Your lanes are 200m wide, but you’ll have trouble with wind direction. If the wind is blowing from directly ahead, or up to about 45deg on either side of that depending on your ship design, you can’t sail forward. That’s ~25% of wind directions. If your lanes are perfectly straight and the wind is perfectly consistent within a day then that rules out 25% of days, but with a realistic path and variable winds it gets much worse.
This is why you see sailboats tack, and why they’re not a good fit for rivers.
My ‘lane’ was only to estimate density of vessels, to show that they would be very far apart—two minutes until you approach a stopped vessel ahead of you. At 5km wide, the entire plateau would be wide enough to tack. And, the Polar Vortex is persistent; check the weather where it’d happen, instead of assuming an evenly-distributed, equal probability of all orientations.
Tacking seems like it dramatically reduces the capacity of the plateau in practice? Unless you can ensure that all vehicles are moving exactly the same speed avoiding collisions is going to be a lot of work.
Um, check the numbers, again—I gave each ship a 2km by 200m ‘zone’ around itself, at my measure of full capacity. That’s not going to lead to them bumping together; it’s a hundred acres per ship. Stopping/maneuvering before you speed past a ship ahead of you? You have 2 and a half minutes to do so, even when they are at a dead-halt. And, you have hundreds of meters to either side. It’s bizarre that claims of ‘too close’ are made, when you don’t give pause to the actual measurements involved. It shows a lack of real effort.
Sorry, I’m not saying that ships are going to hit ones on a parallel track, the problem is that once you add tacking ships are interacting with others on perpendicular tracks which is a lot trickier
You’re welcome to check the numbers, again—I mentioned “5km wide should be plenty” as an illustration. If tacking led to disruptions, you could easily multiply the width of the ice-lane manifold; as mentioned, a spray-wall 10m tall with a 10m/sec arctic wind (the Polar Vortex there) would produce enough ice to cover a mile wide, 400 feet deep, when compacted… each year. You want it ten miles wide, 100m deep? That’ll only take 2.5 years. And, again, the amount of capital required is minimal—a hundred kilos per meter of shoreline, while the amount of ice produced is immense—hundreds of thousands of tons per winter. Similarly, for every 1 joule of energy that you spend spraying water, you are transferring 3,200 joules of heat to the air, as that water freezes. 3,200x inputs is not magically ‘too expensive, too hard’ just because ships zig-zag as they tack. You could claim that I would need 10x more ice, and that still wouldn’t make the design ‘impossible’ let alone unprofitable.
Given the dimensions of ship and ‘lane’ I described, then so long as ships turned-about at the edge of the plateau, then they could literally tack back-and-forth more than a HUNDRED times, blindfold, before the average collision. There is an immense amount of space between each vessel, and it’s a shame that the commenters on this site don’t realize such simple metrics when they claim “you’d crash into each other”—you show a lack of comprehension for the scale involved. Your claim does not stand-up when scrutinized in detail.
Will any of them admit that they were wrong, sans rebuttal? So many commenters here ghost as soon as the flaws in their argument are illuminated… It’s a demonstration of their unwillingness to admit fault, which might be related to why they like to frequent a website claiming to absolve them of error—“I must be correct, when I give hand-waving dismissals and I don’t account for the details, because I’m a regular of LessWrong.” Your site does a worse job of catching fallacious reasoning than the philosophy site I moderated 25 years ago, as a middle-schooler. You’re not less wrong than we were; you’re a huge step down in quality.
With only 2 tons per m2 loading (no stacking) a 500t cargo needs 25m x 10m footprint, which, compared to the ‘lane’ I gave each vessel, is only 1/20th the width of that lane and only 1/80th the length until the next unit of ‘lane’; literally only 1⁄1,600th of each lane is occupied by vessels, as the ‘full capacity’ I listed. It’s bizarre that tacking would magically absorb so much space that vessels occupying 1⁄1,600th of the available area would somehow collide.
Realistically, you can stack many tons on each m2 of ice—it can hold a hundred tons reliably. So, if you didn’t mind a really tall stack, your 500t ship could occupy a much smaller footprint.
Further, 500t was an easy value to plug-in, for estimates of total annual tonnage delivered. I don’t assume that 500t is the optimal scale; doubling each dimension of the ship would only increase its footprint by 4x while providing it with 8x as much room on the ice, 2.8x longer to respond to any disruption or potential collision.
I also have a feeling that “make sure you all don’t collide” is a pretty simple math problem for autonomous sailboats. Especially considering that you can geo-fence the track. Rio Tinto has been using autonomous vehicles on their mining sites, precisely because you’re unlikely to run over somebody’s dog there. Same reasoning applies to a hundred-foot-tall wall of ice at the top of the world; autonomy would allow dense traffic at low capital, again.
I have yet to hear a realistic critique; all these responses stem from the readers’ erroneous assumptions, or lack of digesting the scale described. In contrast, I’m scraping my brain for the best ways to solve the real engineering hurdles, like de-crusting the spray nozzles, because particulates exiting the stream would become seeds for ice-crystallization. You know, the actual problems that would happen, if we did it… not imaginary hyperbole like “you’d run into each other, even though there’s a mile and a half to maneuver between every ship”. The critique I receive on this site is full of fallacies and errors; its commenters are not a reliable source of insight.
Your lanes are 200m wide, but you’ll have trouble with wind direction. If the wind is blowing from directly ahead, or up to about 45deg on either side of that depending on your ship design, you can’t sail forward. That’s ~25% of wind directions. If your lanes are perfectly straight and the wind is perfectly consistent within a day then that rules out 25% of days, but with a realistic path and variable winds it gets much worse.
This is why you see sailboats tack, and why they’re not a good fit for rivers.
My ‘lane’ was only to estimate density of vessels, to show that they would be very far apart—two minutes until you approach a stopped vessel ahead of you. At 5km wide, the entire plateau would be wide enough to tack. And, the Polar Vortex is persistent; check the weather where it’d happen, instead of assuming an evenly-distributed, equal probability of all orientations.
Tacking seems like it dramatically reduces the capacity of the plateau in practice? Unless you can ensure that all vehicles are moving exactly the same speed avoiding collisions is going to be a lot of work.
Um, check the numbers, again—I gave each ship a 2km by 200m ‘zone’ around itself, at my measure of full capacity. That’s not going to lead to them bumping together; it’s a hundred acres per ship. Stopping/maneuvering before you speed past a ship ahead of you? You have 2 and a half minutes to do so, even when they are at a dead-halt. And, you have hundreds of meters to either side. It’s bizarre that claims of ‘too close’ are made, when you don’t give pause to the actual measurements involved. It shows a lack of real effort.
Sorry, I’m not saying that ships are going to hit ones on a parallel track, the problem is that once you add tacking ships are interacting with others on perpendicular tracks which is a lot trickier
You’re welcome to check the numbers, again—I mentioned “5km wide should be plenty” as an illustration. If tacking led to disruptions, you could easily multiply the width of the ice-lane manifold; as mentioned, a spray-wall 10m tall with a 10m/sec arctic wind (the Polar Vortex there) would produce enough ice to cover a mile wide, 400 feet deep, when compacted… each year. You want it ten miles wide, 100m deep? That’ll only take 2.5 years. And, again, the amount of capital required is minimal—a hundred kilos per meter of shoreline, while the amount of ice produced is immense—hundreds of thousands of tons per winter. Similarly, for every 1 joule of energy that you spend spraying water, you are transferring 3,200 joules of heat to the air, as that water freezes. 3,200x inputs is not magically ‘too expensive, too hard’ just because ships zig-zag as they tack. You could claim that I would need 10x more ice, and that still wouldn’t make the design ‘impossible’ let alone unprofitable.
This also might help:
Given the dimensions of ship and ‘lane’ I described, then so long as ships turned-about at the edge of the plateau, then they could literally tack back-and-forth more than a HUNDRED times, blindfold, before the average collision. There is an immense amount of space between each vessel, and it’s a shame that the commenters on this site don’t realize such simple metrics when they claim “you’d crash into each other”—you show a lack of comprehension for the scale involved. Your claim does not stand-up when scrutinized in detail.
Will any of them admit that they were wrong, sans rebuttal? So many commenters here ghost as soon as the flaws in their argument are illuminated… It’s a demonstration of their unwillingness to admit fault, which might be related to why they like to frequent a website claiming to absolve them of error—“I must be correct, when I give hand-waving dismissals and I don’t account for the details, because I’m a regular of LessWrong.” Your site does a worse job of catching fallacious reasoning than the philosophy site I moderated 25 years ago, as a middle-schooler. You’re not less wrong than we were; you’re a huge step down in quality.
For those who don’t want to do the mental math:
With only 2 tons per m2 loading (no stacking) a 500t cargo needs 25m x 10m footprint, which, compared to the ‘lane’ I gave each vessel, is only 1/20th the width of that lane and only 1/80th the length until the next unit of ‘lane’; literally only 1⁄1,600th of each lane is occupied by vessels, as the ‘full capacity’ I listed. It’s bizarre that tacking would magically absorb so much space that vessels occupying 1⁄1,600th of the available area would somehow collide.
Realistically, you can stack many tons on each m2 of ice—it can hold a hundred tons reliably. So, if you didn’t mind a really tall stack, your 500t ship could occupy a much smaller footprint.
Further, 500t was an easy value to plug-in, for estimates of total annual tonnage delivered. I don’t assume that 500t is the optimal scale; doubling each dimension of the ship would only increase its footprint by 4x while providing it with 8x as much room on the ice, 2.8x longer to respond to any disruption or potential collision.
I also have a feeling that “make sure you all don’t collide” is a pretty simple math problem for autonomous sailboats. Especially considering that you can geo-fence the track. Rio Tinto has been using autonomous vehicles on their mining sites, precisely because you’re unlikely to run over somebody’s dog there. Same reasoning applies to a hundred-foot-tall wall of ice at the top of the world; autonomy would allow dense traffic at low capital, again.
I have yet to hear a realistic critique; all these responses stem from the readers’ erroneous assumptions, or lack of digesting the scale described. In contrast, I’m scraping my brain for the best ways to solve the real engineering hurdles, like de-crusting the spray nozzles, because particulates exiting the stream would become seeds for ice-crystallization. You know, the actual problems that would happen, if we did it… not imaginary hyperbole like “you’d run into each other, even though there’s a mile and a half to maneuver between every ship”. The critique I receive on this site is full of fallacies and errors; its commenters are not a reliable source of insight.