by bvod on 7/19/17, 5:29 AM with 68 comments
by austenallred on 7/19/17, 6:19 AM
To know why that matters you have to understand the problem(s) it's solving - it's something people have been working on since the beginning of money, but there are so many edge cases we usually ended up just going back to a centralized authority. In the past you've needed something like a bank or a government or a company to have the authority to declare, with authority, "Person x owns the title to that house."
There are a few problems with that: They could do things to screw everything up (a government printing currency and causing inflation), be corrupt (countless examples), or perhaps not even exist because it's too much effort to create, set up, and monitor some body.
With the blockchain, the software everyone runs keeps track for everyone else in an incorruptible way (yes this is a simplification), so you're free to make changes (or transact) without having someone in the middle to wait for, and with no one that can really screw things up. Just boom, instant marketplace.
The most obvious thing you'd trade on a blockchain is some kind of asset - a cryptocurrency is the main example, or titles or stocks or something valuable.
Say, for example, I have some loans that people owe me on. I can throw those loans on the blockchain, and without a clearing house somebody else can buy them, and when the loan payment comes due you can easily see who owns the right to be paid.
It seems trivial, but allowing people to seamlessly make transactions of any kind without the overhead of a bank or clearing house is a big deal, and has a ton of applications.
by colept on 7/19/17, 6:13 AM
How to trust a single entity by using a magnitude of peers to establish an agreement. In the case of Bitcoin it's a monetary transaction.
One example would be file digesting. When you consider malware: if files were signed and digested on ablockchain we would have better oversight of the authenticity of a download. A file could be confirmed against the blockchain to detect if it matches the certification. Since the source alone cannot be trusted, the confirmation of many adds that layer of trust.
Anything beyond the issue of trust is not suitable for a blockchain as there are more efficient technologies.
by davidgerard on 7/19/17, 7:44 AM
I have a book coming out next week (!) on the subject. https://davidgerard.co.uk/blockchain/ I have a whole chapter on business blockchains, and I looked hard for a real-world example of one in use. There actually aren't any.
The closest we have is .. git! Transaction ledgers, each with a tamper-evident hash, in trees and chains of hashes. Devs routinely throw entire ledgers/repos around, identified by hash. It definitely counts as a "distributed ledger technology", and it's hugely successful. The only thing it doesn't have is a consensus mechanism - it's "here is my tree" or "here is this repo, this is the hash".
As we're seeing here today, most claims for business blockchain are literally the airiest hypotheticals regarding Bitcoin, with the buzzword changed.
The usual concrete posited use cases are interoperability (that it will magically clean up your data and formats) and that it will magically resolve real-world human-level disputation. Neither of these is likely to work out that way.
by retube on 7/19/17, 6:54 AM
The only use case I know, beyond proof of ownership of coins/tokens (and, again, how can that be applied to a real world problem), is proof of document state at a certain point in time (via storing document hashes in the blockchain) which clearly has some uses in law for example, but apart from that I am stumped.
by tbking on 7/19/17, 6:06 AM
️=> Most used application is money-based transactions. Banks are central authority and are affiliated with governments, which makes them charge of your money. Use blockchain (Bitcoin).
=> Supply chains are inefficient due to checks and sign offs by so many people. Use blockchain (Hyperledger)
️=> People don't agree on terms, and they end up in a feud or a lawsuit which is costly. Use blockchain (Ethereum)
Blockchain, and decentralization on whole is seen to be as future. Many want to tap it before others. Of course, large number of startups will fail, but it doesn't say anything about the potential this technology holds.
by eksemplar on 7/19/17, 6:23 AM
Because a shipping form gives you global ownership of a container and because corruption, the security required to handle shipping forms pre block chains meant it was more expensive to ship the form than the container, and still it wasn't completely secure.
The block chain shipping form solved this issue completely because you can never fake forms or hide when ownership of a container changes hands.
You could frankly do something similar for public records of landownership, which might not be important in the west where corruption is low, but could revolutionize the third world.
by haspok on 7/19/17, 7:36 AM
Actually, this is better: http://iang.org/papers/triple_entry.html
My attempt at trying to explain it in one sentence: blockchain provides the technology to connect the double entry ledgers of two (or more) parties by recording the transactions between them, in a way that makes it possible to _prove_ the existence and validity of the transaction independently, without relying on either party's ledger.
by nikanj on 7/19/17, 6:29 AM
It's like XML in the 90s or Web 2.0 in the 00s. Gotta use the technology to get the funding, regardless of actual suitability to task.
by dragontamer on 7/19/17, 6:09 AM
I think for most businesses, the real applicable portion of the blockchain was cryptographic signatures to solidify public ledger. That's about it. Kind of simple, but I think that's what people are paying attention to.
In effect, its less about the technology that BTC and/or cryptominers care about... and more about the boring part that seems to get people excited.
A lot of "blockchain" companies seem to be enabling peer-to-peer transactions for example. By centralizing all transactions to a particular server. There's this one company (I forget the name) which claims to be using Blockchain for exchanging Solar-credits between neighbors.
Having a ledger that is cryptographically reliable, even if centralized, is the main benefit. Also, one that can be fully automated is a big deal. I mean, that's all Visa or Mastercard really are: systems that describe when and where transactions have occurred around a centralized source of trust.
---------
EDIT: And yes, I know what a blockchain is in BTC circles. But I don't think a "Bitcoin Blockchain" is what people are talking about on typical marketing material. Just like "the cloud" has evolved to mean something new... "blockchain" seems to have been picked up by managers and/or venture capitalists to mean something totally different.
by ebcode on 7/19/17, 8:10 AM
If you're old enough to remember the days before the internet, access to information beyond what your parents and peers could tell you was found in the public library. Books. I have always loved books for this very reason. They give me access to information beyond what my immediate community can (or will, when you think about it) tell me.
Of course, the problem with the internet (in its present form) is that the information is so unreliable. There are just too many conflicting interests. With the library, you at least have the librarian, who, being a sort-of gatekeeper, makes sure that the library contains only "good" books.
So, blockchain. What problem does it solve? It's a really good question. The blockchain is a distributed ledger, that is, basically, a key-value store. The keys are the bitcoin addresses, and the values are the bitcoins themselves. What is unique about it, and what makes it such an interesting program, is that the record of transactions is constantly being updated, because there is a bitcoin "reward" for running the program. So, even if everyone stopped trading bitcoins tomorrow, if they were just like, "nah, we're over it", the bitcoin network itself would still generate some new bitcoins, and the ledger would be updated. It's really a fantastic invention, when you think about it.
The value of a blockchain is that it adds a type of "reliability" on top of the internet. Because the ledger is public, anyone can inspect it. Because the network is distributed, all the miners running the program are, for want of a better phrase, "keeping eachother honest". When you look at a regular website (or book, or movie), it can be hard to tell if the information you're seeing is legitimate/honest. Whereas when you look at the record of the Bitcoin blockchain (for example), you can be fairly confident that those keys (addresses) were assigned those values (bitcoins) at that time.
To sum up, the problem that a blockchain solves is storing a reliable record of transactions, by harnessing the inherent self-interest of people to "make money" (ha ha).
by Hamcha on 7/19/17, 9:54 AM
by bmcusick on 7/19/17, 4:20 PM
It solves the problem of financial intermediation. The whole financial world has been built on financial intermediaries (banks and stock exchanges) to verify that you actually own things. Now you can point to a public record (the blockchain) for that, and no one can take it away from you (without your private key).
by ruairidhwm on 7/19/17, 6:58 AM
Blockchain would be excellent for verifying that a bag has been received and processed at a certain place.
Lots of the large GDSs such as Sabre/Amadeus/Travelport are actively looking into this.
by itamarst on 7/19/17, 11:06 AM
by Abimelex on 7/19/17, 2:26 PM
by mshanu on 7/19/17, 9:06 AM
by 0x4f3759df on 7/19/17, 2:55 PM
by shpongled on 7/19/17, 6:12 AM
by Temasik on 7/19/17, 6:44 AM
by contingencies on 7/19/17, 1:51 PM
IMHO the term "blockchain", in current and common use, actually means Merkle trees applied to time-series information that is shared between multiple computers over the internet, hence 'distributed database'.
Merkle trees are basically a way of saying "this precise set of information was used to generate this new precise set of information". Applied to successive timestamped data in a blockchain, this means that participants within a network can provide strong evidence for past state and state change - ie. they establish relative trust in the fundamental integrity of a shared and evolving database (the blockchain) based upon mutually agreed rules.
What is the real world advantage? Frankly, in most cases, there isn't one, but there are definitely disadvantages. Distributed databases are not new, and any other way of executing them is demonstrably more efficient, simple, etc.
The main potential advantage is to remove the ability of a central authority (or small colluding group of bad actors) to "change the rules" by "rolling back" the state of the system, halting the system, seizing control, removing actors, etc.
This property is perhaps at its most potentially impressive when its implications with respect to the conventional business environment, government and politics are considered for large global industries with deep pockets. Simply speaking, right now most regulations are issued in airy-fairy lawyer speak and cost a lot of money to skew/change/interpret/argue/take to court. This is the case for government regulation, case law and non-governmental agreements (industry codes of conduct, informal protocols, MOUs, contracts, SLAs, etc.).
The potential advantage of distributed systems with mutual trust (note that blockchains are only one implementation of this concept) are that all of this overhead can be removed, since an adequately cross-border (ie. cross-jurisdiction, or government-independent) network of an adequate size will be essentially immune from direct state actor intervention, and the rules for all parties remain clear.
To summarize: save lawyer money, ignore regulators, save time/hassle, remove any means of effective government intervention.
The reality, however, is that most industries are not able to agree on a clear set of interactions and rules under which to formalize their activities, and even if they could it would probably not be feasible to do so in all cases, therefore this semi-utopian objective cannot be reached, and in my professional opinion few corporate blockchain projects to date are delivering anything more than an inefficient means of distributing simple time series transaction data.
The big exception (ie. winning example for your last question) is Bitcoin, IMHO because it was the first, and because it quickly enabled safer and more cost effective means of recreational drug distribution, which essentially provided the initial market which funded its growth (now tending to speculation).
My prediction is that we will see a rise in smaller, private systems and asset types providing transaction services with different properties to present era cryptocurrencies and a smart layer will emerge to evaluate and utilize these based upon their objective properties and per-transaction requirements / risk models. Many of these systems will not be based on blockchain at all. Instead of de-facto bank or card systems, we will have a broad range of settlement mechanisms available. Instead of stock exchanges or large VCs, we will have a broad range of funding sources available. Access to capital will improve, business transparency will improve, government regulators and established financial services monopolies will slowly be disempowered.
Further reading - https://en.wikipedia.org/wiki/Merkle_tree + https://en.wikipedia.org/wiki/Blockchain + https://en.wikipedia.org/wiki/Time_series + https://en.wikipedia.org/wiki/Distributed_database + https://en.wikipedia.org/wiki/Bitcoin_network + https://en.wikipedia.org/wiki/Money + https://en.wikipedia.org/wiki/Cryptographic_hash_function
by redy on 7/19/17, 6:55 AM
The problem is that whenever you have a public anything there will be spam. People will come along and they will fill the space up with scam ads and porn and hate speech. See: the internet.
To prevent the database from filling up with spam you need to find a way to make people "pay" for writing to the blockchain. Nothing is free and if you're going to consume resources (storage and cpu) from other's computers then they ought to get something in return. That's what "transaction fees" are and that's why they're measured in costs per byte -- the people who are writing to this public, decentralized, highly resilient database are being charged storage fees.
So blockchains are the solution to spam. And unlike pre-existing spam solutions a blockchain doesn't require storage/delivery nodes to "whitelist" certain certificates or for certificate authorities and CRLs and all that wackiness. This is kinda remarkable.
This innovation enables all sorts of new models of collaboration (read: business models). Just today I saw a very interesting proposal for an Uber-like system built on a blockchain. Everything -- requests for transport, driver availability, driver fees, driver certification, passenger and driver ratings -- gets stored in the blockchain. Anybody can read, anybody can write. Here's the database, here's the type of "transactions" (records) -- go crazy. The interesting thing about this system is that it doesn't rely on huge bureaucracy to protect the database. (And what are many business firms but database maintainers and protectors?) Write whatever you want -- write pics of your cats if you like -- but it's going to cost you.
(There's another element of what's commonly called blockchain which is "block verification." The problem here is that when everybody has a copy of the database how do you decide whose copy of the database is authoritative? This is the concept of consensus and note that it's only a problem if the peers in the database don't trust one another. It's actually not so interesting but it is largely an implementation detail of public blockchains with different strategies proof-of-work, proof-of-stake, electoral-validation etc.)