5 ms·

As somebody in container shipping: good riddance. There are a handful of sensible people who have been calling this out for years and years, and every time ther

As somebody in container shipping: good riddance. There are a handful of sensible people who have been calling this out for years and years, and every time there are wishful thinkers telling them this is the future. The worst of it is that there is an actual business case for a blockchain-like solution, because in flows where the shipper of the goods arranges their own hinterland transportation ("merchant haulage") there is a huge communication gap. The hinterland party and the shipping line need to arrange a handover of the container, and communicate about this constantly, but they are not in direct contact with eachother, because the shipper contracts them independently. T-mining (https://www.t-mining.be/ https://www.t-mining.be/, I'm not associated with them) is solving this for the release rights bit of that issue, but really this problem should be approached much wider to solve for seaship ETA updates and customs documents as well. Crucially, you can't solve this through a trusted third party, firstly because there is none that operates world-wide, and secondly because it would grant enormous power to that party, and no logistical actor is willing to give away this power. Of course the major shipping lines (Maersk, MSC, CMA-CGM) have been pushing to integrate hinterland services in their business more and more, so the shipper won't choose to arrange this themselves to begin with. This means they have no interest in fixing the communication problem. This is okay for shippers in the short term, but long term this will ruin the market. The shipping market is heavily centralized with only a few large players, which will now gain significant control over the hinterland as well and be able to force their policies on inland container depots, barge and train operators, and trucking companies. It's a shame really.

latchkey 4y ago
How do you start with 'good riddance' and then end with 'it's a shame really'? It sounds more and more like these people tried to solve social / political problem(s), with technology. Regardless of blockchain or not... it wasn't going to work until they got past the social issues.
- ArchD 4y ago
  He means the blockchain application in TFA is bullshit but the other situation he described regarding hinterland transportation is one that could benefit from a proper decentralized blockchain system but there is no such system.
  - miohtama 4y ago
    This is correct. I watched a presentation long time ago (way too early) about a blockchain pilot for container and document transfer between Finland and Estonia. https://www.porttechnology.org/wp-content/uploads/2019/05/036-038_3.pdf https://www.porttechnology.org/wp-content/uploads/2019/05/03... The author of the pilot told quite clearly that it is a killer for small and medium sized businesses, like few trucks and a man and a dog companies. Efficiencies could go up greatly. But then, when he pitched the idea to three letter logistic giant, he was thrown out from the board room because any decentralised network would kill the edge large logistics companies have. One of the main competitive edge for a large logistics company is that their internal IT systems work well and there is no problem with the information flow. They do not want to have a level playing field here.
    - sidewndr46 4y ago
      This is something lots of people often miss. The barriers to entry and the difficulty of using any system is someone's else's market edge. It's part of the reason why healthcare in the US is such a clusterfuck. It is the only industry with a degree of government mandates around price obfuscation. If you fix that, most hospitals would lose their edge in the market. Even the notorious automobile industry has clearer pricing.
  - wnissen 4y ago
    Exactly! IBM had a number of these that all involved "blockchain" but somehow also required that IBM host the blockchain in their cloud. I am no blockchain expert, but if you have IBM running the show, why not simply let them be the authority who runs a standard shared database with an API on top? Decentralization takes extra work, but it's worth it when it allows two parties to transact who, crucially, have no reason to trust each other and no convenient intermediary.
    - simne 4y ago
      I'm slightly involved in business, not in subject, but could answer. There are hundreds large ports in this world, and when ship transport your payload, owner MUST communicate with many govt and private structures. And these structures are not friends to each other, remember, for example one of channels is nationalized by Egypt govt, near it strait owned by Turkey, and they are not in war state just now, but near, and they all don't love US, and from time to time make nonfriendly actions, and could achieve US sanctions. And complex software needs good communication with vendor. Now imagine, how consultants from same US company will work, when one country is US ally, but other is under US sanctions? - Communications will be extremely hard. Open protocols and decentralization give perfect answer - nonfriendly countries will use blockchain, which is not owned by any subject, so they have not any reasons to block communications.
- oivey 4y ago
  GP is saying they were bullshit and sucking the air out of the room for fixing real problems.
  - superice 4y ago
    Precisely. I'm saying 'good riddance' to TradeLens, and I'm saying what a shame that effort by Maersk and other big carriers hasn't been put into one of the problem spaces where this might make sense.
RC_ITR 4y ago
> Crucially, you can't solve this through a trusted third party, firstly because there is none that operates world-wide, and secondly because it would grant enormous power to that party, and no logistical actor is willing to give away this power. This is one of the weirder aspects of crypto. Why would entrusting 1000’s of random people with GPUs be preferable to trusting one party (who, in a black chain scenario could be one of the GPU-holders anyway).
- idiotsecant 4y ago
  That's not weird at all because the whole point of crypto is that nobody trusts anyone. A properly designed crypto system is just a game of incentivizing each player to do the thing you want them to do while assuming that they're all out to screw each other.
  - RC_ITR 4y ago
    But what does that mean? Why does it need to be decentralized? Why can’t one party have the same incentive structure? Listen, I’m all for cryptography and Block-chains, but the whole proof-of thing has never made any sense to me. Maybe to make myself more clear, there are levels of trust, and you 100% “trust” that the miners on your random blockchain will do work for you in exchange for money. We’ve seen examples on even big blockchains like Solana where that trust is broken, even. So why exactly is that sort of trust better?
    - aftbit 4y ago
      > But what does that mean? It means that the GPU-holders are trusted only to make sure that all the other GPU holders are following the rules of the network. > Why does it need to be decentralized? Why can’t one party have the same incentive structure? Because if there's only one party, then who is watching them and ensuring they follow the rules? Furthermore, if there's only a handful of parties, who knows that they are not colluding? This is a major problem for proof-of-X tokens as well, as both mining and staking have centralizing forces.
duxup 4y ago
Am I interpreting this right as Maersk / IBM was solving what is basically an integration problem where there's all these parties and all this data not communicating and they decided the common interface was blockchain? I work in logistics and have some visibility to integrations with carriers (not much international, but some) and I find this very amusing. That sounds a bit like "nobody will agree on what cookie to eat, so let's make brownies, but what kind?". Half a dozen of one, 6 of the other.
woah 4y ago
Are you saying that the blockchain-for-shipping stuff might make the market more efficient with decentralization, but the big companies who have the power to sink or float it don't want this?
- superice 4y ago
  Right now the market is decentralized to a decent degree. The core business of Maersk is operating ships, and managing their equipment (containers). Their core business traditionally was NOT the hinterland transports from and to the ports. So traditionally shippers had to arrange this themselves. Around the world the terminology is different, but this is known as merchant haulage, a CY-CY booking, or a port-port transportation. Lots of communication needs to happen from shipping companies like Maersk to the party the shipper has contracted to pick up the container from the port (trucking companies, barge operators, train operators). One solution to this would have been a shared digital platform where the shipper could've granted access to the hinterland transporter to get updates about the container arrival, and to be able to cryptographically prove they are authorized to pick up the container. Whether this is an actual blockchain or something different also leveraging cryptography and distribution, I personally don't really care about. For the non-tech savvy people in the business, I'd call this 'blockchain' just to get it on the radar. The other solution to this handover problem is trying to force shippers to book the hinterland transport included in the sea transport, so Maersk would arrange that bit. This is called carrier haulage, CFS-CFS, or door-door transportation. This approach seems to be what they have settled on, and all the major shipping companies are investing aggressively into the hinterland parts of their business. Now imagine you're running a trucking company focused on ports (I think in the US this is called drayage). You've gone from 50 customers down to 3. At the same time major carriers are making record profits and have been explicitly exempted from the antitrust laws based on an argument of economies of scale in ocean shipping. Looking at it through this lens (pun intended), yeah, I'm a bit worried. If you were a shipping company, would you go with the technical solution, or would you try to get more of the entire process under your control so you can just strong-arm the hinterland parties into doing "the correct thing" ?
yunohn 4y ago
> secondly because it would grant enormous power to that party, and no logistical actor is willing to give away this power. This is a crucial point. When you speak to crypto enthusiasts, they speak about oracles. However, all the oracles they build are for crypto coin prices… not IRL applications.

latchkey 4y ago

How do you start with 'good riddance' and then end with 'it's a shame really'? It sounds more and more like these people tried to solve social / political problem(s), with technology. Regardless of blockchain or not... it wasn't going to work until they got past the social issues.

ArchD 4y ago
He means the blockchain application in TFA is bullshit but the other situation he described regarding hinterland transportation is one that could benefit from a proper decentralized blockchain system but there is no such system.
- miohtama 4y ago
  This is correct. I watched a presentation long time ago (way too early) about a blockchain pilot for container and document transfer between Finland and Estonia. https://www.porttechnology.org/wp-content/uploads/2019/05/036-038_3.pdf https://www.porttechnology.org/wp-content/uploads/2019/05/03... The author of the pilot told quite clearly that it is a killer for small and medium sized businesses, like few trucks and a man and a dog companies. Efficiencies could go up greatly. But then, when he pitched the idea to three letter logistic giant, he was thrown out from the board room because any decentralised network would kill the edge large logistics companies have. One of the main competitive edge for a large logistics company is that their internal IT systems work well and there is no problem with the information flow. They do not want to have a level playing field here.
  - sidewndr46 4y ago
    This is something lots of people often miss. The barriers to entry and the difficulty of using any system is someone's else's market edge. It's part of the reason why healthcare in the US is such a clusterfuck. It is the only industry with a degree of government mandates around price obfuscation. If you fix that, most hospitals would lose their edge in the market. Even the notorious automobile industry has clearer pricing.
- wnissen 4y ago
  Exactly! IBM had a number of these that all involved "blockchain" but somehow also required that IBM host the blockchain in their cloud. I am no blockchain expert, but if you have IBM running the show, why not simply let them be the authority who runs a standard shared database with an API on top? Decentralization takes extra work, but it's worth it when it allows two parties to transact who, crucially, have no reason to trust each other and no convenient intermediary.
  - simne 4y ago
    I'm slightly involved in business, not in subject, but could answer. There are hundreds large ports in this world, and when ship transport your payload, owner MUST communicate with many govt and private structures. And these structures are not friends to each other, remember, for example one of channels is nationalized by Egypt govt, near it strait owned by Turkey, and they are not in war state just now, but near, and they all don't love US, and from time to time make nonfriendly actions, and could achieve US sanctions. And complex software needs good communication with vendor. Now imagine, how consultants from same US company will work, when one country is US ally, but other is under US sanctions? - Communications will be extremely hard. Open protocols and decentralization give perfect answer - nonfriendly countries will use blockchain, which is not owned by any subject, so they have not any reasons to block communications.
oivey 4y ago
GP is saying they were bullshit and sucking the air out of the room for fixing real problems.
- superice 4y ago
  Precisely. I'm saying 'good riddance' to TradeLens, and I'm saying what a shame that effort by Maersk and other big carriers hasn't been put into one of the problem spaces where this might make sense.

RC_ITR 4y ago

> Crucially, you can't solve this through a trusted third party, firstly because there is none that operates world-wide, and secondly because it would grant enormous power to that party, and no logistical actor is willing to give away this power. This is one of the weirder aspects of crypto. Why would entrusting 1000’s of random people with GPUs be preferable to trusting one party (who, in a black chain scenario could be one of the GPU-holders anyway).

idiotsecant 4y ago
That's not weird at all because the whole point of crypto is that nobody trusts anyone. A properly designed crypto system is just a game of incentivizing each player to do the thing you want them to do while assuming that they're all out to screw each other.
- RC_ITR 4y ago
  But what does that mean? Why does it need to be decentralized? Why can’t one party have the same incentive structure? Listen, I’m all for cryptography and Block-chains, but the whole proof-of thing has never made any sense to me. Maybe to make myself more clear, there are levels of trust, and you 100% “trust” that the miners on your random blockchain will do work for you in exchange for money. We’ve seen examples on even big blockchains like Solana where that trust is broken, even. So why exactly is that sort of trust better?
  - aftbit 4y ago
    > But what does that mean? It means that the GPU-holders are trusted only to make sure that all the other GPU holders are following the rules of the network. > Why does it need to be decentralized? Why can’t one party have the same incentive structure? Because if there's only one party, then who is watching them and ensuring they follow the rules? Furthermore, if there's only a handful of parties, who knows that they are not colluding? This is a major problem for proof-of-X tokens as well, as both mining and staking have centralizing forces.

duxup 4y ago

Am I interpreting this right as Maersk / IBM was solving what is basically an integration problem where there's all these parties and all this data not communicating and they decided the common interface was blockchain? I work in logistics and have some visibility to integrations with carriers (not much international, but some) and I find this very amusing. That sounds a bit like "nobody will agree on what cookie to eat, so let's make brownies, but what kind?". Half a dozen of one, 6 of the other.

woah 4y ago

Are you saying that the blockchain-for-shipping stuff might make the market more efficient with decentralization, but the big companies who have the power to sink or float it don't want this?

superice 4y ago
Right now the market is decentralized to a decent degree. The core business of Maersk is operating ships, and managing their equipment (containers). Their core business traditionally was NOT the hinterland transports from and to the ports. So traditionally shippers had to arrange this themselves. Around the world the terminology is different, but this is known as merchant haulage, a CY-CY booking, or a port-port transportation. Lots of communication needs to happen from shipping companies like Maersk to the party the shipper has contracted to pick up the container from the port (trucking companies, barge operators, train operators). One solution to this would have been a shared digital platform where the shipper could've granted access to the hinterland transporter to get updates about the container arrival, and to be able to cryptographically prove they are authorized to pick up the container. Whether this is an actual blockchain or something different also leveraging cryptography and distribution, I personally don't really care about. For the non-tech savvy people in the business, I'd call this 'blockchain' just to get it on the radar. The other solution to this handover problem is trying to force shippers to book the hinterland transport included in the sea transport, so Maersk would arrange that bit. This is called carrier haulage, CFS-CFS, or door-door transportation. This approach seems to be what they have settled on, and all the major shipping companies are investing aggressively into the hinterland parts of their business. Now imagine you're running a trucking company focused on ports (I think in the US this is called drayage). You've gone from 50 customers down to 3. At the same time major carriers are making record profits and have been explicitly exempted from the antitrust laws based on an argument of economies of scale in ocean shipping. Looking at it through this lens (pun intended), yeah, I'm a bit worried. If you were a shipping company, would you go with the technical solution, or would you try to get more of the entire process under your control so you can just strong-arm the hinterland parties into doing "the correct thing" ?

yunohn 4y ago

> secondly because it would grant enormous power to that party, and no logistical actor is willing to give away this power. This is a crucial point. When you speak to crypto enthusiasts, they speak about oracles. However, all the oracles they build are for crypto coin prices… not IRL applications.