Crazy, the paperwork around the shipping often is more costly than the shipping itself:
> Maersk had found that a single container could require stamps and approvals from as many as 30 people, including customs, tax officials and health authorities.
> While the containers themselves can be loaded on a ship in a matter of minutes, a container can be held up in port for days because a piece of paper goes missing, while the goods inside spoil. The cost of moving and keeping track of all this paperwork often equals the cost of physically moving the container around the world.
Two typical scenarios - unions are holding things up until they get a bribe, or there is contraband on board and the holdup enables people to get it off.
So that sounds like at least two parties who would be opposed to an unrevokable ledger. Govt officials awaiting their bribes to let a container pass would be a third party opposed.
If you haven't seen it, season 2 of The Wire is mostly concerned with this. Although the show is fictional so I can't vouch for its accuracy on this, but it goes into a good deal of detail.
This means a blockchain where to participate you need to be approved by a single party or a consortium.
In my opinion if your blockchain is private then you might be better off using an existing database technology centrally controlled and allowing people to participate via an API. Expecting people to install Blockchain nodes locally to participate in this scheme might be expecting too much.
The problem is that when software updates are required, which they will be, you have the job of coordinating updates across a node infrastructure installed in many companies and locations. If it goes wrong you have a chain split.
Another problem is that the people who recommend or buy into the Blockchain hype might not have the skills to properly asses it's appropriateness to certain business cases.
Are you saying that the very use of a block chain is too complex to be deployed by one IT-operational department at a large scale without issues? Just trying to clarify what you mean.
What is the advantage of using a block chain for storing the information rather than just having a database? Is there some intrinsic benefit or is it just marketing hype?
Trustless trust meaning that with a blockchain you don't have to trust any party involved to be certain data hasn't been tampered with. There's a huge number of agents involved in supply chain management not all of whom can be trusted.
There are third party institutes, tests labs and the like whose only job is ensure everyone plays by the rules. Making matters worse not all of these institutes can be trusted either in every case every time. It's turtles all the way down ...
The blockchain perhaps is not a comprehensive solution to all of these issues but it eliminates a significant opportunity for fraud right from the start.
I'd go as far as to say that supply chain management probably is one of the few truly viable use cases for the blockchain.
> I'd go as far as to say that supply chain management probably is one of the few truly viable use cases for the blockchain.
I get that there is a massive amount of hype and bullshit around everything "blockchain", but surely it's way too early to write it off[1]. It could be as prevalent as HTTP or SQL some day (or maybe it disappears, but it's too early either way).
I think one exciting area is voting - imagine the convenience of mail-in ballots, verifiable, instant vote counts, and the ability of anyone, e.g. the electorate, universities, political parties, transparency non-profits and so on to act as a node and therefore monitor the voting in real time. I think it could be a great innovation to strengthen democracy. Imagine a real time, direct democracy that has voting (or even just opinion polls) on a continuous basis to keep politicians in check.
Uhhh… I'm no expert regarding blockchains, just know what the basic idea is. And in my understanding you can't have truly anonymous votes using a blockchain. Or can that be done?
I imagine it could be done if you could distribute one truly anonymous, but verifiable token to each eligible voter and if there is a way to anonymously insert a block "token x votes for y" into the blockchain.
To provide the anonymous tokens, fully homomorphic encryption could be used to have a trusted authority blindly sign a token when a voter requests it. (Maybe these requests could also be recorded in the blockchain, to make sure nobody gets two tokens?)
The hard part will probably be the anonymous network access.
The hard part isn't the network. The hard part is anonymity (secret ballot) is directly at odds with verification. The 'x' in the formula "token x votes for y" is so secret the voter themselves can barely be allowed to have it, never mind their Gmail or Yahoo account. And never mind verification over the Internet, even on a LAN, verification is suspect to being hacked.
Secret ballot requires that it's hard to verify tokens, or else paying or otherwise coercing voters to vote one way becomes way easier to do at scale.
> “As a global integrator of container logistics with the ambition to digitize global trade, we are excited about this cooperation and its potential to bring substantial efficiency and productivity gains to global supply chains, while decreasing fraud and increasing security,” said Ibrahim Gokcen, chief digital officer, Maersk.
> This promotes sustainable transport by integrating shipping processes and partners, and establishing evaluation frameworks through increased transparency and trusted access.
Seems like the benefits are a) transparency, and b) distributed. This is Maersk leading the effort, but it seems the plan is to get the entire chain onto the same system, which basically requires decentralisation.
Edit: The New York Times article [1] lays out more details in their reporting too.
In these kinds of cases it's mostly about simplifying transactions or contracts via hash-based signatures.
Scenario:
1. You request service A from me
2. I confirm receipt of service request
3. I provide service A
4. You confirm receipt of service A and provide payment
5. I confirm receipt of payment
Since each block of a blockchain is dependent on the previous blocks, it makes it impossible for me to go back and adjust the record to say, claim you never paid me, or change the order quantity, or for you to go back and claim you never ordered a thing.
It makes 5 depend on 4 which depends on 3 which depends on 2 which depends on 1 and altering any of the steps "breaks" the later steps in such a way that you not only can tell they're broken, but also at which step did the chain break.
In this sense you can basically use math to provide enforcement of the kinds of transactions that might occur during a sale. With shipping containers you can simply put in movements of the container, or manifest of the containers or track bill of lading [1].
I'm not in shipping - but wild speculation could be once you sign the blockchain it's now your legal responsibility - no dickering over timestamps, is that really our guy's signature, forged paperwork, whatever. You signed it now it's yours until someone else does.
Implementing that may be somewhat more difficult though, and I have no idea if this is really that large of a problem or not.
Maersk will be implementing a new standard of implementation here, replacing a standard that's been bootstrapped over centries. Expect an ISO or similar to be created to define it, with a lot of reach into the workflow of insurance companies and banking (LCs, etc).
This would be an interesting area to 'disrupt' if you could catch early wind of the emerging standards and ride on the back of a new standard coming into implementation.
It likely is a problem given that shipping is often a many to many relationship between multiple providers -- which one do you trust to provide data lineage of the entire shipment? Nobody - use a blockchain
the person who owns the database can write whatever they want in it, where as a blockchain provides a chain of trust which extends to the (hopefully secure) client. An immutable ledger allows circumvention the armies of people needed for rubber stamping documents and verifying rubber stamps.
It is now moving in the direction of becoming a politically only acceptable solution, when no one can impose any negative judgment making the solution provider look malicious even before anything bad happens.
The problem to be solved here is not chiefly technological: It's getting all of those agencies to agree to a single messaging protocol. That's hard! They have long experience of using their own protocols (e.g., paper), and little incentive to switch to Maersk's. Calling the new protocol a "blockchain" makes it sound sexier, and so more likely to be adopted.
I'm inclined to agree with the premise, that "blockchain" is just what the cool kids are using, and adoption is piggybacking on that. Unfortunately the article fails to go into any detail as to why that is or is not the case, it merely quotes from the press release and then Matt points out that central banking exists.
Shipping, especially on Maersk's level, has all sorts of challenges that I'm sure I've never even contemplated, but Matt Levin's piece doesn't mention any details that leads me to think he's working off any more details than I.
Large shipping container docks are horrible environments for electronics and wireless technology. Long distances, lots of big metal things, lots of water present - large bodies of water block and bounces radio waves and salt water is corrosive to electronics, concrete everywhere, so laying cables is expensive, and there may not even be power in a lot of places. Add multiple layers of IT contracting on top of that and I'd be surprised if there's anything Internet except at the main office, miles away from everything else.
While blockchain doesn't address problems with networking technology, I can easily imagine the blockchain, as a solution for the Byzantine Generals Problem, addresses practical issues when you need to give write access to the database, to multiple parties, some of which cannot be trusted.
I think it's very interesting/strange/scary to see what started out as a Cypherpunk dream come true is now being experimented with / used by the industry it was meant to replace.
While I think Bitcoin (et al.) is extremely interesting and definitely has it's use on the internet, it is becoming clear that it is a lot harder to replace money than it is to replace media supply chains (torrents): Everywhere from the political situation (blocksize discussion) to current usage cases (a lot has to do with circumventing either capital controls or AML/KYC for a number of reasons).
disclaimer: I do a lot of work in Bitcoin but am currently employed by a bank for a project very similar to this one.
I have become a bit of an IBM skeptic, what with excessive IBM Watson hype, but this seems like a really good idea. A distributed and secure blackboard type system seems like the way to go, rather than using a database that one of many parties is trusted to maintain.
I just looked at the Java SimpleExample code in the repo and it did not look too difficult to use.
I suspect you can plug in any consensus technique. For a private blockchain like this, you can go light because all participants are trusted not to misbehave.
At that point you might ask "so why are you using a blockchain at all", but you won't rack up billable consultant hours with that sort of thinking.
The first paragraph mentions it's using Hyperledger's Fabric[1] blockchain implementation. I assume they're building their own smart contracts on top of that.
The hyperledger equivalent of ethereum's smart contracts is called chaincode. I'm playing around with it now - so far enjoying it much more than smart contracts.
I wouldn't be surprised if it was a private blockchain with access via paid subscription and thin clients that don't have to pull all of the chain to view it.
> Maersk had found that a single container could require stamps and approvals from as many as 30 people, including customs, tax officials and health authorities.
> While the containers themselves can be loaded on a ship in a matter of minutes, a container can be held up in port for days because a piece of paper goes missing, while the goods inside spoil. The cost of moving and keeping track of all this paperwork often equals the cost of physically moving the container around the world.