Hammersmith Bridge – Where did 25k vehicles go? | Hacker News Reader

Hammersmith Bridge – Where did 25k vehicles go?

nickmaini.substack.com

63 pointstobr50 comments

I think the author was a bit too confident in their farming out of research to Claude. For instance, the claim that "at least 9 Chinese bridges have been built that would span the English Channel" is obviously false.

Railway viaducts (built mostly over land):

- Danyang-Kunshan Grand Bridge

- Tianjin Grand Bridge

- Cangde Grand Bridge

- Weinan Weihe Grand Bridge

- Beijing Grand Bridge

Not actually long enough to span the channel (excluding access roads etc.):

- Hong Kong-Zhuhai-Macau Bridge (main bridge is 30km)

- Jiaozhou Bay Bridge (all three legs combined total 26 km over water)

- Runyang Yangtze River Bridge (total length of two-bridge complex is 7.2 km)

Could actually span the channel (setting aside differences in water depth and whatnot):

- Hangzhou Bay Bridge

Not to say all of these projects are not extremely impressive, or that the article doesn't have a point. But making claims like this undermines the author's credibility, at least in my eyes.

I guess one of the new careers being expanded is rigorous fact checker.

I almost read beyond Part 1's introduction before looking at the comments.

But thanks to your comment I now know it was written with an AI and that makes me want to stop here.

If the author is asking an LLM to confirm his biases in order to make his point it will probably contain sycophantic hallucinations.

Britain is broke. That's probably all there is to say.

You can tell it's mostly AI slop by all of the unnecessary and inconsistent use of bold type throughout.

A couple interesting things I've come across over the years:

1. Western politics seems tragically reactionary and concerned with short-term issues. "Boring" stuff like infrastructure maintenance gets set aside. Deferred maintenance results in a superlinear increased expense: deferring $1 of maintenance today will cost you >$1 in the future (in real terms, accounting for inflation).

2. Some nations massively spend on some infrastructure with results little better than others.

This was an interesting, very long, read! They say of those 25,000 daily trips, most shifted to cycling, walking and public transport, and some moved to other bridges. And then another 9,000 or so were replaced by alternatives that were just better... people tried new transport modes and often found they were better. They do say the closure has created genuine hardship for specific groups.

I don't really buy it. I live in the area, and what happened is that traffic increased dramatically everywhere but Barnes, which is where Hammersmith bridge is. People in Barnes generally love it, as you can read in the author's tone.

London in general has a terrible problem of car commuters who travel 1-2 hours across the city every day. They're going to take whatever route necessary to do it.

The article opens up with a glaring mistake:

> Paris, 15th April 2019, 6:43pm.

> Inside the ancient cathedral all is quiet.

The fire started during the mass, so not fully silent. And a first fire alarm had already been sounded 20 min earlier.

Not disagreeing with the author's conclusion, but the price comparison to the original struck me as a bit odd.

Ceteris paribus, building the exact same bridge will result in the exact same failure. Some of the additional cost is precisely to avoid the present scenario repeating itself in the future.

How big that addition represents and how effective it is up for debate, but asking for a better bridge at inflation adjusted price is not a. apples to apples comparison.

If spending the 1887 price (adjusted only for inflation) got us an identical-to-1887 bridge, which lasted through another 125 years of mostly-neglected maintenance - very few people would refer to that as a failure.

Would we get the lighter 1887 loads and the cooler weather mentioned in the article too?

No - but that 1887 bridge did not fail under heavier modern loads. And realistically, sourcing load-bearing members as weak as the 1887-tech cast iron might cost far more than using "average quality" modern reproductions.

Temperature only seems to be an issue because of the now-seized bearings.

With the 250M price tag I really keep thinking how we in the west just accepted such a massive cost for infrastructure development, especially considering the cost of living has gone down and the Victorians typically built this things by hand.

Most people who say these things frequently do want all the other functionality one buys with that money, though. As an example, in times with lower safety standards, many projects proceeded without incident. The point of modern safety standards is to guarantee to a greater likelihood that a project will proceed without incident. Would you be willing to give that up?

Another concern is the loss of a historically listed structure. Most people today prioritize historical structures over any modern structure. Would you be willing to demolish the bridge? You certainly can't rebuild an identical one because we don't have that many expert workers of wrought iron.

It will have been built to older standards. You'll have to convince a lot of people that the weight standards of then, the fire standards of then, and the disaster management standards of then should be exempted from modern controls and in order for them to be exempted you need to create a framework for exemption if it doesn't already exist. Coordination costs a lot of time and money. Even deciding that you don't need coordination for this project requires coordination because without a framework for exempting coordination you can't do it without allowing for always exempting coordination.

You will have seen this in any other realm. The more people have an opinion on something the harder it is to get done. The union of all requirements creates a project that is the intersection of all possibilities enabled, which combined with the classic aphorism about every additional percent taking as much effort as everything before, means that things cost more now.

We can build better and faster when we don't have to listen to anyone. This happens in emergencies. Take a look at the US MacArthur Maze tank truck fire and rebuild.

I don't see the value in preserving obsolete infrastructure for historical purposes. Photograph it, document it, open a museum to commemorate it if you want, but blow it up and build a modern bridge that doesn't have all these problems and benefits from an additional 100+ years of progress in engineering and materials science.

Nor do I, but we are in the minority in anglophone civilizations (at least, perhaps others as well). And that's where the coordination cost comes in.

I see the value in preserving it as a museum if it's also replaced elsewhere with modern infrastructure.

It's expensive because we don't want the people who are doing the actual work to be living in a shantytown tent with no sewage.

It’s just another example of Baumol’s cost disease: https://en.wikipedia.org/wiki/Baumol_effect

Very few efficiency improvements have been made in bridge building over the past 150 years aside from prefabricating sections offsite and using hydraulic cranes. Inflation pushes wages higher, making it seem more expensive since there’s no efficiency gain, just higher wages for the workers. It’s good that the people that build bridges and roads and buildings can afford to live.

Preserving a 150 year old bridge gets complicated as it’s virtually bespoke work, problems are uncovered as the project continues, ballooning costs.

Well like the article points out - even that outrageous cost of 250M is nothing compared to the national budget. We as a country could easily afford it - but this is the second part of the whole problem, not just that things are expensive, but that no one is willing to actually sign off on any solution, so we just end up with the default of doing absolutely nothing. In a way it would have been better if the bridge actually collapsed(with no injuries to anyone, of course) because then it would have been much easier to replace it or repair it. Right now it's usable at least somewhat so there's all kinds of reasons why nothing should be done(money being just one of them).

A few people I know had moved to houses on one side of the bridge for easy access to schools and jobs on the other side, and were hit hard by the closure.

Their commute times skyrocketed to go to the next Thames crossing.

Putney and Chiswick bridges aren’t all that far, I regularly walk around.

> Six years later, Hammersmith Bridge remains closed to vehicles. The solution proposed by local authorities costs £250m and has no funding.

If this is not a failure of Mayor of London Sadiq Khan, then who? Where I grew up, new bridges were paid for by tolls until the building cost was paid then taxes covered occasional maintenance.

This is an automated AI slop substack that somehow got boosted onto the HN front page - third one of these I've seen in the last 2 months, the AI spam is getting better.

It's got nothing to do with anything, it's AI written slop, and the author is farming clickbait topics and articles with no coherent theme or perspective.

Wait, really? I read this entire article, quite enjoyed it(in fact I sent it to 3 of my friends already, as it's a topic that's very dear to me) and I hasn't even crossed my mind at all that it might be AI generated. Are you sure it's AI generated? If yes, how?

I don't think I'm as hyper sensitive to llm writing as some others here but this triggered me very quickly. I started scanning and selectively reading after the intro which was just completely soulless. After I reached the end of section 1 I closed tab.

It's cumbersome. Things like "This essay examines two questions." Bold is used wrong. Uniform style wall of text all the way, with identical size sections and everything. Section conclusion not reflecting its title. Looking closely I noticed a couple of other tells but I won't share them in case they are reading.

It's got that AI smell. The word choice. Antithesis ("not this, but that"). Persuasion words.

There's meat in it, it's not pure slop, but it was definitely fed through the slop factory.

And as some have mentioned, the facts are a bit dubious.

100% sure. It's low information repackaged bland pap with nothing to add with regard to perspective or novel information, meticulously referenced like a research paper, with the formatting and grammatical quirks of AI, the simultaneously broad, yet weirdly limited vocabulary, and so on. I'm sure if I spent more time I'd be able to articulate exactly which patterns I'm recognizing, but it's very recognizable. There are also similar patterns in previous content from that substack. Some effort has been put in to it, there aren't any emdashes, but it's slop. It's like a "ChatGPT, what's an interesting topic to write about that's relevant to region XYZ" "ok, thanks, write a detailed, interesting essay, but edit it to avoid all the AI tell-tales like em dashes." and so on until you get this article.

If it gets on the HN front page, it gets propagated and an insane level of visibility all over the internet - say, 100 million people see it over the course of 24 hours. If they charge $5 for an annual subscription and 1% (or .1% or even .01%) are so impressed with the content that they sign up, that's a lot of money. I can't find any other possible reason for this showing up, it's nothing to do with tech or AI or silicon valley or the usual weird eclectic stuff that gets discussed, lol.

Kudos to the author for knowing enough about how things work to make such an elegantly targeted campaign, I guess.

This doesn't read as slop to me. I think the author just overuses bold-emphasis and short sentences.

Here's a "Walk with me" video of the bridge: https://youtu.be/DA9NPmwWDWE?t=385

What is funny is that we can easily guess that the "blocked bridge" for car might look "harmless", but it is probably compounding on the cost and difficulty to do anything. And that it is probably similar small things that resulted in the current situation.

For example, restoration of a church or school nearby was maybe costing like 5 millions pounds, and will now cost 5.5 millions because construction material, workers, trucks and tractor take that much longer to reach and leave the place having to take deviation now that the bridge is not available anymore.

"...which is still closed and could take 20 years to fix, causing a major headache for drivers."

This isn't April 1st is it? Just contract the Chinese to fix it and it'll be done in a few months.

You would need Chinese laws and regulations. This is one of the reasons why when building Belt & Road Initiative projects in SE Asia, Africa, etc, China demands exclusion from local regulations and insulation from local politics, at least after initial negotiations and before work begins. In many nations the problem is corruption and kickbacks, but in a country like Britain the problem is bureaucratic red tape and "community input" (i.e. every Tom, Dick, and Harry effectively has veto power).

The pacing of this article is exhausting. It reads like shows that have to fill 30 minutes so repeat themselves over and over again.

And now I see it's AI slop. Great.

Article Summary: Why we can't have nice infrastructure any more. :(

I expected that to be the conclusion, but it's not. They could spend £250m on the bridge, but they're not. And it appears to be the right answer since it wouldn't provide anywhere near £250m worth of utility. They'd spend £250m to make things worse -- right now it's an awesome cyclist/pedestrian bridge, and after spending £250m it'd be much worse for that.

That's the takeaway I had as well. Spending a quarter of a billion pounds to get more cars into a traffic choked downtown is a bad investment. Spending that money on improving public transit options would improve the quality of life far more.

Barnes is not really “downtown” it has a rural village feel.

to me, the argument that we can't just print more money and do both because of a fear of inflation falls flat. We can't have nice things because it might be nice?

Reply on news.ycombinator.com