Bitcoin Forum

I am a fan of Bitcoin, I believe that to maintain the robustness of the Bitcoin network, independent and distinct full-node clients are particularly important. Imagine how horrible it would be if all the nodes of Bitcoin used the same software, the same version. Can Bitcoin still talk about decentralization?I noticed that libbitcoin, a standalone full-node client for Bitcoin, now has only one node running. I have a cloud server that runs bitcoin full node all year round, running bitcoin core software, and would like to switch to "libbitcoin". My cloud server is Windows operating system. I downloaded the related software for libbitcoin, but it does not work, prompting me to initialize it first. I still can't run this client after a lot of effort, is anyone willing to help me?

Running the same software is not a problem, as long as the consensus rules that this software are decentralized.  Nobody really has the power to chance those consensus rules alone.

And there are also dozens of different bitcoin nodes, not only core. There are many spv wallets, prune nodes,  other full nodes implementations, etc...

Why do you want to run libbitcoin?

If you want to run a full node, you should run bitcoin core which is the most used one, unless you really need something else.
https://bitcoincore.org/en/download

 bitcoin is the most decentralized cryptocurrency that exists and there are thousands of full nodes running 24/7

Creating a full-node is very hard work which is akin to maintaining a large enterprise software repository. That's probably why you don't see that many Bitcoin clients popping up everywhere. Unlike, say, Ethereum, where it is companies who are the ones maintaining all the different reference clients.

But if you're still looking for an alternate full node client, you can use Bitcoin Knots (by luke dash-jr). I cannot vouch for it though, as I have never used it.

Agreed. If we couldn't trust Bitcoin Core, we'd have bigger problems to worry about. And even if the developers ever go evil and add bad stuff, we'll just stick to the older version and go from there.

The problem is not just trust. If everyone is running the same software then there is a risk of a serious bug that could take down the entire network. With multiple implementations, that risk is limited.

You mean bugs like back in 2010, when someone created 184 Bitcoins? Different implementations might indeed reject those blocks instantly, but if any of the implementations would have a share >50%, it still wouldn't stop it. One way or another, it would need to be fixed and trigger a chain reorganization for anyone who's on the faulty chain, and don't think it's an improvement if different people see different blocks as the most recent ones.

Still, in this scenario, that only applies to the pool that rejected the blocks. It would basically be a fork, and anyone on the wrong end would still need a reorg.

Not just that kind of bugs. The robustness, as he says. See what happened with LND last year. Twitter user Burak did a 998-of-999 Tapscript multi-sig (https://twitter.com/brqgoo/status/1579216353780957185) and crushed all LND nodes. The lightning network continued running, because other implementations (such as c-lightning) were existent. Sure, in terms of node connectivity, the network was damaged, but it didn't stop, and that's valuable if we're valuing decentralization.

Excuse me, which libbitcoin-server you used for guide you've written? I've tried libbitcoin in past (with version 4.X) and steps you mentioned is very different compared with what i did[1] and libbitcoin-server guide[2]. For example, the guide use binary bs and config file bs.cfg[4].

[1] https://bitcointalk.org/index.php?topic=5329445.msg56771929#msg56771929
[2] https://github.com/libbitcoin/libbitcoin-server/wiki/Quick-Start (https://github.com/libbitcoin/libbitcoin-server/wiki/Quick-Start)
[3] https://github.com/libbitcoin/libbitcoin-server/wiki (https://github.com/libbitcoin/libbitcoin-server/wiki)
[4] https://github.com/libbitcoin/libbitcoin-server/wiki/Configuration-Settings (https://github.com/libbitcoin/libbitcoin-server/wiki/Configuration-Settings)

It's better not to respond to those AI spammers!

https://i.postimg.cc/DfYjVr48/AI-Written.png

thank you

Is there any information to back this? Are full node operators actually in their thousands? I have always has the fear that the number of people running full node is reducing drastically.

This is a good question.

Yes, there are tens of thousands or more.

According to bitstamp (which is the oldest exchange still in operation  quite reliable imo), there are even many nodes which are not reachedable 

Roughly 16K, at least according to BitNodes though there isn't an accurate methodology to prove this; their crawler sends getaddr to nodes to get a list of IPs which are running a full node and the realtime number can vary drastically. Anyhow, the number shouldn't matter but the geographical distribution is by far more important.

A vast majority of full nodes are Bitcoin Core and for a good reason as well. They are generally well tested and each commit is vetted fairly rigorously and the likelihood of a severe bug or an exploit is unlikely. Though that has happened in the past, most of it are not as severe or easy to execute as the development matures.

Thanks for the information.  Both are pointing towards same direction. There is atleast 10k active full nodes. According to the quote provided by bitmover, the number could be higher and I believe it is higher.

Let's assume there are 10k active full nodes, and 85 percent of these nodes are concentrated in one country. What is the implication?

• Will it mean any form of centralization?
• I think the major concern should be if the government of that country rises against bitcoin mining

Centralization and redundancy goes hand in hand. If for some reason, a certain country experiences a natural disaster or any form of attack that disrupts the internet connection, electricity, etc. You would see a pretty big drop in node count. It wouldn't pose as a critical threat but it is certainly not what we want.

Bitcoin mining and running a Bitcoin node is quite different. Latter is harder to crackdown.

Here are link to up-to-date number,
https://bitnodes.io/ (https://bitnodes.io/)
https://luke.dashjr.org/programs/bitcoin/files/charts/software.html (https://luke.dashjr.org/programs/bitcoin/files/charts/software.html)

Although it's true the number is far lower from decade ago.

Hm? IIRC, the numbers that we have today is much higher than when I first discovered Bitcoin (in the range of 7-9k) and Bitnodes was running a campaign to encourage people to run full nodes. The only problem is that there is far higher proportions of nodes on datacenter ASNs than before.

Beautiful map.

I think we have a nice geographical decentralization of bitcoin nodes. We don`t have as much nodes in developing countries because fewer people can afford one there.

But developed countries have nice distribution.

https://www.talkimg.com/images/2023/07/12/Zk8BG.png
source: https://bitnodes.io/

May I know the statistic for this? As far as I can tell, according to Bitnodes the last few years never surpass the current number of nodes. Well, technically the highest is around 17k according to them. I can't find the data for the last decade (back to 2013, for example)[1]. Other sources also seem to give the same picture[2]. CMIIW.

[1] https://bitnodes.io/dashboard/7y/
[2] https://coin.dance/nodes

More than half of the nodes are onion nodes, so in terms of clearnet decentralization, not so much.

If you check the ASNs that the nodes are on, you will find that the overwhelming majority are datacenter related. There isn't a lot of people running it at home or on a different network which still introduces a centralized point of failure to some degree.

Ah fair enough, I remember seeing this sometime ago.

I think the disparity is pretty big as compared to the other statistics that you find around and the accuracy can be quite skewed, even today (4k vs 6.4k). That is taking into account that you're only looking at the listening nodes. Anyhow, he has never shared the methodology and the last time I checked, he preferred to keep it a secret so I can't really vouch or check for the accuracy.

Noted, but in the case of redundancy, the glory would be in the pass because at present, it might not be in use. I understand we should only be afraid of attacks or natural disaster if the nodes are concentrated in one region. So, this has nothing to do with centralization in terms of decision making.

Thank you bitmover for this pictorial highlight. It shows that node operators are evenly distributed across the world. Maybe it is not time for Africa yet, because there is relative small number of node operators compared to the size of the continent.