Hacker News new | past | comments | ask | show | jobs | submit login

The blockchain's (current) prime use case: triple entry accounting. In double entry accounting, a ledger only contains debits and credits. With the blockchain, the blockchain's attestation of the validity of the debit and credits is the third entry. This removes the need to trust whether the credits and debits were correct in the first place. Were financial systems to be based on blockchain design, an Enron situation would not be able to occur. It might even prevent the likes of Madoff. This technology is only getting started. While there is hype, this hype is necessary for the market to test out many ideas and applications, until those that make sense remain standing.

No. I support the blockchain and think it will be revolutionary. However, Enron's issues weren't attributable to some discrepancy between credits and debits.

Accounting, surprisingly to folks who've never studied it and think it's simple arithmetic, requires judgment, based on rules that are sometimes grey and subjective. When you add in legal complexities, it becomes... well, more complex and therefore open to manipulation.

For instance, Enron was able to hide many liabilities by marking it (in hindsight) below market, since... there was no market for said liabilities. So it was impossible to value them. Furthermore, it hid other debt in obscure subsidiaries that were only tangentially, legally and fiscally speaking, connected to Enron.

The blockchain confirms, with better accuracy than existing systems, "X transaction occurred between Y and Z partners." It does not confirm "X's assets were appropriately valued and marked as such to a non-existent market, and X is most definitely a legal subsidiary and is overseen by the fiduciary duty of A Holdings, Corp."

The blockchain has the potential to be a far more efficient settlement system. But saying it prevents Enron or Madoff is simply not true at all.

It's a step towards preventing corruption.

The automated clearing house system, our current means of settling payments, is pretty secure. The bigger threats are someone getting ahold of your credit card or checking account info, which is the same threat level whenever you want to convert your Coinbase balance back into your local currency (or someone getting ahold of your public key).

The blockchain, rather, represents a better clearinghouse merely by being a decentralized platform to account for payments rather than a centralized one.

OP says this prevents accounting fraud. Not one iota.

It won't prevent corruption. It will alter the logistics of corruption though.

Uh, no. Enron’s debits and credits always added up. It was what they claimed as credits that was the problem.

It's not the only the correctness of the debits and credits that is possible, but also determining their nature. Of course these implementations would have to be molded to fit for each use case. The technology is nascent.

Yea, the hard part is when it comes to determining the value of the collateral for some obscure project in rural Africa, vouched for by the trader who sold the trade and signed for by somebody 10,000 miles away who might be the traders pal.

Yes, OP seems to think the blockchain somehow solves the (extremely subjective) issue of valuation, as opposed to simply "did this transaction occur between counterparties?


I think you are a little optimistic about blockchains preventing Enron type situations. From wikipedia their issues were:

>by the use of accounting loopholes, special purpose entities, and poor financial reporting – were able to hide billions of dollars in debt from failed deals and projects.

which you could probably spot if you had access to all their bank accounts including those of the special purpose entities but no large company is going to make all that stuff public. If it's on a blockchain it'll be encrypted some how. The people who are supposed to have access to that stuff and check it are the auditors, in this case Arthur Andersen who screwed up and were destroyed by it going from 85,000 employees to mostly closed.

I think I understand the advantages in this specific example and I can see how it's a great solution, even if I don't believe it's the future of banking/currency.

However, I was really asking for the benefits of blockchains more generally. I've been hearing a lot of smart people say it's the future, and I've seen news articles about numerous companies looking into building their own blockchains, but I don't really see what the general use case is for it. The only example that's ever made much sense to me is the example of cryptocurrency.

What's the general use case of the internet? Any answer to that still leaves you to connect the dots with real-world effects. It sounds like you wants some of those real-world effects for blockchain technology listed for you. Here are some far-reaching possibilities...

* Puts the whole world on the same level of financial freedom and participation - farmer in Africa without an internet connection permissionlessly invests in a Chinese tractor company via text.

* Solves the problem of ownership - no need for house, car, etc title companies.

* Brings a new level of honesty to the financial world - auditing industry goes away because it's not possible to cook the books.

* Every movement of value becomes near-instant and doesn't require trusted middlemen - elimination of clearing houses.

* Breaks down national borders - vastly improved supply-chain logistics.

It has the potential of reshaping almost every corner of the underlying structures of modern economies. Just like the internet, it's the foundation for a new chapter of innovation. Who knows what will come of it?

I’ve seen some pretty useful models for delegation of authorization come up. For example, a contract deploys trust to a root node in the chain (somewhere in the chain), then the chain allows public distributed rules regarding that trust encoded in the contract via transaction. Effectively, the holder of the trust “pays” the currency of privileges to the receiver. The advantage here is the ability to transmit privileges is entirely autonomous.

Do you have a source on this I can read more about this? It sounds interesting, but as someone who works in security and understands the basics of blockchain/contracts, your description was highly confusing.

Please spend sometime reading up on Enron before saying this. As it has been pointed out Enron's book was clean. The problem was mainly due to "Revenue Recognition" [1] and valuations. It basically describes when a revenue can be added on to the books. Enron booked it's revenue as soon as deal was booked. So, if they got 100 million deal for next 5 years, they should be booking 20 million each year. What they were doing was to show 100 million as revenue for this year. This is not a single entry as many people believed rather a series of credit and debits which always balanced.

The amazing thing is blockchain, specially things like Monero or Zercoin, makes it even more difficult to spot this as Enron would have been able to open accounts as they wanted.

I am writing a piece on blockchain transactions and I still don't understand how this triple account thing applies.

Let me explain how double accounting works - You and I have 100 in Bank A. The Bank A notes:

Bank's Liability to: You - 100

Me - 100

Total - 200

Asset for:

You - 100

Me - 100

Total - 200

When I pay you 50 out of it, the book shifts:

Bank's Liability to :

You - 150

Me - 50

Total - 200

Asset for:

You - 150

Me - 50

Total - 200

Both side in balance.

Now this book is obviously internal. Can you use this same example and explain how triple accounting works in blockchain?

[1]: https://www.investopedia.com/terms/r/revenuerecognition.asp

This is indeed not what blockchains prevent. If you have bad rules (ie. government is complicit either accidentally or intentionally) there is nothing that can be done.

And because governments really don't want to admit to their constituents how much they really spend and what they spend it on, the rules are ... less than perfect.

But a blockchain can vouch for the fact that the ledger was made according to the (supposed to be published and fair) rules, and that the rules didn't change except with total agreement.

So enron would have been able to book things in against the rules, but an easily written automated program would have detected such shenanigans, and there would not be anything Enron could do to prevent that program from raising red flags when the transactions don't follow the rules (so it would not be possible to book in such transactions, then not book in payment in full when it's due. If you're wondering how that's possible, consider that in bitcoin there just isn't any credit. So the only way to place an order using a bitcoin directly is to stake they money ahead of time, optionally with an instruction that it'd be paid back if an "oracle" (anyone else on the blockchain) says it's not delivered correctly)

One question - Before you pass a judgement on government "bad" rules, let ask you something, have you ever run any company of decent size? Because if you do, you will realize things are more difficult than you realize.

Let's take an example of a retailer. She buys stuff from a company to sell in her stores. And she has How or when does her book it as a revenue? When it is bought? When it is sold? Common sense says after it is sold. What happens if there is a return? There are so many rules to cater to so many things. It's something you can hardcode into program code.

> bitcoin there just isn't any credit.

And far from you belief, it is not a good thing.

> optionally with an instruction that it'd be paid back if an "oracle" (anyone else on the blockchain)

I like how a statement defending cryptocurrencies goes all over the place.

Pray tell, how are "oracles" implemented in "bitcoin"?

Also, why should people not trust the government but random "anyone else on the blockchain"?

That said, I am still waiting for someone to explain me this "triple accounting" thing on blockchain.

I do not think lack of credit is a good thing. Or rather, I have the game theory opinion. It is not a survivable thing.

Given 2 economic systems. One with credit and one without. The credit based one will crush the non-credit-based one.

Does that mean credit is good ? No, but it makes it a moot point: we will have credit. Point. Bitcoin will not survive, in the long term. I get that.

Triple accounting is explained. You have an incorruptible third party that attests to the accuracy of the computation.

I am sorry but given that you can't explain using the simple example of double book keeping makes me doubt if you even know accounting at all. Just repeating the standard blockchain lines over and over again.

The accuracy of the computation is pretty much never the issue anyway. That's checkable by any auditor. As far as I can tell triple entry accounting adds one buzzword..

When was the last time you checked the books of the bank where your money is stashed ?

Because for your bitcoins, that would be "45 minutes or less".

The essential point of the third entry has been made.

How about using the above example and showing where the third entry is made?

The blockchain is the third entry.

you are also assuming all of the accounting is on a public ledger. It's not necessarily the case, is it?

Well said.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact