Are trust-based private blockchains really trustworthy?

罗伯特•沃伦斯基is the Senior Managing Director of the Genesis Project, where he conducts blockchain research and leads a technology company focused on commercializing private blockchains.

In the article, Volensky offers his take on the increasingly popular permissioned blockchain design, arguing that permissioned blockchains do not have the ability to provide immutable record keeping.

More and more people believe that solutions to replace POW efficiency exist, and that traditional countermeasures can be used to maintain the authenticity of historical records on private blockchains, and even believe that a "regulated" or "trusted" group of people will not collude with each other to change historical records simply because they are regulated.

No distance from the trust mechanism

As the desire to commercialize blockchain technology grows, two different views have emerged on how to ensure the authenticity of historical records in blockchain systems: (1) POW and rules; (2) trust or authorization and rules.

In addition to the above two, many people believe that POS and its derivative mechanisms are the third security mechanism. However, after a detailed observation of the underlying mathematical algorithms and rules of this mechanism, you will find that POS is just a variant of the second mechanism mentioned above, so we classify it as trust.

Yale statistician Paul Sztorc once wrote an article about the economics and security of POS, which goes into great detail, so I won’t go into detail here. Suffice it to say that POS is less efficient, untested, and may be less secure than POW.

Does trust work?

Many people support trust-based blockchain systems and give a variety of arguments, including the basic principle that "rules" govern the interactions between parties, and in the case of certain financial technologies, all parties are under the regulation of the government and guided by higher authorities.

For the sake of clarity, let’s look at the quality of regulation by each party. Below, we cite a small number of examples from media reports to look at the behavior of a “trusted” and “regulated” group of people.

Bloomberg reported, July 24, 2015:

U.S. Treasury bond dealers accused of conspiracy... Bank of America, Goldman Sachs Group and JPMorgan Chase & Co., among 22 other financial companies, were accused of colluding to manipulate primary market U.S. Treasury bond auctions... In a similar case, the banks involved were fined $6 billion...

CBS News, June 20, 2015:

In a rare guilty plea, Wall Street banks admitted to manipulating the market... The five largest banks admitted to multiple charges related to manipulating foreign exchange and interest rates... Agreed to pay approximately US$5.6 billion (approximately RMB 35 billion)...

The Wall Street Journal, September 12, 2015:

“Wall Street’s largest banks have agreed to a preliminary settlement of charges that they conspired to manipulate the credit derivatives market…Twelve banks and two industry groups reached a preliminary agreement…to pay $1.87 billion…The consequences of this behavior on Wall Street may be more far-reaching than we think…

In all of the above cases, traditional responses (lawsuits, fines, settlement agreements, SEC investigations, and even prison) were used to address the problem of “collusion” between a “trusted” and “regulated” group of people.

So what does “trusted” mean in blockchain technology? Does it just mean that you can rely on a group of people for everything?

No, it’s not like that. In blockchain, “trusted” means a group of people are “trusted” to follow the rules of the blockchain.

This reasoning leads to the misconception that all trust-based systems are viable, but as demonstrated above, this is false.

Trust is outdated

Let's put aside that fantasy and face reality.

Any rules created by a blockchain (i.e. “rings of tokens”, “longest chain wins”, etc.) can be broken by a group of people who are in collusion. Why? Because there is nothing substantial stopping this group from breaking the rules, except their own beliefs.

If this group of people on the blockchain colludes, they can rewrite the local record regardless of the interests and protests of other parties. Others may not even notice that this group has tampered with the historical record.

Worse, because there is no way to prove which party has the correct record (i.e. the objective state of the ledger), the system will be fragmented and divided. Using dates to prove the authenticity of the distributed ledger? It doesn't work. Because data can be traced, it can also be rewritten.

So what happens if this group chooses not to follow the rules and forks the blockchain history? What mechanisms should the affected parties use to deal with this situation?

In a trust-based blockchain environment, there really is no alternative to the outdated traditional countermeasures that were often used to enforce contracts in the pre-blockchain world: lawsuits, fines, mediation, SEC investigations, prison, etc.

In summary, we can see that adopting traditional countermeasures to promote the enforcement of blockchain rules goes against the overall conceptual potential efficiency of blockchain technology.

POW mechanism

Under the POW mechanism, Satoshi Nakamoto anticipated that this group of people would commit fraud or collusion, which is the well-known 51% attack. However, Satoshi Nakamoto introduced a cost equation through the POW protocol as a non-traditional countermeasure against fraud or collusion.

The distributed ledger and trustless environment brought about by the POW mechanism is a huge shift that is fundamentally innovative and transformative.

The costs are predictable and huge, and it requires no traditional countermeasures to enforce blockchain rules, only a simple equation (the longest chain or the one with more computing power wins), thus creating a "trustless" environment.

"Miners' alliance" is another topic for discussion, so we won't discuss it here. Understanding the 51% attack is important, and it is an important entry point to understanding Satoshi Nakamoto's Bitcoin innovation and successful blockchain technology derivatives.

In fact, a 51% attack is a risk assessment mechanism that accurately calculates the cost of POW computing power required to overturn historical records.

POW makes it costly to overturn historical records. Without POW, alternative mechanisms are inherently subjective and fail to provide a practical and empirical way to quantify the fidelity of historical records.

Therefore, the immutability of the historical record becomes a theoretical academic debate rather than a scientific or objective fact. This unique (non-theoretical) ability to quantify risk makes POW very attractive from a tax and audit perspective, and drives the inherent efficiency of blockchain technology.

POW – A Paradigm Shift

Let’s take a moment to unpack the massive paradigm shift that PoW creates by creating a trustless environment.

Many people believe that existing crackdowns on illegal activities, such as fines, should be sufficient to curb illegal activities such as fraud or collusion.

As I have shown in the examples above, we know that this reasoning is wrong.

Because traditional countermeasures create a non-deterministic and dynamic environment, regulatory costs inevitably become an estimate of costs versus benefits, that is, zero cost for successful evasion versus higher costs for unsuccessful evasion in some cases.

POW is completely deterministic, meaning that the parties know the costs of fraud and collusion and must decide whether to support those costs in advance.

When the countermeasures are the most expensive and direct, and the efficiency is high, then the distributed ledger and trust-free environment spawned by the POW mechanism is a huge shift that is fundamentally innovative and transformative.

By now, it should be clear that trust-based systems are insecure and non-empirical “solutions” compared to the real security solutions provided by POW. In addition, it should be clear that those who do not support POW do not do so because they think it is safer to build a non-POW distributed ledger, but because they cannot obtain POW in an economical way.

POW is the key to unlocking a massive paradigm shift and efficiency in blockchain technology, without the need for traditional countermeasures and without any viable alternative to POW.

Without the POW mechanism, the system you build is still an outdated (and inefficient) distributed database.

Here’s a word of caution for companies seeking to leverage blockchain technology: Beware blockchain salesmen who tout “immutability” and understand the inherent elegance of Bitcoin. It’s not the software that makes Bitcoin so efficient, it’s the economics.

Because when it comes to historical immutability, we know there is no free lunch.

Original article: http://www.coindesk.com/can-trust-based-private-blockchains-be-trusted/
By Robert Wolinsky
Translator: printemps
Editor: printemps
Source (translation): Babbitt Information (http://www.8btc.com/can-we-trust ‎)