Hacker News Reader

Comment by stephen_g | Hacker News Reader

It's a little bit different than building a McDonalds - they're building an underground rail line in my city, and just the acoustic sheds and site offices (before they could start even digging a shaft to drop in a tunnel boring machine) on one of the six or so major sites is like building a dozen McDonald restaurants in itself! Not to mention water treatment, spoil handling and treatment, ventilation systems, moving utilities etc. etc.

View on HN

In the old days, the New York subway systems were mostly built by cut and cover, where they more or less dig up the street, build the tunnels and stations,throw dirt back on top, and pave the road again.

Is that always possible? No. But when it is, it's almost certainly going to be much cheaper and faster.

For bonus points, throw all the ancient utility lines in conduit so the next infrastructure project isn't a disaster.

Nowadays we get the DC Metro, where a mostly above ground extension in a highway median was delayed for years because....I'm not sure exactly.

The DC metro is ridiculous. I'm cynical enough to think that the only way the Silver Line would have been done faster would have been to shutdown DCA so politicians would be forced to use Dulles. Even then they would probably just have drivers.

IIRC, there are issues with the concrete used on the new platforms too.

they're doing cut and cover on some fraction (maybe 1/3) of the purple line subway extension here in LA. it's still a bloated, expensive, late mess of a project, so technique matters little relative to incentive structure on total cost.

If you travel internationally you quickly realize all the excuses for why things take so long are just that. Be curious how long Tesla would take to build a huge full factory in China. In ca the permitting alone is on the order of years.

They built their factory in China in less than a year [0]. Tesla pushes equally hard everywhere which you can see in Germany where they pulled out every stop they could to build quickly and still took about 2 years [1].

0 - https://www.pmi.org/most-influential-projects-2020/50-most-i...

1 - https://en.m.wikipedia.org/wiki/Gigafactory_Berlin-Brandenbu...

The Berlin factory was larger then the initial factory in China. Its more like the second extension in China.

Seems like in the US Tesla can get something up and running within a couple years.

  May 2020, a selection process was underway by Tesla
  July 2020, Tesla selected Austin as the site
  July 2020, Construction began
  end of 2021, limited production of Model Y starts
  April 2022, initial deliveries of vehicles built at the factory took place at an opening party called “Cyber Rodeo”

https://en.wikipedia.org/wiki/Gigafactory_Texas

Most states and municipalities are pro business so they will move mountains for jobs but not for housing. This is a big problem because there are a lot of jobs in Silicon Valley but not enough houses (despite the layoffs).

It's not an excuse, and I am looking internationally (i.e. this is a project not in the US). Just explaining why the kind of solution proposed (ask McDonalds or someone similar how they build a small building reasonably quickly) is not going to help work out how to build major infrastructure quicker or cheaper!

They have a lot bigger budget than a single McDonald's as well.

What you say sounds good until you look around the world snd discover lots of other places have built similar in harder environments for much less and much faster. If all you know is McDonald's this might seem complex, but if you learn about other rail projects you discover this isn't that complex.

Reply on news.ycombinator.com