There are plenty of reasons why the Lightning Network is great and is considered one of the best things that happened to Bitcoin in years. First of all, “unfairly cheap”, instant, and more private transactions are finally possible in a scalable way. Secondly, riding the Lightning also requires running a Bitcoin node, which strengthens the security of the network and creates a greater incentive to keep the device running. Thirdly, running the said node finally pays, as routing payments through your channels adds small amounts of satoshis to your wallet. This may not be much yet, but in the future it can at least cover the electricity costs.
However, the onboarding delay is still pretty inconvenient for many users, as the amount of time required to synchronize a full node is a deterrent factor that may stand in the way of efficiency. The risk of this conservative model is that newbies will simply use custodial wallet solutions like Tippin.me, Wallet of Satoshi, or Bitrefill’s rented channel opening. As opposed to waiting for longer than a day for a computer to synchronize the entire Bitcoin blockchain (which is as large as 209 gigabytes at press time), impatient users will simply take the shortcut and think that they’re using the Lightning Network while actually accessing the services of a centralized company.
In this context, Node Launcher lead developer Pierre Rochard expressed his advocacy for BIP 157 and its merging into the Bitcoin Core client. With this addition, users will be able to operate non-custodial Lightning wallets in a matter of seconds, and then experience the truly fast nature of the network. It also enables a convenient way of running clients on mobile devices, which has the potential to exponentially increase adoption.
On the other hand, there are vocal critics who remain conservative and express concerns on the incentive model and security.
What is BIP 157?
In a nutshell, BIP 157 (officially titled “Client Side Block Filtering”) is a 2017 research project conducted by Bitcoin developers Olaoluwa Osuntokun, Alex Akselrod, and Jim Posen. It seeks to improve upon the light client model introduced in BIP 37 and removes many of the security concerns associated with Mike Hearn’s previous work. Unlike previous attempts, this proposed implementation is more robust and requires a lower amount of trust.
For context, CoinJoin light wallet implementation, Wasabi Wallet, also uses BIP 157 in order to facilitate a secure, private, and trustworthy way of mixing your bitcoins that doesn’t require downloading the entire blockchain. While the developers of Wasabi did include a way for the client to get connected to the user’s full node, the operation is not necessary and can be regarded as an extra step for precaution.
Now, one can argue that there is a big difference between intending to onboard millions of people on the Lightning Network through a light client, and serving the few thousands who want to increase the fungibility of their coins. There is an issue of scale, and the biggest problem that may emerge concerns the idea of running a node: why would anyone consume electricity and make use of over 200 gigabytes of storage when the shortcut is way more accessible?
Neutrino, despite being a potential trigger of mainstream adoption, is problematic for the incentive system. Also, as the next section is about to present, some bitcoiners are concerned about the security model too and suggest that every light client should ideally be accompanied by a full node – so you ride the Lightning within minutes, but at the same time leave your computer on to finish synchronization for the sake of your security.
The pros and cons of Neutrino-enabled light wallets
Node Launcher lead engineer Pierre Rochard stands out as the most vocal of BIP 157 advocates, and he seems to be convinced that Lightning should live up to its name from the first minute. A Neutrino-enabled light wallet would turn the sync days into mere minutes, which means that mainstream adoption can potentially become a rapid and feasible phenomenon. With the right user experience and enough applications published on all major app stores, Lightning might really take off.
In regards to the issue of bypassing the full node requirement, Mr. Rochard presents BIP 157 as a preliminary method or “gateway drug”. Essentially, the implementation is all about convenience and the first contact with the technology, so anyone can test Lightning and see its capabilities without the initial long sync-driven delay. At a later point, a full node can be engaged to complete all security requirements.
With the Node Launcher, there's no tradeoff with decentralization, as LND uses #BIP157 for rapid on-boarding while your Bitcoin full node takes hours/days/weeks to sync in the background. Once your Bitcoin full node is ready, #BIP157 usage stops and you use your Bitcoin full node
— Pierrrrrrrrrrrrre Rrrrrrrrrrrrocharrrrrrrrrrrrrrrd (@pierre_rochard) March 10, 2019
However, the big issue with BIP 157 – and particularly its implementation of Neutrino – is to be found in the security model. Bitcoiner Nicolas Dorier has taken his time to outline his criticism on the proposal. In his article “Why I don’t celebrate Neutrino”, he identified four different systemic risks: trusting a random third party node to not lie, appealing to multiple random nodes on the network to make sure that the block in question exists, relying on the block with the most Proof of Work, or using your own full node.
Ideally, all Neutrino light clients should be accompanied by corresponding validating nodes. However, assuming that users who find absolute convenience in their interaction with the Lightning Network will eventually consider maximizing the security implies a lot of wishful thinking. In the worst case scenario, too much trust is going to be vested in the miners, and we might end up in a situation where malevolent, unchecked actors damage the entire system.
From a light client point of view, it makes most sense to trust one or more random third party nodes. But given the premise of acquiring financial sovereignty through a trustless system that puts the used in charge, there is a strong ideological divide between accessibility at all costs and a radical view on security. Nicolas Dorier goes through the entire game theory by analyzing all possible outcomes and the damage they cause.
In the end, he concludes that merging Neutrino in Bitcoin Core is useless for its established purpose of running light clients on mobile devices, and suggests workarounds involving more conservative ways. The last phrase of the article, which basically says that Samson Mow’s hats won’t save the protocol from B3X (a random name for another miner-driven contentious fork), he strengthens his view that long-term security should reign supreme.
Community reactions: toxic maximalism or cautious conservatism?
As previously stated, one of the biggest virtues of the Lightning Network is that of increasing the number of full nodes being run across the world. Currently, all users have to choose between a custodial method (trusting a third party with their Lightning transactions) and running a full validating node. Maybe that Neutrino aims to specifically eliminate the centralized services by creating a convenient light client which has the potential to earn all the feats of decentralization and financial sovereignty.
However, most bitcoiners who have voiced their opinions in relation to BIP 157 and Neutrino didn’t seem to be thrilled with the proposal of merging the code into Bitcoin Core. On the contrary, many criticisms have been addressed to tackle the uncertainties brought about by such an implementation.
For instance, Italian Bitcoin maximalist Giacomo Zucco has introduced the issue of poor timing, given the increasing number of full validating nodes running specifically to grant Lightning access.
I really hope this will not pass in Core, and that if it does people will move to a better implementation without it. Incentivizing people for not running Bitcoin nodes is the worst thing to be done now.
— Giacomo Zucco (I have to change my pronouns) (@giacomozucco) March 10, 2019
Cypherpunk Peter Todd, on the other hand, has referenced the big picture from which such proposals stem: SPV (or Simplified Payment Verification) which is the 8th chapter of Satoshi Nakamoto’s final draft of “Bitcoin: A Peer to Peer Electronic Cash System“. He basically argues that simplifying the consensus method is a terrible idea for the checks and balances of the network, and purposely positions Neutrino as one element that originates from this long line of attempts to choose convenience over security.
SPV was a massive political and technical mistake.
— Peter Todd (@peterktodd) March 11, 2019
Furthermore, Nicolas Dorier outlined his pessimistic views on BIP 157 by implying that such a change to the consensus protocol is a potential gateway to another SegWit2X that’s driven by the miners. In its current state, the Bitcoin protocol is more balanced from a political point of view and makes sure that all powers are kept in check.
Yes. He assume that the longest chain is valid. BIP157 increase the chance the the chain will be invalid, as miners might be tempted to do a B2X second edition.
— Nicolas Dorier (@NicolasDorier) March 11, 2019
At press time, the result of these debates is inconclusive, as both parties stand their ground. However, as time passes and there is a greater requirement for easy Lightning onboarding, it’s very likely that the issue will be brought again with greater support. It’s also probable that mining pools would support the idea just because Neutrino grants them more political power in the event that light client users never switch to a full node solution.
Yet regardless of the fate of BIP 157, this is an important chapter in the history of Bitcoin, and one that once again puts the resilience of the protocol to the test. This time, there doesn’t seem to be any bad actor involved, and it is mostly a clash between idealism and conservatism. All claims can be valid in a given context, as human behavior is unpredictable and it’s impossible to tell how Lightning users will react when offered the chance to bypass the full node requirement.
In all technological sectors where innovation occurs, the urgency of popularizing a new discovery often clashes with the cautiousness of those who oppose radical change and favor time-tested solutions. And in the case of Neutrino and BIP 157, this might just be the beginning of a new political battle for Bitcoin.
Crypto Insider has contacted Nicolas Dorier, Giacomo Zucco, and Pierre Rochard in order to further summarize their views in relation to Neutrino and BIP 157.
At press time, Mr. Rochard has responded by pointing out to the tweet at the top of the article and this one as being representative for his views. Furthermore, he mentioned that Node Launcher will only connect to his Neutrino server and not a random (and potentially malevolent) one.
Also, Mr. Dorier has stated the following:
In a nutshell, SPV will be one of the biggest threats to my sovereignty if it becomes too popular. If you are running a SPV you are contributing a little bit to the destruction of my own sovereignty 10 years down the road, and I hate you for this. Use an Explorer wallet or keep your money on an exchange.
Yes, I know you don’t care about my self-sovereignty as much as you care about your own money. You don’t want your money to be hacked, seized, and data sold. I get it. I know that you somehow feel more in control by using SPV and owning your keys, and I know that my future self will have to pay for your behavior if I do nothing against it.