Topic: Does Bitcoin become more decentralized over time? - page 3. (Read 710 times)

After certain (unknown) limits are reached, decentralization will be negatively affected by increasing the block size limit. Assuming widespread consensus, raising the max block size to 4M +1 weights will not affect decentralization. The same is also probably true if the max block size was increased to 1.25MB/5M weights. If the max block size were to be increased to (again assuming widespread consensus) say 100MB, then decentralization would very much be negatively affected. I am not sure the point in which decentralization is affected after raising the max block size to a certain amount. There are of course costs associated with raising the maximum block size, so I would not necessarily advocate for raising the max block size merely because doing so would not harm decentralization.

In terms of the number of full nodes, I think new technology, especially privacy technology is having a greater impact on the number of nodes. Users can more reliably use light wallets, including those that have privacy features, so there is not the same incentives to run a full node as there were in the past. I also get the feeling that bitcoin's user base has evolved over time, and the current user base it more concerned with utility and less with privacy, and less with maximizing security regardless of the cost, and this demographic favors light clients over full nodes.

You have 2 things decentralized in terms of nodes and decentralized in terms of mining.

For mining:
I thought it was gong to happen when the block rewards dropped down to 6.25 last year but I guess it's not going to happen till 3.125 or 1.5625 when a lot of the larger pools start to shut down. Unless the BTC price goes above $100k USD and stays there it's going to cost more to run big pools then they can generate. So I think a lot of smaller pools will start to pop up. I also think a lot of miners will drop out since it's going to be harder to make money. Once again that's if the price stays where it is, $100k or $200k BTC and all bets are off.

For nodes: Drive's are falling in price, 5g is coming online in a lot of places, starlink and Project Kuiper are going to bring high speed internet to a lot of places. I think once that happens, when you can get a RPi type box with 2 TB of storage you are going to see a lot more project nodes and things come online all over the place.

Just my opinions, I could be wrong :-)

-Dave

This came through my mind so many times before, yet all I could think of was the factors you've mentioned. Besides the growing number of full nodes (which is probably going to keep lowering down in years until storage goes cheaper due to the quickly increasing blockchain size), hashrate etc.. do we have anything else that makes it more decentralized?

Something we may not be thinking about is development and products that could help with the decentralization. For example, doesn't goTenna help with it by taking the Bitcoin network into a mesh network rather than an internet service provided data? Does that count? To me, I think it does..

On the other hand though, we have exchanges and custodial wallets (or wallets that aren't even truly wallets such as PayPal or Revolut) that will likely have a growing mass of customers coming in and preferring custodial wallets over non-custodial ones for convenience. Doesn't that basically decrease decentralization since we put our trust into a single entity?

Bitcoin Core development-wise, I'm not sure we've truly had any improvement in the past few years at least. I think we've reached the current peak of decentralization as we know it, considering the current technology and knowledge we and experts have. The best thing we can do next is probably just moving on and trying to go even more off-grid by taking BTC off the centralized internet.

But, if the blockchain size is not very large like almost 400 megabyte now, many more people will run it in a way there will be more nodes and decentralization.

Oh, it is easy:
Let n be the cardinality of the smallest set of actors that are potentially able to collide for committing a successful adversarial behavior against any subset of users by somehow violating their rights for joining/continuing/exiting the network deliberately and safely. We can trivially define a decentralization index as:
DI = 1 - n^-1

The concept probably won't have any suitable metrics to be used as a gauge so I don't think there is any absolute answer to this.

With the growth of Bitcoin, you're going to inevitably eliminate the smaller scale miners who can't afford to keep their costs down. It would result in having miners probably being centralized around certain regions and farms. Mining would probably never be any more decentralized.

As for the full node count, I suspect that some nodes actively block Bitnodes crawlers as they publicly list their IPs and thus resulting in a much lower node count. Running a full node is obviously not so viable for most people; even in 2021 360GB+ is fairly significant if you consider that some laptops and/or desktops only ship with 1TB of storage. That, combined with the very few advantages over SPV (which I assume most actually wouldn't consider), will probably reduce the full node count over time. I can't see the point of most people running Bitcoin Core when using Electrum or Wasabi would suffice for their use case while wasting no time to synchronize whenever they're using it.

OP,
I read you as follows:
Given the argument about small blocks helping significantly with decentralization is correct, we expect a constant improvement in the hypothetical decentralization index of the network due to the constant hardware improvement trend that is expressed in the Moore's law. Is it actually happening?

Answer:
No, it is not. But it doesn't make the initial argument false, because it doesn't imply that it is enough to keep block size under control and everything goes well. Other measures should've been taken as well, decentralization of pools and support for light clients (not SPVs) being the most notable ones.

Unfortunately, we are facing an exaggerated "nobody is in charge" situation in bitcoin, where a strength is proving to be a weakness at the same time. I hate what is going on in the Ethereum camp, the stupid "road map", the "Foundation" and the "genius",  but the lack of any sign of responsibility in Bitcoin is concerning too.

Real indicator for more decentralization should be growing number of full Bitcoin nodes, but I don't think we can have real accurate data for that.
Some say there are around 100,000 nodes, than there is Luke's script showing there are 75260 nodes at the moment, and bitnodes website is showing around 10380 nodes.

I don't think blockchain size should be such a big issue anymore as disc prices are going down and capacity is going up very fast, so we can expect more people will run full nodes and only this can make Bitcoin more decentralized over time.
Other issue is that I think most people don't even understand importance of running a full node so they never even tried running one, but we surely can't expect newbies running full nodes.

The biggest contributor to decentralization is "more people joining in". That has been happening with price rise and they are not all 100% speculators wanting to make quick money. We get more of everything, more people running full nodes, more people buying ASICs and becoming miners, maybe even more mining pools (a more distributed hashrate), more developers/contributors to reference implementation and a lot more of other useful open source projects.
All of that makes bitcoin more decentralized over time.

As for block(chain) size, we already have a lot of options that make things easier for different type of users. For example pruning for those who want to run a full node but don't have enough storage to keep historical blocks. And the IBD which is the slowest step that people usually complain about should be performed only once, the daily network and hardware usage could go up but there is also options to reduce those too (smaller mempool size, higher fee/byte min relay fee, not relaying anything, ...).
Considering the fact that we have 70k to 80k full nodes with 10% listening for incoming connection I'd say things are looking good. With the price rising, I don't think node operators mind a slightly bigger cost of running a full node considering the benefits bitcoin and running a full node for it is providing them.

In my opinion, bitcoin is decentralized, what makes some people do comment not to be decentralized is the issue that more miners are mining in China, but as 75% miners were mining in China in 2019, this reduced to 65% in 2020. And with recent news around the world, there is more chance that miners in percentage from China will still reduce this year. And that does not make bitcoin decentralized as some people thought.

About this topic, about the issue of scalability, before was only 1MB for block size, but the introduction of segwit make use of weight which is 4M weight units which is around 2.3MB in size of all trrasactions from segwit. This is a sign that there can be more approaches in the future that will enhance bitcoin transaction scaling that will get low fee transaction confirmed early.

Another is the lightning network, this is another alternative solution just to make the scaling issue solved. Developers will look for more ways to make bitcoin remain the possible best  decentralized coin existing.

Another approach is the taproot and the use of schnorr's signature that will significantly help in multisig transactions.

Some people says raising Bitcoin block size hurt decentralization because it increases cost to run a full node. Bitcoin block size limit remains at 1MB 4 million weight units in past few years, while hardware and internet continue growing. The growth could be higher speed, higher efficiency or lower cost. With that in mind and ignoring other factor (such as blockchain size growth, hashrate distribution and total of full nodes), does Bitcoin become more decentralized over time?

P.S. This is just my shower thought, so obviously it has flaws.