-terrorism-and-gaslighting--if-you-downvote-this-post-you-are-part-of-the-problem
Depends what the goal of the chain is. Although the blocksize for Hive is ok for now in future years it will become obsolete and too small..
Look at how much bandwidth has grown over the years and the size of data now sent. It will only get bigger in the future with VR and who knows what else will be developed.
Posted Using LeoFinance Beta
Well explained! Bitcoin cash is one of my favorite crypto currency and now I find some more information that's going on in the market right now.
SegWit method is good for scalability, otherwise bigger data chunks can slow down block generation and eventually BTC transactions would suffer a backlog.
Maybe sharding comes handy someday.
Posted Using LeoFinance Beta
Oooo, i feel the commitment in your sentiment @edicted. It's super cool to be on the cutting edge of what is becoming!! So exciting. 😁
Posted Using LeoFinance Beta
The photos is so cute!!!
Although I agree with you regarding the latent benefits of bitcoin from an RNG perspective, I can't say I agree with you on the block size issue. Sure, it's important to keep the block size manageable, but BCH's blocksize (32 MB compared to BTC's 1 MB) is still very manageable and can process 30x more transactions per second than BTC (and BSV takes that differential to 700x).
It’s manageable but has lost 99% of value compared to bitcoin. The people have chosen. It was always going to be layer 2 solutions. Just my opinion.
Sort of, but not really. If BTC had opted to increase its block size to 32 MB and Craig Wright had insisted on keeping it at 1 MB (i.e. forking the chain just to maintain the previous status quo), then BTC would still be the dominant crypto, but with a 32 MB block size instead of 1 MB, and BCH would still be pennies compared to BTC. BTC's dominance is not because of its block size, but because of its 'brand'. (My $0.02, anyways.)
Craig ? He’s such a fake.
I don't think we are in disagreement.
I've just realized there may be a benefit to the bullshit that went down.
If Bitcoin increased blocksize x4-x10 I certainly wouldn't complain.
Sorry, I guess I was a bit confused by the title: "Why I'm Glad Bitcoin Cash Lost the Hash War."
It seems to me that if Craig Wright had won his first battle (as in if he had succeeded in convincing BTC miners to switch to a 32 MB block size) then BTC would currently be the same (or stronger) as a result, and RNG capabilities would be the same or better.
Similarly, if Craig had still lost that first battle but won his second battle (i.e. if BCH had been able to dethrone BTC after the hard fork), then BCH would be as strong as BTC is now (and BTC would likely be in the toilet), but RNG capabilities would still be the same or better (on the BCH chain instead of BTC).
Or, perhaps the best outcome would've been for BTC to keep its 1 MB block size but switch to 20 second blocks. That would've resulted in a 30x increase in transaction throughput and would've increased the RNG capabilities by 30x as well.
Everything you said is accurate except for this statement.
30 times more blocks means 30 times less security and 30 times easier to hack the RNG. If we needed faster RNG at the cost of security we could just use another chain or a combination of chains to achieve this. Also I think it's a mistake to completely ignore the cost of running a node. Bigger blocks is automatically higher cost.
But yeah really in this post I'm reaching quite a bit with my logic.
Perhaps I'm just justifying what happened because we can't change it.
I don't think the security of BTC would be 30x less if 1 MB blocks were validated at 20 second intervals instead of every 10 minutes. In fact, I think it would be stronger, not weaker. Satoshi included calc's to that effect in his bitcoin whitepaper:
The "assumption that p > q " refers to the 51% assumption, meaning that the attacker controls less than 50% of the total hashpower (where p + q = 100%, p represents honest nodes and q represents coordinated dishonest nodes)
So, if the blocks are being validated more quickly (albeit at a reduced difficulty), the 'luck' required to cheat the system actually goes up, at least with respect to time.
where z is the number of blocks a recipient waits before considering a transaction valid.
So, with current BTC protocol, it is prudent to wait 3 blocks (30 minutes) before considering a transaction valid. This corresponds to a probability of 1.3%, that a determined double-spend attacker could cheat you out of that transaction, if he controlled 10% of the hashpower and was specifically targeting that block. That 98.7% 'certainty' requires waiting 30 minutes.
If blocks were validated every 20 seconds, you would achieve that probability in 1 minute, instead of 30. And, you would only need to wait 200 seconds (3.33 minutes) for that probability to reach 0.00012%.
This is from the following table in Satoshi's whitepaper (looking at z = 10 vs z = 3):
However, if the attacker controls 30% of the hashpower (q = 0.3 instead of q = 0.1), the probability of being cheated after waiting 10 blocks goes from 0.00012% to 4.2% (looking at z = 10):
So, after checking the math, I stand by my prior statement:
In fact, the improvement in RNG capabilities would go up by way more than 30x. If you still only relied on one RNG every 10 minutes, the improvement would be many orders of magnitude better. It represents the difference between z = 1 and z = 30 in Satoshi's calc's.
None of this is relevant to a lottery based off a single block.
All of this math is double-spend security.
Of course the lottery could just be rewritten to allow multiple hashes to seed it
but I feel like there must be some kind of miscommunication error here that I'm not grasping.
Still, this is some good info.
Feel like I should bookmark it or something.
nope im wrong i just thought about it some more and adding more blocks to the rng doesnt increase the security. all security for pow rng is derived from the hash power required to mine one block. if one block is 30 times easier to mine it's 30 times easier to hack. again this has nothing to do with reorg.
Seems like the lottery cheat would use the same method as the double spend cheat.
For the lottery, I create a fake block that has my fake, not random, lottery numbers and I try to pass that block off as valid in lieu of the honest block.
Or am I still misunderstanding something?
I have seen people talk about increasing the block size here on Hive. Do you think that is a good idea? I know you said it's great for BTC and I agree.
Posted Using LeoFinance Beta
I think it's a terrible idea.
Hive has proven time and time again that the blocksize is already too high.
Witnesses can't handle full blocks.
We've seen this confirmed multiple times (most recent was splinterlands bots).
This we agree 🍻
I feel like I used to see a lottery on hive that paid out based on the last numbers in of the blocks
Yes but a lottery built on Hive can be hacked by one person (the witness that mints the block).
A lottery on Bitcoin is much more scalable and secure.
I certainly was not comparing the two, merely making an anecdote
i think the worse thing was bitcoin cash losing the hash war..the reason bch lost was because institutions and venture capitalist wanted to make bitcoin a profit making machine for their business when we needed an actual functioning currency..so the world really lost for those who sophisticated enough to understand the importance of a currency
Posted Using LeoFinance Beta