Slashdot Mirror
IBM Completes Blockchain Trial Tracking a 28-Ton Shipment of Oranges (coindesk.com)
IBM recently completed a trial of blockchain technology to track a shipment of mandarin oranges from China to Singapore. From a report: Announced last week, 28 tons of mandarin oranges, or 3,000 cartons containing approximately 108,000 fruits, were delivered ahead of Chinese New Year celebration on Feb. 5 (mandarin oranges are a symbol of prosperity, IBM explained). The main shipping document, the bill of lading, was recorded on a blockchain. This document serves as a proof of ownership of goods, as a receipt of goods and a contract of the shipment, and normally it's mailed to all parties involved in the shipment, including banks providing trade financing.
For the pilot, IBM created an electronic bill of lading, or e-BL, which helped reduce and speed up administrative processes "to just one second" as the document flow is automated, the company claims -- while the standard paper-based procedure takes five to seven days. "By using the e-BL, we have seen how the entire shipment process can be simplified and made more transparent with considerable cost savings," Tay Khiam Back, the chairman and CEO of fruit importer Hupco, said in a press release.
For the pilot, IBM created an electronic bill of lading, or e-BL, which helped reduce and speed up administrative processes "to just one second" as the document flow is automated, the company claims -- while the standard paper-based procedure takes five to seven days. "By using the e-BL, we have seen how the entire shipment process can be simplified and made more transparent with considerable cost savings," Tay Khiam Back, the chairman and CEO of fruit importer Hupco, said in a press release.
but why do that?
...but were the oranges 3D printed?
Only if you think Consensus beats Authenticity. I'd somehow feel safer having signed documents where I can verify the digital signature, and trust the certificate chain. Even if the parties pre-arrange to use self-signed certificates.
Or maybe use the Consensus approach to Generate your self signed certificate, thus avoiding any CAs. That is a lot less computational work and waste of resources. You only generate your certificate occasionally. But you could use it to sign much more frequently.
I'll see your senator, and I'll raise you two judges.
Blockchain tends to work well when you have a bunch of people that don't trust one another, but trust the group. This scenario is simply tracking packages, where you trusting one central authority (the maintainer of the DB) seems perfectly fine.
So why would you use blockchain for this rather than a database, or some other centrally managed data storage?
block chains either have to go to a central authority or they have to be distributed among a limited set of nodes that that sync to maintain the longest chain.
If you want it to be secure then I think that you either have to use a very expensive hash and pay the Miner (like bit coin proof of work) or you have to limit this to a distributed set of miner nodes that use a shared secret.
one of the other has to be in place I think because otherwise anyone coutd forge the chain if it's free to hash and not a closed set of trustee nodes.
Since were talking oranges , individually, then you need the cheap method with a centrally authorized set of approver nodes.
But if you are going to have such a centralized authority. Why not a regular old fashioned database?
I just don't understand what is special here about the block chain.
It does have some robust aspects of syncronizing a distributed set of approver nodes by consensus (longest chain wins). But that isn't really special and can be done with databases that synchronize too.
Some drink at the fountain of knowledge. Others just gargle.
Its a 1 page PDF with like 3 relevant fields to anyone looking at one. It takes one second to read one. They simplified nothing.
And in this case, the weakness will be linking the blockchain to the physical goods. Anyone can still crack open the container and steal oranges during transit. A drug smuggler can still swap a shipment of oranges with a shipment of drugs by switching the labels. And the delivery guy can still take a snapshot of your package on your porch as "proof" it was delivered, then load it back up into his van and take it home.
Basically, the rest of the world is going to start recording their transactions in git repositories.
It provides compounded verification and cryptographic authentication (e.g. commit signing).
However, nobody would understand "IBM completes Git trial". In fact, people might be offended. The cool word is "blockchain".
How does a blockchain entirely owned by one company (IBM) guarantee the transaction an immutable transaction history? Is the code open source and available to be run / scrutinised by others?
- The Kessel run is for nerf herders. I can circumnavigate the entire Central Finite Curve in a lot less than 12 parse
The bubble's burst, everyone's losing their money, blockchain for anything is like using activex on IE6.
I can easily imagine that a heavily-stage-managed and custom documentation process would be way faster than the lowest-common-denominator legacy paper arrangement; but I find it much harder to understand why the 'blockchain' based updated electronic records process would be faster or more efficient than a similarly updated electronic records process that's based on some boring database instead.
It's easy to compare updated processes against legacy ones and get good results(especially given that there are probably way more techs watching this order and making sure nothing bad happens than any normal one enjoys); but if you want 'blockchain' data what matters is how it stacks up to a similarly updated process that uses a different backing mechanism.
I'd also be curious about how this beats the 'garbage in/garbage out' problem: Blockchain storage makes it very tricky to munge records after the fact without being noticed; but have absolutely no effect on any of the steps where someone needs to accurately and honestly observe the state of the world at a given step and write that into the record. The fun cryptographic integrity doesn't kick in until you have the electronic record; the process by which that record is produced is unaffected; and where a variety of opportunities for adding lies present themselves.
Isn't this just an electronic document? Couldn't you just put a QR code on the containers that linked to a copy of all the paperwork online, or fire a trigger that good have been received? Once the code is scanned at its destination an email of the documents are sent to all parties that the good have been received. You also log the ip address of the device making the web request as proof that it was done on the dock (or receiving warehouse, maybe include a pin that has to be entered as well).
If you wanted every party to sign off, you could make those e-mails instead link to a document management system that as each email link is clicked you record the party has seen the document, and you give them a simple "accept"/"deny" links to click. I seem to remember Sharepoint having something like this. Why the need for blockchain? Seems like they created a convoluted way of using blockchain so they could say blockchain.
The last time IBM was slightly good was when they got on the Linux hype train early during the dotcom boom. But then they blew it by going all in on outsourcing to India. I guess they think they can get some of the magic back if they jump on the blockchain boom early, too bad the blockchain boom already busted. Womp Womp. Glad I dumped my IBM stock a long time ago.
It doesn't. Blockchain is one of the most over hyped technologies of the last two decades. There are some things where it might be useful but most of the time these companies are just using where a standard database would work just as well. Sometimes better. This is one of those times.
I read at +2. If your post doesn't reach that level I will not see or respond to it.
You can't go back and forge early documents at a later point.
You can't go into the DB and just decide to edit a port of origin as all of the metadata that goes into a record also depends on all of the metadata prior to it.
Give me root access to your databases and I'll make your products show up from anywhere in the world. What happens when Orange company gets sold and Orange2, Inc wants to scrub any incriminating data showing Oranges from region X caused cancer? Edit the database.
If you can figure out how to Edit the blockchain you'll be richer doing that than bothering with oranges.
Holy shit. Good point. How is that supposed to work exactly? Might as well just use a fucking regular old database...
The Bill Of Lading Electronic Registry Organization (BOLERO) has been allowing shipping companies to use electronic, cryptographically secured bills of lading for more than 20 years. Unfortunately they don't use !!!BLOCKCHAIN!!!, so obviously the IBM solution is the only real alternative to using paper Bills of Lading.
If intelligent life is too complex to evolve on its own, who designed God?
I suppose the question is how much corruption/fraud requires a party to go back and modify records rather than falsifying them in the first place.
XML is like violence. If it doesn't solve the problem, use more.
For their next demonstration they are going to track a shipment of apples (the fruit). This will allow them to claim that chain has conclusively proved that you really can compare apples to oranges. (The next press release.)
I'm a little confused. Which part of this actually requires a blockchain?
And lets just call blockchains what they are.. Basically, a secure distributed hashtable.
I understand that blockchain is a nice buzzword similar to "the cloud" that encourages people to be interested in it. But we really need to start calling out dumbasses (who in this case appears to be a blockchain company) who pushes this trash.
I'm sorry the 90s called and they want their "e" back.
Harrison's Postulate - "For every action there is an equal and opposite criticism"
mandarin oranges are a symbol of prosperity, IBM explained
That needs to be explained? By IBM?
Everything is a "symbol of prosperity" in Chinese culture. There isn't anything in Chinese culture that doesn't come down to being about wanting to have lots of money.
Those who do not learn from commit history are doomed to regress it.
ROFD beats Blockchain every time. Regular Old Fucking Database.
"falsified in the first place." This is easy to guard against at discrete points in time, for example, when the shipment of oranges changes hands. This is possible by having all parties sign off on the transaction at that time, and a brief wait while several more blocks from other parties are added to the chain. At this point, the data cannot be retroactively falsified, and a single customs agent can assure this with a one-time signature.
And that is all immaterial. We have ePedigree for drug tracking though supply chain, it was signed and non-repudiated and the final recipient had all the history. The distribution companies shot it down (like had a regulation repealed). Now we have a DSCSA, distributed brokers for searches, information about one transaction exist in only one repository, the implementation are horrific, and there are retractions and corrections.
No industry will work on public non-repudation where every member can see the entire history end to end, even if it is a regulatory mandate. All the big players will have their conies elected to power and the regulations changed.
Block chain is some zitsy tech nerd wet dream. The actual world doesn't want it.
The point is these aren't actual blockchains because there are no thousands or more of computers calculating hashes.
To edit these so-called "blockchains" you just edit what you want and re-calculate the hashes, no problem there. Then you need to make sure to replace all references to the "final" hash.
It's about the same level of "difficult" as rewriting the history of a git repo. A PITA and if someone has the original final hash they might notice, but not really a challenging task.
* Immutable technology already exists for 'ordniary' DBs. Think something like git.
* 'Editing' a blockchain can well be possible depending on the consensus algorithm.
* Really, no one so far considered mutability a problem with DBs. There are several more important considerations - e.g. speed, availability, (eventual) consistency - which are far more important.
It only delivered "approximately" 108,000 fruits.
but if I had to make a wild guess I'd say it lets them offload some of the database processing to external "nodes", e.g. to another company.
Think about how much computer time costs for a large database. Now imagine if you could shift that cost to somebody else...
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
YES YOU CAN VERY VERY VERY VERY EASILY FORGE A BLOCKCHAIN. The art is in coming up with a way to stop that (e.g, bitcoin). When you catch up with the rest of the class you can raise your hand and post.
....and we've been using express B/Ls (ie no originals "mailed" anywhere) since you had to fax them before email was a widespread thing.
When original docs are required, ie for a letter of credit, they still use consigned BLs to really they're docs that bottleneck the PAYMENT chain, not so much the goods at all.
As far as proof of ownership, perhaps some people use them in sketchy parts of the world, but "to order" BLs that *actually* are a negotiable title to the goods...I haven't seen one in at least 20 years. Otherwise all transactions are consigned bills of lading in which case title is transferred to the consignee according to the Incoterms.
Short version: this is a complex solution to something that's not a problem, and solves nothing (that I can see) that IS a problem in this business space.
-Styopa
"...For the pilot, IBM created an electronic bill of lading, or e-BL, which helped reduce and speed up administrative processes âoeto just one secondâ as the document flow is automated, the company claims â" while the standard paper-based procedure takes five to seven days...."
Um, BLs are required to be produced within 48 hours of sailing, and that's been the standard since at least the 1980s. Usually it's less than that, and in fact with the implementation of ISF carriers are forced to issue BL numbers (the key bit) BEFORE THE VESSEL HAS EVEN SAILED (ISF is only relevant to the US, but I believe this process has ended up speeding BL# issuance worldwide anyway).
And a vanishingly small % of any shipments are issued with paper BL's anyway. I haven't seen a BL in the MAIL for 15+ years.
-Styopa
OK, so they created an e-BL and stored it on a blockchain. Whoopee, I can do that with my toy Multichain setup in 30 seconds. Now what?
How many participants were on that blockchain? Was the transfer of custody of the oranges recorded via asset transfer on that blockchain each step of the way? Was the e-BL used as a contract document by legal entities who received it via the blockchain, and did a bank connected to the blockchain use it as proof of collateral, as the article implies?
Because there have been a lot of big-name exercises in practical applications of Blockchain which then turned out to be a single node using it as a simple database, like the much-ballyhooed World Food Program blockchain exercise did.
Updated by IBM:
"You get 28 tons, and what do you get?
Another day older and deeper in debt.
St. Peter don't you take me, I won't go,
I owe my soul to the company store!"
There is a lot of difference between a blockchain consortium among a fixed set of semi-trusted parties, and a centralized database.
In the former, the parties trust each other, but only as far as they can throw each other. The network is full of competitors and those who could see substantial if fraudulent transactions could occur in their favour.
This is why Blockchain has a large advantage over a central database for this kind of scenario - not only are all the transactions on the chain and immutable, the code that operates the chain, is on the chain itself - which means that all parties on the chain have to agree to any code changes around the transactions, which means they know no one is messing with anything without everyone else knowing about it. You can't do this kind of scenario with a central database.
Turbo laser 5000 tracking system! It's the shit!!
Please... This is a paperwork issue and blockchain changes nothing, it's just one of many possible implementations. Seems overly complicated for this application. As others have pointed out, a secure auditable DB is all that's needed.
Yup, if the nodes in the blockchain are all controlled by a single entity, an ROFD would do. Or maybe just a text file & notepad.
The problem blockchains solve is one of distrusting peers having a way to perform trust-able transactions with a trust-able transaction history. If you remove the "distrusting peers" part, surprisingly, you end up with untrustable transactions and transaction history.
- The Kessel run is for nerf herders. I can circumnavigate the entire Central Finite Curve in a lot less than 12 parse
Sure you can. You can write whatever you want in a block chain. The only restriction is that to make everything line up, you have to recompute the hashes for all the subsequent blocks. Thatâ(TM)s not hard either.
Now, if you publicize the data, someone might catch you doing it. Just like they would if you published a blog entry or tweet you regretted and deleted it.
Block chains arenâ(TM)t magic.
git isn't "immutable".
If you and I share a git repo and you force commit to remove a change, I could detect it but still have no way of determining to legal standards who did what.
Just wait until someone figures out a way to combine AI and blockchain! Then you're going to hear hype.
I will agree to blockchain tracking of shipping if you mail me the bytes or kilobytes, black ink on paper as hexadecimal or base64.