Getting more node participation

[Steve]

In that other thread...Jeff commented that Bitcoin could not withstand a targeted cyber attack.  I expected him to talk about polluting the block chain or a DOS involving lots of spammy transactions.  But instead he seemed to imply we just need more full nodes and miners.  On the miner side, I assume that Bitcoin probably already has sufficient power, but perhaps it needs more diversity (something I expect cheap and easy to setup ASICs will help with).  In terms of full nodes, I would think a few thousand nodes would be sufficient to thwart a large scale DDOS attack (most DDOS attacks are successful because they target a very small set of servers...if you have to start targeting thousands, I think they would be much less effective).  However, if we want to encourage more people to run full nodes, here's an idea. 

Make a full node that is designed to serve multiple web based wallets (with keys never stored on the server except in encrypted form, etc).  Web based wallets are convenient but unsafe, even the ones that never expose keys to the server (the service provider could always change the Javascript code at any time).  For this reason, I'm reluctant to recommend such services to my friends and family.  I'm also reluctant to recommend they run a full node.  I personally want the convenience of a web based wallet, but I also want to run a full node.  I want to run the full node on a server I control that's only indirectly connected to the broad Bitcoin network through a grid of other full nodes.  I want to connect with a web browser to that node for my day to day wallet usage.  And, I want this to be the wallet I recommend to friends and family (that way I don't have to ask that they place trust in someone else that they, nor I, might not know that well).

I think this model would be preferable to having just a few large web based wallets that everyone uses.  It would strengthen the network because many of us would operate such nodes for the benefit of our circles of family and friends and it would reduce the dependence on the few web based wallets that currently exist.

[gweedo]

I think one huge problem also is people have no idea what a full node means. I saw 2 threads, this past week that both asked "What is a full node" and "How do I start running a full node" and they both were running full nodes. Also I think we need to start telling people to start with the bitcoin-qt and bitcoind then move into web wallets. Then just starting off with web wallets or light clients.

Another huge problem is scaling, there are some threads that shows that bitcoind doesn't scale very well when you start getting into these huge numbers, and when some bigger companies are going to want to start entering bitcoin world. They will learn very quickly they just have to get bigger servers to have the bitcoind work correctly. This is a turn off to run it.

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out. Cause if wordpress was made to run a bitcoind, I am pretty sure they have some good pipes could have been a good full node.

[gmaxwell]

I think one huge problem also is people have no idea what a full node means. I saw 2 threads, this past week that both asked "What is a full node" and "How do I start running a full node" and they both were running full nodes. Also I think we need to start telling people to start with the bitcoin-qt and bitcoind then move into web wallets. Then just starting off with web wallets or light clients.

There are enormous speedups happening right now which might stem some of the alarming movement to centralized clients... but to really close the gap the software will need to be able to be instantly usable and sync in the background, and thats going to take more work.

We're development resource constrained all around and especially testing constrained.  Getting more of the geekier parts of the community running nodes with the bleeding edge code (including testnet and secondary nodes; to keep their real funds off the experimental software) and teaching them how to beat it up and produce good bug reports would help the software mature faster and help get the improvements into the hands of new users.

[Steve]

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out.

That is not at all our business model.  Strengthening the Bitcoin network in every way possible is in everyone's interest, including ours.

[gweedo]

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out.

That is not at all our business model.  Strengthening the Bitcoin network in every way possible is in everyone's interest, including ours.

You do make it so people/companies don't have to run a full node, so you are contributing to this issue. The network needs companies like yours making it easy for some people, but I disagree companies like wordpress and reddit using your company those are two companies that could really strengthen the network by running a bitcoind as the backend.

[Steve]

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out.

That is not at all our business model.  Strengthening the Bitcoin network in every way possible is in everyone's interest, including ours.

You do make it so people/companies don't have to run a full node, so you are contributing to this issue. The network needs companies like yours making it easy for some people, but I disagree companies like wordpress and reddit using your company those are two companies that could really strengthen the network by running a bitcoind as the backend.

We actually evaluated setting up WordPress with a completely separate set of servers and software (including their own cluster of bitcoin nodes).  It's just a bit premature for it.

[jgarzik]

We actually evaluated setting up WordPress with a completely separate set of servers and software (including their own cluster of bitcoin nodes).  It's just a bit premature for it.

My standard advice for merchants:  run at least two nodes:

• Node with empty wallet, connected to Internet
• Node with wallet, handles all your bitcoin activity, connects only to the other node

Large merchants would be well advised to run multiple empty-wallet nodes, connecting their internal nodes only to those semi-trusted nodes they control.

(Note: free business idea!)

Another option is a semi-trusted "backbone"  This is a project I worked on for a first, but did not have the time to build it into a real business.  Run a set of nodes, and permit your merchant-customers to connect to these nodes.  In an ideal design, you would have multiple layers of security:  merchants connect to an internal backbone ("cloud A"), and the internal backbone connects to a DMZ/public backbone ("cloud B") that talks to the Internet.

You would have to consider the possibility of a rogue merchant on the internal backbone, but any payment business in theory does a bit of customer vetting.  And even so, it's still got the standard bitcoin level of security... full block and TX verification, etc.

[bc]

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out.

That is not at all our business model.  Strengthening the Bitcoin network in every way possible is in everyone's interest, including ours.

You do make it so people/companies don't have to run a full node, so you are contributing to this issue. The network needs companies like yours making it easy for some people, but I disagree companies like wordpress and reddit using your company those are two companies that could really strengthen the network by running a bitcoind as the backend.

I doubt Bitpay is preventing Wordpress from doing anything. Maybe they'd like to see a few bitcoins come in before they invest more.

Regardless, Wordpress acceptance of bitcoin dwarfs their potential contribution to network health.

[Raoul Duke]

Stratum/Electrum

Only need to solve the web-based wallet part.

[gweedo]

Also I have to call you out on something, you do run a service that cuts out the need of a bitcoind at all. It is ironic you want more node participation when your whole business model is cutting that setup out.

That is not at all our business model.  Strengthening the Bitcoin network in every way possible is in everyone's interest, including ours.

You do make it so people/companies don't have to run a full node, so you are contributing to this issue. The network needs companies like yours making it easy for some people, but I disagree companies like wordpress and reddit using your company those are two companies that could really strengthen the network by running a bitcoind as the backend.

I doubt Bitpay is preventing Wordpress from doing anything. Maybe they'd like to see a few bitcoins come in before they invest more.

Regardless, Wordpress acceptance of bitcoin dwarfs their potential contribution to network health.

So the health of the network is less than a PR stunt? Yes think about that one. LOL The health network should be first in everyone's mind.

[Atruk]

Stratum/Electrum

Only need to solve the web-based wallet part.

I think a great way to encourage people running nodes is by making it easy to connect a feature rich and network light mobile wallet to you own personal electrum-type server. That way your phone, tablet, or frequently disconnected because you drag it everywhere laptop can have an instantly available wallet without trusting third party servers.

The electrum server and stratum are freely available, but it would be nice if some of the bigger bitcoin news and information sites put together a guide for the mobile bitcoin user doing business in person.

[cunicula]

Stratum/Electrum

Only need to solve the web-based wallet part.

I think a great way to encourage people running nodes is by making it easy to connect a feature rich and network light mobile wallet to you own personal electrum-type server. That way your phone, tablet, or frequently disconnected because you drag it everywhere laptop can have an instantly available wallet without trusting third party servers.

The electrum server and stratum are freely available, but it would be nice if some of the bigger bitcoin news and information sites put together a guide for the mobile bitcoin user doing business in person.

This remote computing seems like a good idea. Mobile Phone stores private keys and connects only to home computer. Home computer handles blockchain, node, etc., but not wallet.