Topic: Time to boost Bitcoin circulation, Million Transactions Per Second, more privacy (Read 314 times)

After introducing Sabu protocol as a solution for Bitcoin scaling (https://raymo-49157.medium.com/time-to-boost-bitcoin-circulation-million-transactions-per-second-and-privacy-1eef8568d180), I shared this idea with Bitcoin developers through the bitcoin-dev mailing list.
I got some constructive feedbacks and critiques leading me to add this part to the proposal which I was skipped due to brevity of proposal introduction.

Here I will investigate on more real live scenarios, general usages and corner cases, and the consequences of some attacks or buggy implementation of protocol, as well as different actors (malicious, irrational, profit seeker, griefer, stupid, reckless, incompetent, etc.) activity effects.

In proposal introduction (previous post), I did not talk about Lightning deliberately, although it seems that this solution is an alternative to Lightning.
Most of readers misunderstood Sabu and asking what differs it from Lightning?
Indeed, Sabu has nothing with Lightning. It has totally different design, network architecture, security model and implementation. The only thing in common with Lightning is both are intended to cover micro payments.
The good thing about Bitcoin is that it does not require any kind of permission. Consequently, related products do not need to ask permission too. We are in a permission-less free market. I think Sabu will work perfectly and if a group of users think like me, we are done. Sabu will work parallel the other scaling solutions without need to drive them out.
However, I have made a comparison between Sabu, on-chain and Lightning transactions to get a clearer understanding of the advantages and disadvantages of Sabu and answer to “why we should implement and use Sabu in our day-to-day deals”.
Most probably this paper is not comprehensive document, therefore this article will be updated.

read here for complete post:
https://raymo-49157.medium.com/scaling-bitcoin-by-sabu-protocol-risks-and-benefits-62157f8a664e

You don't seem to be familiar with the Lightning Network, it already solves the problems you're trying to solve, already has a large user base, and it much more secure than your proposal.
If you already know the Lightning Network, why do you think your proposal is an improvement?

Your protocol has many security flaws but what baffles me the most is that you want to rely on mobile wallets. What happens if my issuer is offline? I can't spend my bitcoins anymore (unless I go onchain which is what we're trying to avoid here).

Please use the terms "Issuer" and "Creditor" to avoid misunderstanding.
The issuer is Bitcoin owner who can issue a valid Bitcoin transaction and inside the transaction promise an output to creditor.
on the other hand the creditor is a person who accept a transaction as a kind of debt-document, and do not send it to Bitcoin network. like you have a valid check from me but you do not go to bank to cash it. instead you spend my check directly and give it to the other person.

So when Alice wants to send 1 BTC to Bob, Bob has to risk losing 1 BTC provided for the Guarantee Transaction? (Alice might be rich and malicious.)

So I have to pay 25% in fees to avoid fees on the chain?
And not only that I have to deposit double what I will be allowed to spend?
Where do those fees suddenly pop out from?

This while I can open an LN channel for a few satoshis right now with nobody taking half of my money and only guarantee me through a 3rd party system I will see them back, without being tied to some measly 20k satoshi but mBTC and pay 0.01% in the fees I create there?


Nope, it's not simple at all.

The average Joe has to install another app on top of his wallet, he must learn how to create those transactions, he must create and link an email, he must then switch to the other protocol and spend those, and I will repeat TOKENS, cause that's what they are tokens, and do this each time for 20k satoshi, 7$ at the current price. That's why I'm telling you don't understand the average customer, they don't want this, they don't want to deal even with electrum or bitcoin core, they don't want addresses, just a QR code, they don't care about satoshi but what that means in USD. Look how hard for some to adapt to LN, most will switch to solutions like a strike or blue wallet that do all the complicated stuff for them and you're coming with a solution three times more complicated than LN.

By this rationality even using pure TCP/IP is a centralized solution!
Please note the fact that decentralization can be interpreted in different depth of abstraction. By the way if the email bothered you, as you mentioned the wallet can use Matrix, Nym network, Onion, I2P, classic central servers,... these are all communication mean and the wallet can simply use one or more in parallel.
Although I still believe email is a perfect choice, and in future we will see the wallet users will choose which solution.

If you intended a malisciuos "issuer" as an attacker, your asumption is wrong, since in Sabu transaction standards Issuer can sign maximum 20,000 Satoshi as his debt and for this transaction the issuer must spend a UTXO worth at leaset 40,000 Satoshi. Each Main Transaction(MT) has to pay fixed 10,000 Satoshi as Bitcoin transaction fee despite the length of transaction or amount of transaction. So your cheating scenario fails. It is almost what you sugessted
please study the images (6. Aggregate credits and 3. Transaction in detail).
But! who is going to entrust you with twice the amount of bitcoins just to be able to spend on that amount in a token

Please re-read the proposal main idea. There is no "token". it is just Bitcoin and valid Bitcoin transactions. The only thing is different is the amount of outputs. Issuers sign transactions and earn very small fees(10 Satoshi per transaction) but they can handle thousands of transactions per day and by this earn a significant income which incentivize them to use this system, on the other hand the creditors can buy (and transfer) small amount of Bitcoin(even under 5$) with no dependency to exchanges or KYC or high transaction fee. It would be a mechanism for small saving and once the creditor has enough funds(near one or thousands of different issuers), he can transfer it to Bitcoin blockchain.

The email service cost will be 20$ per year or less. Meanwhile Email can be replaced (or pralelly exist) by central servers or some peer-to-peer IP based system. It is only a communication mechanism which I still prefer email because I have a particular point of view about this project and its future features. So some third parties can easily provide the communicate service for wallet.

The system must be super user friendly, the average joe will not aware of all these complexities. He know his friends email and add them to his wallet and starts to send and receive funds. All these complexities are behind the scene and on our technical shoulder. We are here to make the things make the life easier for people.

You simply keep denying the obvious!

The attacker brings 10 000 satoshis to the table, he receives the same amount in SABU which he spends, then he issues a transaction with a 9500 satoshi fee, which your GT model is unable to overcome. What's the end result?
Attacker 10 000 satoshis in goods bought with SABU, 500 satoshis saved from that UTXO. He is not losing money he is gaining it!
The creditor has -10 000 satoshi in Sabu and..nothing else!
Miners enjoy those fees which make the whole scheme unfold as fast as possible.

The only way you would even be able to make this bulletproof would be to go the way defi loans are created when the users would guarantee with 40 000 satoshis per example but receive only 20 000 satoshis worth of tokens to spend in your network, making the collateral a barrier to profit from the scheme. But! who is going to entrust you with twice the amount of bitcoins just to be able to spend on that amount in a token nobody has heard of when there are hundreds of other solutions out there far cheaper, he could simply switch to a damn altcoin and keep half of the money safely in his pocket not in some really really complicated scheme.


So either the risk of google disabling my accounts as it doesn't like those mails or pay for a paid email service that would cost me more than the fees. 
You're seriously downplaying the disadvantages of your system, I understand that you're trying to defend your model but you should really look at it from the perspective of the customer! What would the average Joe think of this?

I have to correct your hypothesis, then explain the scam scenarios.
Lets say Bob owns 1 Bitcoin and sign a valid transaction and send it through the Sabu protocol to Alice. It is a private and encrypted communication and no one will be aware of this agreement.
Now Bob has the fiat money (or goods or services) of Alice, and Alice has a valid signed Bitcoin transaction in her hand. Note that there is nothing like Sabu token! Both Bob or Alice have this transaction and both are able to send it to Bitcoin network immediately (of course if they will).

Scam scenario 1: Bob spends the promised UTXO (a part of this UTXO is promised to Alice) and "wants" to buy something. He spend entire UTXO and pay it to someone else, he signs and sends the cheating transaction to Bitcoin network. And now the initial valid tx Alice had is no longer valid.
As I mentioned in paper, in order to prevent this scenario, the Sabu protocol initially forces Bob to prepare and sign 2 transaction. One is Main transaction(MT) which contains the just balance and the second is the Guarantee transaction(GT) in which both Alice and Bob will lose a portion of their coins in favor of miner fee. If Bob deny the initial agreement and spend the promised UTXO, Alice will be aware of spending same UTXO in few seconds (since Bob has to broadcast new cheating transaction to Bitcoin network, and Alice Wallet watches full-nodes or mining pools through API), So Alice will send the Guarantee transaction to defend herself. Now we are ending in a situation by which there are two(or more) transactions in miners mempool which are using same UTXO. Miners are in business because of money, so they will put the transaction with higher fee in next block. The higher fee payer is the Guarantee Transaction. In this case both Alice and Bob will lose a portion of their money in favor of miners, but the numbers are precisely calculated to rationally hinder both "issuer" and "creditor" from cheating. You can find more detail on https://github.com/raymaot/transaction-numbers-and-coefficients .

Scam scenario 2: Alice send the valid transaction to Bitcoin network.
It is ok too, since the valid transaction contains just and fair outputs for both issuer(Bob) and creditor(Alice), the only barrier is the Bitcoin-transaction-fee which is fixed to 10,000 Satoshi and would be divided between issuer and creditor(s) in proportion to their outputs. So if they like to pay Bitcoin transaction fee (what Sabu is invented to avoid it ) they are free to do it. It is not scam at all.

ok. lets say Bob owns 1 bitcoin and signs a valid transaction but sends it to you privately without settling it on chain. now he has SABU tokens, right?
so how are you going to prevent these two scam scenarios:
1. Bob spends his SABU tokens, for example buys a car, and the SABU transaction is supposed to be fast. then immediately he spends the same 1 BTC UTXO that was never settled on chain or even broadcast to the network so he takes his 1 BTC back too. now the initial "valid tx" you had is no longer valid.
2. You never send him SABU tokens but instead broadcast his transaction which will confirm. now he is both out of BTC and SABU.

It is exactly what I refered as our classic internet. It is our current internet draw back, and it is exactly why I decided to use email, instead of centralized servers and domain names. As I told before, you can run your personal mail server. I designed a system that works independently by using the most independent protocol in the internet.

Please read the paper carefully and ask me what part you didn't understand. By the way I explain it here again.
You install the wallet and config a fresh dedicated email address in your wallet. Then you can add your friends wallet email (as payer or payee) to your connection. Now your mobiles are communication together by PGP-encrypted emails. You want to send Bitcoin, you just sign the transaction and your mobile deliver it to your friend mobile. On the other side your friend mobile decrypt and parse your email and show his balance in his wallet. That is all.

It has nothing with faith. It is exactly a cost benefit equation, while in the world there are both rational and irrationals, we are all agree rationals are far more than irrationals, and if some irrationals wants to hurt other in cost of their money no one can stop them, until they ran out of money. These attacks will be system costs and will be very small comparing system benefits.

You are right, they are few businesses that accept zero confirm transactions already. In fact they will switch to use Sabu transactions instead of classic Bitcoin transaction. The use of Sabu will be more beneficial, flexible and reliable for them.


As I told before,  there is no central server (except google play or apple store to download the mobile wallet) at all. So every one pay her/his part. If you need a good email service, pay for it, or just use free services. Doc-watcher is a light client(like torrent) which has no cost except a low internet bandwidth, and Bitcoin full-nodes and mining pools already existed.
The system is highly decentralized. Please study architecture carefully.

You are either from Mars or some other planet when you can keep saying that email server providers are decentralized... they are not decentralized.
Do you know that few days ago when fastly servers went down half of the internet also went down, including reddit, amazon, paypal, amazon, ebay, vimeo, etc.
If you know that huge majority of people use gmail and google services you can't really call anything like that decentralized.
There is one research showing that over 75% of all Gmail users are connected with their mobile devices...so much about decentralization of your protocol.

PS
You are still avoiding to answer my question to explain how your mobile nodes are actually working.

It is something quite serious with the sabu protocol though as it proves it's cheap for a bad actor with a lot of funds to double spend once they get where they want with "in-app" funds...

---

You kinda went on to suggest this was decentralised, trustless and Pro privacy but I can't even see a way it's any of those.
Especially considering the "masternodes" that need to run to maintain the network.

There are thousands of email service providers, and it is highly decentralized. No one is forced to use or not use them and no government can shout down all of them. You can use one or some simultaneously and if you do not trust none of them, just run your email server or even better run your freedom box https://freedomboxfoundation.org/
please note that the email is the ONLY neutral, free (non proprietary) and open protocol/technology for communication in the world that its infrastructure is well-established and is accessible all over the glob.

No, using emails is not decentralized solution all data is stored on one or two centralized servers, and I can also create million of domains if I want, including some free domains.

Shut down just one server and all your ''decentralized nodes'' are dead, and even worse than using emails would be usage of mobile phones as some kind of ''nodes'' and you didn't explain how thot would work but it all reminds on that junk Pi network.

You are right, most email service providers are centralized nevertheless you can find good independent email service provider as well. Usually they are not free.
But you misunderstood the concept of email and mobile as a server. The basic idea is:
You need a way to find out by others and finding others by you. In our classic internet you have a domain name which is totally centralized. Using email is a decentralized solution. You can create and change your email address million times with no cost. Of course email service providers usually do KYC, but it has less effect on Sabu protocol since all communications between mobiles are PGP-encrypted. As we all knew protecting privacy has different levels and if you want more privacy, just use privacy-care email providers. It is your choice, meanwhile there are some improvement proposal which I prefer to talk about it later.
For now we found a network of mobile wallets which are working purely peer-to-peer. Each mobile has a list of its friends and communicate with them(send and receive Bitcoin) and do not need any central server.

So you have no way of actually protecting your "customers" other than the faith of others not wanting to attack your protocol.

Also, you have not covered out the possibility of people using an accelerator in case of higher fees.
The next block confirmation fee is 50sat, the attacker uses a 20sat/b fee your protocol outbids it with 21 sat/b but the attacker uses an accelerator and gets the lower fee tx confirmed before our although higher has a chance of getting naturally confirmed.

There are multiple ways of cheating this outbid defense system it will simply turn out to be way too costly to keep it up. Why do you think nobody accepts zero-confirmation transactions? You're underestimating human nature, if it pays even a few cents per contract, someone would build a bot that that would wreak havoc on your system.


Nope, it's not!
Besides, you're relying on centralized services such as email servers, for broadcasting contracts so, who is going to pay for all of this?

If you (as a bad issuer) decide to waste your money, absolutely you can do it and you can harm The related creditors maximum 20,000 Satoshi in each transaction as well. In this case no one will cover creditors lost. But how many of these crazy guys exist?
Instead if you ask me "What if issuer is miner as well", I can address this on paper the answer of "What if issuer is miner as well?"

No one can force the issuers(UTXOs owner) to use and spent the UTXOs in a certain way. Issuer can deliver to a creditor a valid transaction, and later try to spend it in a counter-promising manner,
But because of limitations that wallet already forced to transaction format and amounts cheating chance is in minimum. Take a look at https://github.com/raymaot/transaction-numbers-and-coefficients and also other transaction control.

The Sabu protocol is the most decentralized system ever. Please study "The network architecture" in article.

If you can please explain a bit better how exactly are you using mobile phone wallets as ''nodes'' and emails, because this sounds to me like a privacy nightmare.

You say there is no central authority but most email providers are centralized and they track and read your emails like gmail for example that most people are using nowadays.

Sorry but I see a lot of holes in your theoretical medium post.

So, what happens if I send a transaction of 1BTC with a fee of 0.99 BTC?
I will be a bad creditor but at the same time, all the money is lost, so who will cover for this loss?

Also, as far as I understand the creditor will only give you access to the UTXO that covers the amount in the SABU protocol, it's all nice in your models to overcome an "evil" creditor but you have only counted a simple tx, not if he plans to combine inputs and use it to pay to different outputs, he will have another fee advantage over you as as a 1input, 2 outputs tx is 180 bytes and a 2 in / 10 out is 552 bytes so he will outbid you and still manage to save on the fees for the other transactions he was supposed to do.

And now to the SABU protocol, it's a centralized system, right?

Thanks jackg for your comments, but what you sommented, has nothing with Sabu protocol. The Sabu protocol doesn't care about hash-rate or "Current cost per phs" etc.
Please read the article carefully and "think out of the box". The protocol is based on UTXOs and potential valid transaction which people use them as a kind of debt-proving-document instead of sending it to Bitcoin network.
Potentially nodes would be millions or more with no limitation, since the nodes (mobile wallets) are standalone and it is not necessary everyone know everything about every transaction.

Have you thought of the international limit you're going to set on how much a node can hold on your network?

At current, I think I worked out a resync of every 2 weeks (the point where a new utxo is made) would take 0.5btc to be able to confirm within that time with neutral luck.

Hashrate current: https://www.coinwarz.com/mining/bitcoin/hashrate-chart
Current cost per phs https://www.nicehash.com/pricing

Values given are 116EHs and 0.0062/phs

116000x0.0062=719.2btc to take over the whole network for a day (obviously this is theoretical).

719.2btc/(144*14)=0.357btc

Thus, at 3000 transactions a block, each channel at the moment should be limited to 0.0001btc?
(blocksize ref: https://www.blockchain.com/btc/block/0000000000000000000489d050be1cb87b11ff2e77c9c6768b44f47f6eae6136)

Hey guys,
Sabu has nothing with Lightning at all.
It looks you do not know how lightning works, neither how Sabu works!
As a short answer in lightning both parties(sender & receiver) have to open channel and put the fund in a multi-sig account, but in Sabu no one has to replace its fund. The issuer just creates a valid transaction and deliver it to creditor. Indeed in 99.99% of time they do not need to record their money transfer on Bitcoin blockchain!
Please! read this post first.
https://raymo-49157.medium.com/time-to-boost-bitcoin-circulation-million-transactions-per-second-and-privacy-1eef8568d180

I am humbly looking forward to your conscious critique and helps about Sabu.

This looks really similar to LN, why not do a comparison to tell us more?

edit: looking at the paper is this not just the lightning network? Is it compatible with the lightning network?
You can check my link to the left of here for my summary of it, I linked the paper in the op too.

Could you ass to your post an explanation of how it would be decentralised (mainly with fund spending, would all coins be burnt to produce the one your protocol is issuing or are people able to sign funds out, how is this done with bitcoin having a limit of 15 signatures per multisig address - afaik?) , how it incentivises people to keep it decentralised, and how it stops attacks against decentralisation? Additionally is privacy fully assured - there's no gaps for reverse engineering how the system behaves to determine who has what funds?

by Ray Makan Otas (Raymo)


I am going to introduce a solution for Bitcoin low throughput (TPS) and its week privacy. It solves Bitcoin scaling problem and help its prevalence, particularly in small payments. It also highly increases the privacy level of Bitcoin users. It is a layer two protocol named “Sabu” and works perfectly with current Bitcoin core protocol. Sabu protocol is fully decentralized although it does not implemented based on blockchain or side-chains or any kind of DLT. In order to use Sabu protocol users only need to download the mobile wallet app (Gazin) and install it. No need to pay Bitcoin transaction fee, no need to open/close channels, no need to record transaction in Bitcoin blockchain, no need to run any server, no need to deposit or block money or Bitcoin in smart contract or stacking or any other third parties interference, even no need to have technical skills. And the last but not least no KYC at all.

How Bitcoin transaction works?
Owning Bitcoin, means having some UTXO (recorded in Bitcoin blockchain) under your control. That is you can sign that UTXO to prove you are the legitimate owner of that money. So if you want to spend your Bitcoins, you create a transaction by which sign your under-controlled UTXO(s) and represent your desire to transfer this ownership to the other person. This transaction is a document that issued by you and provides a legitimate order for this money transfer. In order to execute this money transfer, you need to broadcast your signed document to Bitcoin network aimed to record it in Bitcoin blockchain, otherwise, no money transfer has taken place. After recording this transaction in Bitcoin blockchain, “Everyone” will be aware of the new owner(s) of that particular spent coins.

How Sabu protocol works?
You -as a UTXO owner- are an “issuer”, and always can issue a document(AKA transaction) by which you represent your will to transfer some of your UTXOs to others. As long as this document is not registered in the Bitcoin blockchain, it is nothing more than a debt-document. i.e you owe some Bitcoins to someone else. That guy naming her/him “creditor” payed money to you or provided goods or services for you, in exchange of this transaction. Thus s/he has a copy of this transaction in her/his wallet. The creditor can send this transaction to Bitcoin blockchain network aimed to record this money transformation in Bitcoin blockchain, or keep this transaction in wallet. The creditor always can broadcast this transaction to Bitcoin network, but due to the high transaction fee on the Bitcoin blockchain and the insignificance of the amount transferred (a few Dollars), the creditor will not send the document to the Bitcoin network, instead s/he prefers to use this document as a payment method and exchange these documents in Sabu protocol and in an off-chain manner.
In this exchange process, you as the issuer will be informed of this credit transformation between two Sabu users and you have to issue a new document in which you owe the new creditor(s).
Sabu protocol is an off-chain protocol in which the UTXO owners (issuers) can issue debt documents and give them to creditors in exchange for fiat money or goods or services. The creditors can spend these documents and give them to other creditors or other issuers in exchange of money, goods, or services.
The issuers earn small Sabu-transaction-fee per each money transfer (10 Sat per transaction). Millions of issuers and creditors can exchanging these documents (transactions) in a pure peer-to-peer network continually, with no central authority. There is no blockchain nor public ledger. Users do not need to open/close channel or pay Bitcoin transaction fee neither routing fee at all.
After each dealing, the issuer cancels the old transaction and creates a new document, and updates the creditor balances. These documents will be in circulation between issuers and creditors in the Sabu network forever meanwhile less than one percent of these transactions will be recorded on the Bitcoin blockchain.
Either issuers or creditors in order to use Sabu protocol need to install Sabu mobile wallet (called Gazin) and start to deal. That is all they need. No technical skill or extra cost needed.

read full paper on
https://raymo-49157.medium.com/time-to-boost-bitcoin-circulation-million-transactions-per-second-and-privacy-1eef8568d180