The Suez Canal will be widened
businessinsider.com
businessinsider.com
https://www.popularmechanics.com/science/environment/a359688...
The accident is probably just a nice catalyst or PR move.
Here's an expert whose discussion is very interesting
https://www.youtube.com/watch?v=CZ14E289jvw
Also, the canal generates critical foreign dollars for Egypt. They're probably slightly panicked about the proposed Israeli canal, particularly as transit companies are (one would assume!) much more open to alternatives.
It's probably what they think is achievable at acceptable costs. SUEZMAX is 400m long, but there's no way they can increase the canal width to 400m (at depth) to ensure no ship ever touches both banks at once. The minimum width is currently 205m at 11m deep (and the bank slope is 1:3 to 1:4 so a surface width of at least 270m), the work to double that would be insane.
Edit: maybe there is a curve in the canal and the radius of the curve limits the length of ship for a given width. Widening it at the curve would let you build longer...
Each individual sees a small disadvantage when a bridge is closed, but the overall economic effect might be very substantial, and hard to measure.
Basically with car, I went from home -> work -> home and unless I went really out of my way (pull into shopping center, park, get out of car, walk to store) then I wasn't near anything to buy. I basically had to make the conscious decision to go to a store.
Where as now, every time I enter or exit a train station I pass by 15-20 store fronts (book stores, bakeries, coffee stands, dessert stands, lunch stands, clothing stores, drug stores, grocery store, etc...) and it's literally 1-2 seconds off my path to be in one and buy something.
In various countries sprawls, shopping districts and office-only areas are prohibited by building regulations in order to create mixed areas where people can live/shop/work and build local communities.
Do you have any links? that goes entirely against my intuition. SF is generally considered walkable and has high street crime. I'm pretty sure the same is true for NYC, Paris, Berlin, other highly walkable cities where as say Irvine, Brea, (E.T. like neighborhoods that require a car to do anything) have low-street crime.
Update:
The first 2 hits I found on google found higher walkability correlated with more crime
https://journals.sagepub.com/doi/abs/10.1177/001391652092184...
https://link.springer.com/article/10.1057/s41284-018-00161-7
https://en.wikipedia.org/wiki/Walkability#Socioeconomic
As the sibling poster wrote, by comparing the only walkable area in the downtown of a car-centered city you are [involuntarily] cherry-picking the populace that cannot afford a car. You have to compare whole countries.
I can't remember the link, but one study I read found that it mainly affected whether people took multiple trips or one to do the same number of things.
E.g. if no traffic, you'll pop to 5 different shops on 5 different days because it's easy.
If tons of traffic, you save it all up to go to the same 5 shops in a single trip instead.
So obviously the effects are complicated, and the consequences aren't obvious. Also, if it's harder to make it to the middle of the big city, that can stimulate more vibrant jobs/shops/activity/etc. in towns on the outskirts that weren't there before.
So while all the examples you've chosen are detrimental, you can also reframe them all as positive.
On the other hand I don't think it's funds that are the problem here, there's plenty of money chasing more urban lifestyles. The limits are on the zoning and permitting for urban housing types in the locations that need them. Once there is capacity for people to urbanize, then we can talk tactics to nudge the reluctant.
If you just make the suburbs painful without upzoning the city, all you've done is decrease quality of life.
If that's not the case, I only know of one other solution: charge directly for usage. That way you can set the price such that demand stays at a high level, but at one supported by maximum available capacity.
If someone knows of a different solution, I'd like to know -- I have an induced demand problem in a type of infrastructure that I would like to operate at high capacity, and charging for it would be unconventional to the point of making it hard to find support for that solution.
> If someone knows of a different solution, I'd like to know
Some cities have areas that allow unlimited access to locals and deliveries and limited access to others (X times per year)
Making things free at the point of use is not always the best way to help the poor or public. The inefficiency of no charge first come first serve could outweigh equity benefits. Having everyone pay in time (traffic jams) is a large hidden cost. Time is valuable even for those who lack money. The US Department of Transportation came up with a benchmark of $12.25/hour (50% of median household income) to evaluate the expected time savings of proposed transportation projects (2014).
With free roads only time, fuel, and vehicle depreciation are considered. Roads are considered zero cost. There's minimal consideration of mass transit, carpooling, or moving closer to work. Just demands to add another lane and more parking everywhere. If roads were tolled or a lane was re-allocated to buses that could help poor people by making carpooling or riding a bus a realistic lower cost alternative to owning a car (or second car). Carpooling is a great example of how tolls incentivize high efficiency: 2 coworkers coming from the same neighborhood can cut their vehicle ownership expenses, the toll only hits half as hard, the people/hour throughput of the road goes up, and they take up less parking. Two people get to avoid congestion doubling time savings.
Infrastructure is always shared. There are hard normative questions about how capacity should be allocated, and who should be subsidized. Making it free at the point of use doesn't eliminate those trade off questions.
> If not then all roads should be toll roads or cars can be taxed per mile via odometer readings
...or you tax the combination of car usage and fuel efficiency by taxing fuel consumption. Which is already done with steep taxes on fuel sale in some countries and it's discouraging car use. For some people.
To be clear: the infrastructure I'm working on isn't transport, but it'd still be odd to only allow paying users. I was thinking two similar copies of this piece infrastructure, where one is priced to avoid congestion, and the other will be as congested as induced demand allows.
I'd like the discussion to stay general because I suspect any solutions will be reusable.
If you could get to Japan in 20m, I would probably pop over for lunch right now or Vladivostok for a Sunday stroll.
As transport improves, people work/shop/etc further from home. Commuting range is measured in time, not kms. So, capacity (eg traffic) tends to be the balancing force.
That's more than the typical Jevons Paradox, I think.
You can own 2, or 20, or 200 houses, and block the usage of them from anyone else. (induced demand with unlimited consumption). You can own 2 or 20 or 200 cars, but you can't physically drive more than one at a time. This is a hard cap on transportation consumption (you can't literally drive two cars at once, you can't be going to two different places at the same time)
This is why the induced demand philosophy doesn't work for cars. Demand for transit has a low hard cap. If your society is broken such that there's lots of latent unserved demand, it will artificially look like there's induced demand happening near the beginning. But that demand will vanish as soon as you get close to a functional level of capacity.
For a more practical example: If you could get to Japan in 20m, then you might choose to do so, And if you could go to lunch in Japan, you can't also be having lunch in California, which means you can't be waiting in traffic on the Bay Bridge. The induced demand philosophy obsesses over the "downside" of improved transit (that you get to go to cooler places faster and cheaper), but intentionally ignores the upside (that you free up the capacity of whatever you had chosen to do prior, and now no longer do).
You can own 20 houses, not renting them out. People just wouldn't, because it's pointless. You could also own 97 iphones, or 400 pairs of jeans. Latent demand just that people would buy more if it was cheaper. That's normal, not broken.
IDk what "artificially look like there's induced demand happening" means. If we could we would travel many more KMs. 1000s of times more, if teleporting existed and made travel instantaneous. The cost of travel, is largely time.
Urbanists sometimes intentionally undercount existing traffic and overcount later traffic, to try to paint public transportation infrastructure in a bad light. "We widened the public transit freeway by an extra lane each way, and now it handles ~33% more people per day. See, freeways make traffic, public transportation is bad"
The public transit freeway in this instance did not make any new traffic. It moved traffic from other sidestreets, and it allowed people who were underserved by transport to finally get served by transport. But nothing new was created here.
It's like saying, "Hospital A has 100 sick patients and is full, but we can't build a new Hospital, because it will make more sick people". And then a new hospital B gets built, and they say, "see, before we had 100 patients, now we have 200 sick patients, the new hospital induced more sickness!" But in truth, the Hospital didn't "make" any sickness, it's treating patients that were previously uncounted or not getting treatment they needed -- that's why we wanted to build a new hospital in the first place.
> If we could we would travel many more KMs. 1000s of times more, if teleporting existed and made travel instantaneous. The cost of travel, is largely time.
Right, and that acts as a hard cap. You can't travel to two places at a time, and you aren't going to want to spend every second of your life traveling somewhere. So, unless you think Winnebago's are going to be a thing everyone lives and works out of 24/7/365, there's a very explicit hard cap on transportation usage.
If you make a teleporter between California and Japan (for example), then an urbanist could paint that as "your teleporter is bad, because it induced demand for trips to Japan". But it didn't -- for every one person in your teleporter, there's one less person on an international flight. Even if your teleporter makes trips to Japan more popular and common (because it's now faster + cheaper to get there), it still doesn't break this hard cap, those trips are disappearing from somewhere else.
I live in Michigan, I would normally never grab lunch in Japan. But if I leave to go to Japan for lunch (using your new teleporter), then I can't be driving on US-131 to pick up lunch back at home (a totally normal thing I would do today). I'm not going to suddenly eat 2 lunches just because your teleporter exists, so even though someone would argue you have "induced demand" for trips to Japan, you haven't really done so, you've just taken other trips and converted them.
The only way "induced demand" will ever apply to cars, is if that hard cap is broken somehow. Driverless cars, for example, could actually induce demand, since a car no longer has to have a person in it to take up traffic space, and plenty of people would send driverless cars around, there's no longer a natural hard cap on transit usage.
This is why autonomous vehicles are concerning. It's going to become possible to consume a lot more transportation without using your own time. (Think: empty vehicles driving around rather than being parked, or empty autonomous taxis just driving around.)
The underlying argument behind induced demand is if you spend $XX million USD on widening a highway from 6 lanes to 8 lanes and now capacity has gone up NNNNN people per hour, that can be a worse investment than spending $XX-YY on million commuter buses going down an HOV lane to get the same NNNNN capacity boost, or maybe spend the same amount but get NNNNN+MMMMM boost.
We should optimize for people per hour, not cars per hour.
Work from home gets us another step further: instead of moving office workers bodies downtown so they can sit at a desk, why not eliminate the commute entirely?
Cool, that means they no longer live in their old city, so the transportation infrastructure of their old city is now freed up. No congestion was created, it just got moved (by place and/or time). No "demand" was "induced".
It really comes down to whether you want to encourage or discourage the use of the infrastructure.
In a suburb, the solution isn't anything except building more and better roads.
People aren't bicycling 10-15 miles to the grocery.
They should be, but, that's a different societal problem.
When I lived in a tiny Cambridge commuter village 30-minute cycle out from where I worked, the local grocery store was closer to me than one corner of Milton Tesco car park is from its opposite.
Here in Berlin? First place I stayed, out in the suburbs, the basics were a bit further — 300m for a bakery — but there were also seven supermarkets within 1 km.
Instead of cars making trips within cities faster, we started tolerating the same commute but spreading cities out more.
†5-10 minutes (?).
That's not a solution. Suburbs are self-bankrupting by definition: they're so sparse the taxes can't cover the infrastructure cost.
The solution is to relax zoning and make the suburb denser and liveable.
Removing and narrowing roads is a good way to do that: it increases space available for non-car infrastructures and produces free surface for "neighbourhood businesses". It also naturally routes through traffic around the area, rather than having to force that through inconvenient culs-de-sacs and making the suburb into a maze which requires driving 10 minutes to go to the neighbor's house.
Apparently you can't just put a generic "grocery" as the destination, but if you search for "grocery" and check different results, you'll get 10-15mn for all of them.
That's pretty standard US suburban sprawl, usually under the insanity that is euclidean zoning: entire neighbourhoods zoned exclusively for detached single-family homes, not allowing any business (no bakery, grocery store, florist, no nothing) and no multi-family dwellings.
There are, of course, other supermarkets much closer (2 within walking distance) but they are not quite so big, have a limited choice and are situated in areas which attract a certain class of clientel (known in the UK as Chavs). For him, the inconvenience is worth it - although it does take 3 hours out of my day.
I don't get going to a supermarket at all. We've had groceries delivered once or twice a week from Sainsburys, Ocada, Tesco, Asda, etc for about 10 years.
The concept of walking around a supermarket on a regular basis fills me with dread - all that wasted time and effort.
Not in the US. In the US "in the middle of nowhere" is grocery runs of 50+ miles. 15 miles is a "far out" suburbian hellscape e.g. https://i1.wp.com/cdn-images-1.medium.com/max/1080/1*AEfVhyv... or https://bostonglobe-prod.cdn.arcpublishing.com/resizer/h2N-h...
Speaking of which, whatever happened to the EV, pedestrian and bicycle tunnel under tje bosfora starting at kabatas which he claimed would be finished 2 years ago, along woth the 4 levant metro extension.
Are you suggesting that Istanbul or London would be just fine with a single bridge?
1. You can demolish large chunks of the city so that fewer people have a reason to visit, and spread development paper-thin over a wide area so that it stays beneath the incredibly low density threshold that causes car congestion. This is the approach taken by most US cities after WW2, however it has only worked in cities that have gone into economic decline because the places with meaningful growth consistently fill in more densely than cars can accommodate even with half the old downtown reduced to parking lots and strict zoning laws etc.
2. You can charge for the roads. Congestion pricing deters people who don’t really need to drive at a particular place and time from doing so, and thus allows the roads to flow freely even in urban centers. Surprisingly, it doesn’t take a huge toll to eliminate rush hour, as (1) the tipping point for traffic congestion is roughly the last 10% of cars that can fit on the road, and (2) it turns out that there’s a lot of low-value driving on the margin which simply stops when there’s a price attached to it.
The smart option is #2.
Option 4: Build excellent public transport so that cars are not the best choice.
Option 4 is also very difficult to achieve. Even Tokyo, with its legendary transit coverage, also has a ton of car traffic. What is much easier is to recognize that if everyone plows into their car for every trip we’ll all be stuck in gridlock, then put congestion charges in place so that everyone who can feasibly get their trip accomplished without driving is incentivized to do so, and the trips that really require a car are able to pay the fee and then have a nice trip quality.
I lived a few months in Tokyo, without a car of course, and got by very well with a shitty mamachari and the subway. I wonder what the modal split in Tokyo is like, compared to other cities. I bet most person kilometers are met by trains and subways.
That's less about removing congestion and more about clearing the roads of poor people to make room for everyone else unless you pay road toll as a % of income and assets.
It becomes more important to reserve "transfer nodes" (especially at the edges) where drivers can drop their car and hop on other transport modes though. That is something which is often lacking.
So you wait a little bit at the meter but your (and everyone else's) overall travel time is actually faster. I recommend this YouTube video (and the whole channel) for great explanations of the traffic dynamics.
When the line from the meters spills onto the adjancement roadway, some people will take alternate routes, but mostly people want to get onto the freeway because it's the best route, even if it's congested.
I'm not sure Jevon's paradox applies to houses, whereas it absolutely applies to car traffic.
> adding one or more roads to a road network can slow down overall traffic flow through it.
You can't have a car free metropolis. That's a contradiction.
It rarely delivers enough utility to even compensate for it costs.
> You can't have a car free metropolis. That's a contradiction.
It's absolutely not, your assertion is outright nonsensical.
Car-centricity is by far the worst way to build a metropolis.
And as points of difference (aside from size), Amsterdam has a much smaller metro network, but a larger tram network (Paris completely shut down its tram system in the late 30s and has been rebuilding it since the 90s, whereas Amsterdam's tram network was never completely shut down, and while a number of Amsterdam's tram lines were switched over to buses in the 40s and early 50s, trams got repopularized very quickly, as soon as the late 50s).
Makes sense. London is blessed with apparently empty high rise apartment blocks. 'Investment' flats, bought by foreign owners, who haven't seen their places and have no interest in them. It's just a safe place to store their wealth and they don't care if they screw up everyone who's actually trying to buy a home. Tax empty homes, should help things a lot.
Now, my idea to solve congestion is a bit different. For useless, bloated big SUV's, place restrictions and taxes on exterior options like alloys and paints. Incentivise smaller cars. That should help congestion a bit. And is it just me or are even supposed minis absurdly bloated and elongated these days?
Containerships don't conform to the canal. Both the ships and the canal are made to conform to the standard.
There is a document titled Rules of Navigation issued by the Suez Canal Authority which defines the maximum dimensions for the vessels authorised to transit. You can read it yourself: https://www.suezcanal.gov.eg/FlipPDFDocuments/Rules%20of%20N... (it's on the 90th page as the pdf reader numbers them, or 68th by the document's internal numbering)
Based on what you say it sounds like you think the current limits are experimentally found. As if ship owners build all kind of ships and then try to cram them through the channel and see which one goes. That would be madness.
Containerships absolutely do conform to the canals. Lots of ships are built to the maximum size that can transit the Panama or Suez canals. It's called Panamax and Suezmax ships. When the Panama canal was expanded, that caused larger ships to be built to the new specifications of the canal (called new Panamax).
If this expansion causes the maximum dimensions allowed through the Suez canal to be increased, ships will be built to that larger standard.
When you order a ship and you stipulate that it must be able to cross the Suez Canal the shipyard won't send out two dudes with a theodolite to measure how wide the canal is. They will look up the standard (the very same I have just linked!) and build the ship to that specification. That is to say the ships conform to the standard not to the canal.
Similarly if you are engineer working for the SCA how do you know when to dredge the canal? Easy. You read the same standard and you dredge where ships of the published maximum size would not fit. Likewise you check the published air draft maximums before designing over-water electric transmission crossings. That is to say you make the canal conform to the standard.
> If this expansion causes the maximum dimensions allowed through the Suez canal to be increased...
True. But there is zero reason why the SCA would be coy about announcing a neo-Suezmax standard if that is what they would be wanting to do. In fact they would float any such plans years and years in advance, because vessels are not built in a day. I have seen no indication for that.
In reality, the maximum air draft (how much the ship is sticking out over the water) is also limited by the Panamamax standard. It is constrained because the ships has to go under bridges and some cables too while crossing the canal.
So for the Suez that'd mean there shouldn't be any danger of that happening. The EverGiven was stuck length-wise, and if they're not getting longer then at worst they can become square, which is probably not the right shape for a ship.
The limits are way beyond what's at issue here. There are lots of ships which are already too big to go through the Suez canal.
Canals tend to limit dimensions based on:
* the canal width (limits beam)
* the canal curves (limits beam and length)
* the canal depth (limits draft)
* locks (limits length)
* bridges over the canal (limits height)
"Capesize" is the general category of ships which can't go through the Suez or Panama canals[0], and must therefore go around Cape Agulhas and / or Cape Horn.
[0] although with the deepening of the Suez canal it's now possible for many capesize to go through Suez, just not Panama.
How does a ship being longer cause stress on the steel?
The longer the ship is the stronger these forces can be.
Also, drag on a hull is such that longer hulls tend to be more efficient.
Follow this thought to "let's just make these trains as long as possible." You can imagine some wild consequences, both positive and negative.
Still, a far more effective solution would be increase width and draught likewise to reach your target volume. And that's where we're back at ${canal}max classes (locks also impose a length limit but I'd assume that those are much easier to adapt than width/depth).
If Suez Canal authorities are serious about trying to prevent future Ever Given incidents (how serious they are likely depends on the exact outcome of the ongoing liability fights) they should keep some parts of the canal at the existing profile (locks perhaps?) to prevent suezmax 2.0 (or whatever their current version+1 is)
We could make a much bigger , lighter efficient ship with carbon fiber entirely, however there is no economic incentive to spend 100x on the ship, fuel costs and economies of scale rtc don't help recover that capital cost
Expensive would be OK, of course a bigger ship costs more to build than a small one, but you hope to make the investment back by carrying more cargo and additionally benefit from some economies of scale.
However, the biggest ships we have are at the limit of how thick we can join steel plates with current welding technologies. So there's some non-linearity here.
The costs of all of it needs to be economically feasible even if the technology is there, I don't think welding is only reason we have been limited in the size
Who pays for this dredging and these cranes? The localities. It's difficult to defray these costs with shipping fees. American President Lines pulled their headquarters from Oakland to Phoenix when Oakland did that. [2]
[1] https://en.wikipedia.org/wiki/CMA_CGM_Benjamin_Franklin
[2] https://www.prlog.org/10551489-apl-phoenix-headquarters-relo...
I am not saying the limit doesn't play a major role, I am saying if it's was the main bottleneck there would be a lot of r&d in solving it, solving it would certainly help, but the economics and other limitations mean solving welding won't mean a revolutionary change to ship sizes
Any attempts at standardization would actually be worse, because now they build each ship for exactly one actual canal. With standardization they'd get built to some document that will get interpreted slightly different in every place.
The alternative is simply to continue as it is, with ship builders squeezing every inch just because they can, and it becomes a matter of if, not when, another EverGiven happens.
But as the Evergiven incident has shown, it has a limit. We were lucky it didn't get stuck worse than it has.
At some point it makes sense to just blow up the ship to free the channel.
I think it's always interesting to get another perspective and I am somewhat disappointed that people have downvoted you without even wanting to hear your reasons. Such conversations are always an opportunity to either learn or to teach for those that are open to it.
Then I’ll refine my statement to bridges and such.
https://www.businessinsider.com/us-planned-suez-canal-altern...
Otherwise carpet-bomb with nuclear warheads should do the trick as well.
https://en.wikipedia.org/wiki/Mediterranean_Sea#Invasive_spe...
> A first look at some groups of exotic species shows that more than 70% of the non-indigenous decapods and about 63% of the exotic fishes occurring in the Mediterranean are of Indo-Pacific origin, introduced into the Mediterranean through the Suez Canal. This makes the Canal the first pathway of arrival of alien species into the Mediterranean.
[0] https://www.theguardian.com/world/2017/jun/24/asian-carp-gre...
https://en.wikipedia.org/wiki/List_of_Lessepsian_migrant_spe...
Wider could have a major effect on terrestrial non flying mammals but zero effect on aquatic animals (except maybe for the biggest whales). Deeper yes, would made a difference for aquatic species.
Maybe would be a good chance to made two roads separated by a chain of islands instead. More expensive but not without extra benefits. By the way reduces the ever present risk of too long ships. With two isolated lanes you can rely always in the other to keep the road opened.
Those chain of islands in the middle could easily hold infrastructures or attract tourism. And would be a great opportunity to became a sanctuary for migrant birds and fishes without blocking so much the migrant mammals.
And if in the future they need to wide even the channel they could just keep the old docks and remove one island in strategic points to connect both lanes and make space.
And could be done in several steps without blocking the activity or income in the old channel.
If anything they may want to limit Max crossing to calm days. Also the crew's judgement should be examined: wind forces in tight quarters are a big concern for any vessel and yet they decided to enter.
If you ask for a Xm long ship, you get an Xm long ship. That's not a design flaw because it crashes more easily than a smaller ship when not maintained/piloted properly...
Edit: if the Suez officials are serious about this, they should limit the size of vessels to the current maximum.
So shipbuilder builds a ship THIS BIG and it doesn't fit. [0]
I think that's the canal's fault, not the shipbuilder's.
[0] in fact, most of the time it fits, but sometimes it doesn't.
That’s a heck of a difference.
It is also likely cheaper to widen than dig a while new canal. They will revise prices for wider ships to justify the expense of the widening project. There is not enough transit traffic to recover a full two lane system today.
If there are two directional channels you have the option to use the other for both directions when one channel is blocked.
Or maybe the idea is to leave something for it to ground on that’s easily removable.
(saw references in comments, but c'mon, this needs to be a top-level comment, this is totally obvious)
Feet, knots, miles etc are still used for boats and anything that flies.
[0]: https://en.wikipedia.org/wiki/Nautical_mile
[1]: https://upload.wikimedia.org/wikipedia/commons/5/59/Nautic_m...
The article on suezmax[1] uses metric as the primary units, even in images/charts. The article on deadweight [2] explains that long tonnes were used historically but now metric tonnes are used. I don't know much about the units used for altitude in shipping
But if you check other sources on the same news (eg Chinese state media company Xinhua [1] used only metric, or the Russian state-owned RT [2] uses only metric, while the Canada-based Reuters [3] uses a bizarre mix of only-metric, only-imperial, and both) - They're all writing for their target audience. Without watching the original TV address, I couldn't say which was originally used.
[1] http://www.xinhuanet.com/english/africa/2021-05/16/c_1399483...
[2] https://www.rt.com/news/523483-suez-canal-expand-southern-st...
[3] https://www.reuters.com/world/suez-canal-chief-says-southern...
Edit: I just checked even the French ship registration forms have tonnage in long tons (imperial) and boat dimensions in feet...
/s
They are asking nearly a billion dollars to unstick it... doesn't matter how wide you make the canal when it's a cesspit of corruption.
<< A few days after the Ever Given was dislodged, the SCA impounded the ship and its cargo and lodged a compensation claim of $916 million.
The canal authority since reduced the claims to $600 million. However, the insurer of the vessel said this amount is still too high. >>
Not billions. Also, what do you mean by "cesspit of corruption"?
- You can reduce the program's memory usage, or increase the program's / system's alloted memory. But there's usually some hard limit to this - if you've got 100GB of data, you've got 100GB if data, and if you've got a non-distributed problem you'll eventually run out of RAM slots. Plus you usually just get induced demand when a customer shows up the next day with 200GB data.
- You can make the program handle the too-large-data case better. This requires deeper systemic changes usually across multiple components (the program, your monitoring and error reporting, upstream and downstream integrations). But this also tends to have knock-on improvements across the system. And a lot of times it's cheaper than the first option when you compare total rather than marginal cost.
Or to be more direct: Why fuss over 131x10 feet and not the fact your only recovery plan is one excavator and a handful of tugboats?
Also it seems like a win-win approach: harder for ships to get stuck, and room for more traffic.
The issue with your analogy is that here the "program" did not OOM as part of normal operations, instead it OOM'd because of a rare condition you can't really avoid (even calling it a bug is debatable, as a bug implies misbehaviour in the canal).
In that case, increasing your safety buffer so you can recover before swapping to death or the OOM killer -9-ing programs left and right is a good idea.
> - You can make the program handle the too-large-data case better. This requires deeper systemic changes usually across multiple components (the program, your monitoring and error reporting, upstream and downstream integrations). But this also tends to have knock-on improvements across the system.
How would you make "the program" handle "the too-large case better" exactly? As in, how would the canal handle a ship getting stuck sideways after having lost propulsion save by being wider than the ship can be long?
> And a lot of times it's cheaper than the first option when you compare total rather than marginal cost.
We're talking about a 200km canal.
> Why fuss over 131x10 feet and not the fact your only recovery plan is one excavator and a handful of tugboats?
Maybe they're working on both but increasing the canal width to provide more buffer in case of rare issues is a much more complicated and thus interesting thing, and thus gets more visibility?
One should identify both corrective measures and preventive measures in equal measure.