I often think about that canal future that never happened. How sick would it be to get a boat to work everyday
post
So is the move by the car industry to get rid of mass transit in cities long ago a parallel to AI being swept in to replace people as some better solution only to find out later than it's not so great but now getting people back is difficult?
The thing about the transit conspiracy is that at that time many people wanted rid of street cars and wanted more space for more cars. It's not like Ford came to town in roaming gangs, ripping up street car tracks. They fed a desire among the public for a modern car-brained society and the public were happy to help.
There's no way that the long term billing for AI is tokens. It's not budgetable, it's arbitrary, and eventually people will figure out ways to either game it or evade it.
It's much more likely that they get crammed full of ads and become useless than it is that they become profitable charging via token IMO.
Free Version: Full of Ads
Premium Version: First adertised as Ad Free, then within a few years Full of Ads
Extreme Version: Ad Free for now, but features segregated into additional subscriptions.
The plan is for them to make agentic AI and charge a yearly salary per agent. They are obviously not close to that goal, or you would hear more about it.
Just had a funny thought:
User: Please write a boilerplate function that does
LLM:
//XYZ Function: This function does the thing...
//Subscribe to Incogni to prevent your valuable personal data from ending up in the hands of Data Brokers
function XYZ () {
...
There is applications and ways to make money. But competition isn't hard to come by and profit will be under expectations.
So my expectations is "skill-based programming" will become too expensive as a long-term solution and people who went that route will quickly become uncompetitive.
Whats funny is enshittifying is a suicide button when you can run deepseek or a different open weight model on your own machine. That i own. In my house. Without ads or censorship
Eh, it's close enough?
If they really wanted to, they could charge for "GPU time" directly. Usage gets a bit complicated with batching and context caching, but its definitely a metric that could be precisely derived.
Don't get me wrong though. That whole business model is doomed.
Pretty much all cloud compute is charged by usage (CPU time, storage, ingress, egress, class X api calls, etc), and is notoriously unpredictable. Businesses deal with it because, despite the unpredictability, it is cheaper than any other model could viably be offered at.
Consumer facing products will likely adopt different models; but "AI" is not a product, and most AI products will be bought by businesses.
Canals at least it was pretty easy to see a valuable use case.
Get commerce from New York to the Great Lakes without have to sail up past Maine and then still get past Niagara Falls somehow? That's amazing!
It just got outpaced by technology
Western Canada might not exist if it hadn't been for canals.
After the war of 1812, Canada / Britain (there wasn't really a separate Canada at that point) decided there needed to be a way to get military supplies and equipment to Fort York (Toronto) and west of there without using the St. Lawrence river. The river had a number of issues, but the biggest one was that the US controlled the southern shore. So, as a result, it built a canal from Ottawa to Kingston. Montreal to Ottawa was possible with existing rivers, and it was safe to get to Montreal from England without going too close to US territory.
So, the Rideau Canal was built. It connected Bytown (now Ottawa) to Kingston. For about 30 years it was the main way of moving people and supplies to the upper shore of Lake Ontario. After that it was replaced by railroads. But, without all the people who had moved west thanks to the canal and now could buy and sell things, there might not have been the justification to build those railroads.
So, yeah, it got outpaced by technology. But, it was a good idea, and was very useful for a few decades.
Also, many of those canals are now tourist attractions and pleasant natural spaces for residents. Virtually everywhere I've travelled where there is a canal, it is now used by people for pleasure boating, or houseboats. Almost always there's also a nice walking path or bike path next to the canal. Rivers are lovely, but rivers tend to have sloped banks which means there isn't a natural path that's right next to the water. With canals, there's often a man-made edge to the waterway, so you can walk / bike right next to it.
And the Panama and Suez Canals are STILL incredibly important to global shipping, greatly shortening trade routes where rail or road simply isn't an alternative
Aren't these both projects that had very strong long term benefits on shipping and exporting?
Shoutout to the Clinton Kalamazoo Canal
Likely the most ambitious and worst infrastructure project in state history.
A few counties like the UK and Netherlands got in on the canal game early. Nowadays, the remaining canal networks there are mostly used for leisure (there are a lot of houseboats in Amsterdam, London, the English Midlands and such), though occasionally they hold their own for specific freight (the spoil from tunnelling for high-speed rail lines in England was often moved by barge rather than convoys of lorries).
I live in the Netherlands, they still transport lots of cargo via canals. It’s one of the cheapest ways to transport heavy freight like cement. The canals connect to rivers so the barges can travel up the Rhine or Meuse to Germany . Canals aren’t just the waterways inside the cities.
Also a €7BN mega canal is currently under construction in Europe https://youtu.be/VdWQQ1OCUPo
Canals are still used to transport freight in Europe. For example there is a canal in southern Germany that connects the Main to the Danube, allowing ships to travel from the North Sea to the Black Sea through rivers and that canal. Germany generally has lots of navigable waterways including canals.
Yeah, I live in Jersey (New), and we have the Delaware and Raritan Canal, named after the rivers it connects. Was actually used for a bit, and the idea of using it is fascinating. But now it's a 44-mile path and waterway through history, it's very nice.
I used to live in an area that had a canal through it with what seemed like a nice walking path alongside it. Those paths are there because many boats going down the canals were actually pulled by horses walking along the path.
Unfortunately, the path next to the canal in my area was full of groundhog holes, and it was nearly impossible to bike it or run it without riding or stepping in a 8 inch hole, which is just big enough to do real damage. And since it's a useless canal, there is nobody assigned to maintain it, which is why it's become a haven for groundhogs.
Yeah, ours is fortunately a State Park, and gets maintained. The path is about 4-6 feet wide, depending, pea gravel the whole way, so good for walking, biking, whatever. You'll see people on the canal in the warmer months. It's a great resources and I'm fortunate to live so close to it.
Go out there and make the Friends of the Canal group, and demand the change to a maintained canal!
Oh good, so are you saying in about 7 years, this data center bullshit will end?
Depends on whether the AI boom is like the canal boom (fizzled), or the railroad boom (still going, arguably underinvested).
I'm pretty sure the above rhetoric was about the canal and data centers. Railroad was a side topic. So I choose to interpret it as within 7 years, data centers are gonna be repurposed into big box stores or filled with cobwebs & left to crumble to dust.
Except the railway boom isn't still going... it unarguably, unambiguously did fizzle, just like the canal boom before it, and for exactly the same reason too: it was replaced by a superior technology. In the case of canals, that was railways. In the case of the railways, that was motorised road-going vehicles.
all 69 comments