Edward Swernofsky

Karma: 52

Edward Swernofsky 14 Mar 2021 21:17 UTC
1 point
on: Make more land

Much higher densities are possible but not a good idea.

Not with that attitude! Also, Manhattan daytime and Paris densities are already much higher (~200k / sq mi).

Edward Swernofsky 9 Oct 2020 0:24 UTC
3 points
in reply to: eyesack’s comment on: Industrial literacy
Thanks again for the perspective! These are good things to note and provide a lot of context. I still wonder what qualifies as “family owned” and whether it’s really just farming that brings 60 million to rural life.

The median household income in rural America looks to be only a bit lower than urban. Otoh, the rural poverty rate was 16.4 percent in 2017, compared with 12.9 percent for urban areas.

Jason Crawford mentions farms worked with trains and horses before trucks. The scenario I mentioned with trains would still use (intermodal?) trucks for the last mile and just replace rural highways. I could believe farming transport demands are too strange for this, but I could also see standardization insignificantly increasing costs. And do people in town often travel to the farms or mostly just to other towns or cities?

Pollution isn’t just a local issue, and I agree rural areas have no obvious pollution—but a carbon tax (for global warming) would make fuel more expensive, increasing the already significant costs of rural gas transportation.

I imagine the biggest subsidy of rural areas is the highways, which are ³⁄₄ of the paved lane-miles in the US. The maintenance of these highways appear to amount to ~$3600 per capita annually, with a subsidy of ~$1200. I’d believe that utilities aren’t subsidized more than their urban counterparts.

If as this implies there really isn’t much subsidy, I stand corrected! Thank you, cars. And of course, any other technology (electric, PRT) would still function like the automobile. And AC / climate control is necessary in many states.

Now I wonder how rural areas look in other countries wrt population share, infrastructure, economy, and farm finances.

Edward Swernofsky 6 Oct 2020 22:16 UTC
7 points
in reply to: eyesack’s comment on: Industrial literacy
This is clarifying of rural life. Thank you!

I think you make a lot of assumptions of what I believe here.

Large family owned farms constitute about half of total farm area. It’s not really clear to me what qualifies as “family owned” here: I imagine most still have a number of workers.

I’m also not sure if farms are the primary driver of rural economies. They certainly occupy most of the area. Rural areas appear to take the form of vast swaths of nothing but farms surrounding tiny suburb-density towns. I think there’s a good chance that without the subsidy and with more direct infrastructure burden, the tiny towns (which seem to be most of the rural population) would significantly shrink in population.

That said, I admit (electric? micro?) cars or motorcycles seem like a perfectly reasonable transport system for the remaining rural population. Farmers might need higher clearance for field roads. PRT would likely work for small towns and inter-town transport, but not for farms, and wouldn’t have been possible until recently. If highways are much cheaper with smaller vehicles and without trucks, I imagine rail would be a good alternative for farmers to ship produce.

The main point I was making is that rural areas have significantly more road, utility cost, and transport cost per capita. Owning a car is expensive, and unnecessary in many cities.

You’re not wrong that NYC is a bit insane (and I don’t think “a hundredth” is out of the question in many cities), but the added value seems to generally outweigh the increased waste of most cities even today. Pollution can be mitigated with incentives, and I’d be surprised if rural areas don’t pollute more per capita.

Jerry-rigging everything is a compromise you don’t need to make in many cities! I’d argue that that sort of thing is just a market inefficiency, and actually more wasteful. If you wouldn’t jerry-rig in a city, it’s probably because paying someone is actually a better deal overall, ignoring regulations.

Edward Swernofsky 1 Oct 2020 0:13 UTC
8 points
in reply to: jasoncrawford’s comment on: Industrial literacy
No, just that the following doesn’t make sense if people have more kids as a result:

half of everyone you know over the age of five is alive today only because of antibiotics, vaccines, and sanitizing chemicals in our water supply

I suppose this is fuzzy, but you could also argue no one you know would be alive in such a counterfactual because all their genes and experiences would be different as well.

This is pretty much pedantry, but you could’ve phrased it just as “there’s been a huge reduction in childhood death”.

Edward Swernofsky 30 Sep 2020 21:54 UTC
4 points
on: Industrial literacy
If infant mortality was higher that’d be terrible, but I assume people would have more kids to compensate. Automobiles other than trucks don’t help much; they enable rural settlements, but not many live there and they are disproportionally expensive. I bet they’d be a lot less populated w/o all the subsidy. Cars also may not last long with a carbon tax. And cars actively harm cities substantially.

Professionals in these fields don’t know some of these either. And why should everyone else? I’d go the other way: Untangle policy from the masses.

Edward Swernofsky 9 Sep 2020 0:15 UTC
1 point
in reply to: mako yass’s comment on: A Scalable Urban Design and the Single Building City
I attempted to address rent stuff in a reply to your other comment. I generally agree on psuedoexterior spaces.

I expect if intentional communities or micro placement coordination are valuable enough and rent/space unregulated, communities will outbid people with weaker ties both in construction and allocation. It’s possible you can overcome the frictional costs of all this allocation being centrally planned, but I’m skeptical of the overall value.

Edward Swernofsky 8 Sep 2020 23:28 UTC
1 point
in reply to: mako yass’s comment on: A Scalable Urban Design and the Single Building City
I think the issue I have with this is: Density is expensive and may not be necessary. Let the market decide if it’s a good idea. Deferring the density question to another system begs the question of how efficiently it matches the market and how it works. The city can still tax away land value increases and give that money back in other ways, or limit the total tax. A new city doesn’t exist in a vacuum. It competes in a broader space market and overall market of tax burden. Pricing these too low may just limit city growth.

The price on usable space being close to the value is an effective means of prioritizing who lives where as well as initial motivation to build. That value can lower if the supply increases, or homogenize with good transportation. Land value lowers some from land value increases being taken by the city. A Harberger tax may also cap property prices.

Land prices generally can’t rise significantly until multi-story wood construction saturates buildable land. Depending on whether cross-laminated timber is about as cheap as traditional wood construction and built space per capita, that should happen at a threshold of either around 200k or 1.4m ppl per sq mi. Even if land prices rise significantly, that price is amortized over the number of floors afforded.

Edward Swernofsky 5 Sep 2020 7:22 UTC
8 points
1
on: Will autonomous cars be more economical/efficient as shared urban transit than busses or trains, and by how much? What’s some good research on this?
Hey there! I researched a lot of these things recently and made a post here about it.

Some particularly relevant things to your post: The yet non-existent (at scale) system you describe is Personal Rapid Transit (PRT). The best podcar design is likely a one-seater with a fully-reclined passenger. Fully-reclined minimizes air resistance along with used road volume, but people may prefer partially-reclined for comfort.

One great measure of efficiency is total cost of ownership per mile. Cars have a minimum total cost of ownership per mile 3x ($0.5 vs < $0.2) that of trains, but are admittedly pretty big.

Space is at a premium in cities, and mass transit makes 15-30x more efficient use of it than cars. It’s possible PRT can match this with smaller vehicles and tiny headways. Busses go a lot slower than trains, limiting their effectiveness. 75% of bus cost is the driver in the US. They tend to be used more as feeders for trains in many cities.

You can look at other countries for answers to some questions: Paris and Tokyo have pretty cheap housing and amazing train systems and still have commuting.

One cool thing about PRT is the vehicles can be super light, possibly allowing the road itself to be cheaper (especially bridges and raised sections). I think PRT is most promising as a transport system for lower densities since trains seem almost as good in cities. Otoh lower densities come with higher road network and utility cost.

Alon Levy has a blog on everything transit related and is absolutely meticulous. He’s not infallible, but a great resource for tons of different metrics.

Strong Towns is a great book on municipal finances and incentive structures.

CLT construction may result in 35-floor buildings as cheap per sq-ft as typical 5-floor wood buildings today.

I’d love to talk sometime.

Edward Swernofsky 30 Jul 2020 22:17 UTC
1 point
in reply to: SurvivalBias’s comment on: A Scalable Urban Design and the Single Building City
Thanks again for your input!

Sunlight/windows/view/air

OK, I might not understand what you’re saying here. I agree that this is the primary issue people raise and that means this isn’t discussed enough, but I figured that would dominate the article if I focused on it.

A few things… Circulating air even from all the way outside is a lot cheaper and easier to deal with than the sunlight issue, so fresh air is really not a huge issue.

You mention “worst neighborhoods” near the ground implying there is a section of the city where crime is high, but there are no streets near the bottom: Any poor people living below are living in a small space accessible by stairs or elevators directly from the streets at the top.

Idk if this is complete, but here’s a capitalist argument: People vote with their feet. They can always live further out and commute in. If the conditions are actually worse than another place, poorer folks will try it out and leave to enjoy life somewhere more reasonable. If the full negative sunlight/window economic externalities caused by putting up a new building are assessed and charged to the new owner (I should include this), the owner won’t build if he (and others) can’t get enough return on the lower sections. If people don’t like not having sunlight and they’re still there, maybe it’s legitimately worth it to them? For example, maybe they’re earning way more money than they would otherwise.

Indeed you mention dense thin rows of buildings, but doesn’t it change the whole calculus here? And more or less turns this into a Manhattan with underground car tunnels, or something close to it? I’m also not quite convinced this approach will work that well even in California, in the sense of making people happy with their view. Plus, when there’s a lot of another problem arises with these shafts—they heat up very quickly without wind. Yes, ventilation, but that costs money and I don’t think I’ve ever seen such a “ventilated courtyard”, so probably nontrivial amount of money.

I think dense thin rows of buildings can still achieve very high per-floor area ratios, maybe ²⁄₃ or so, allowing most of the ridiculously high density Manhattan doesn’t come close to. View can probably be included in the negative externalities charged to new construction. Hot air rises, giving a free ventilation force. Just want to note here that moving this amount of air with fans really isn’t that expensive anyway.

What about truck noise?

Yeah, 50 ft is just the figure I found. Trucks would also likely be going slower than 30 mph. “For every doubling of distance, the sound level reduces by 6 decibels.” Soundproofing is a combination of reflection and absorption; Add absorption materials to reduce echos. People live next to streets with trucks using them already, right?

Snow

Fair enough, how about pushing snow into chutes or tubs and having trucks ship it out? Or throw it into the trash AVAC system?

Robo delivery theft

Tbh fast response times, dash cams, high use streets, and police should be enough to handle this even with mixed traffic, although the separated infrastructure pays for itself readily in robot speed and throughput. Navigation can be done either with paint and optical sensors, magnetic markers, or fixed digital broadcasts from frequent points.

Seeing Like a State

OK, the point I’m trying to make here is that these critiques apply as much to existing suburbs and cities as they do here in the sense that: Centralized systems like mass transit, utilities, and emergency services are everywhere. Regulations and codes around building and transportation are everywhere. Government spending on centralized systems and many more things is also everywhere. The amount of top down planning used in existing municipalities is already quite high, and I think this about matches it, not being notably more other than just actually addressing super-high density development. I just don’t think the comparison to master-planning is actually reasonable.

If anything this design moves more things from a flat “no, everything has to be this way” to a “sure if it’s actually worth it” approach to regulation. I’m sure there are things about peoples’ preferences I’m not aware of here, and it’s important to make things a little flexible and open to later change because of that. However, I’m not sure you have a better understanding, either. Existing systems have made the decision for us on a lot of matters and may realistically not have a lot of reason to those decisions.

Edward Swernofsky 30 Jul 2020 20:09 UTC
1 point
in reply to: lincolnquirk’s comment on: A Scalable Urban Design and the Single Building City
Well, the Charter Cities Institute is trying to help make stuff along these lines happen, but they’re probably not as interested in super-high density things. I’ve heard Singapore tends to have better urban planning than other countries. Many European countries and Japan have nice train systems, but still have a lot of cars as well.

Edward Swernofsky 28 Jul 2020 19:18 UTC
1 point
in reply to: lsusr’s comment on: A Scalable Urban Design and the Single Building City
Thanks, updated.

Edward Swernofsky 28 Jul 2020 19:14 UTC
7 points
in reply to: SurvivalBias’s comment on: A Scalable Urban Design and the Single Building City
First, I want to thank you for thinking critically about this. I appreciate your efforts and line of reasoning.

Poor people tend to want sunlight more and are less able to afford it here.

So first, note that this is correlation and not a direct relationship. I’d love to see a study here.

I think this also is mostly resolved by the market: Poor people wanting sunlight will live more crowded in sunlit sections. Middle class people not needing sunlight will live further down in less crowded conditions. Rich people will just pay the premium and live in spacious conditions with sun.

If sunlight turns out to be a huge constraint, it reduces maximum density considerably but still allows for densities multiple times the densest cities we see today.

Columns of outside don’t work in high latitudes for people that want sunlight.

I next mention dense thin rows of buildings lining streets. The streets would be north/south to handle this.

You need streets with not much shade for some climates.

Yeah, the streets with shade thing is just an example to show walkability in nice climates. I agree you can’t do all of these in such climates, and that may make them poor choices to starting such a city there.

What happens if you need to rebuild something on the 3rd floor underneath a walking street?

You can have thinner walking streets supported by the buildings to the side as well as the roof of the building supporting it, but in general you may have to shut down that street when rebuilding it.

How far can cranes reach anyway?

Typically 230 ft, so ideally you have a grid of crane points spaced so half the diagonal is that distance. With a square grid that gives 320 ft between adjacent grid points. You can either space streets to line up with this grid or the grid will be more constrained somehow.

Police and medical need to move quickly; fire and medical need to carry stuff.

Police, fire, and medical can still use truck lanes to carry a bunch of stuff quickly in an emergency. 15 mph is faster when things are much closer together. Elevators can bring them with equipment to the street level, where so long as their equipment is relatively lightweight, it can still be electrically powered and easy to move.

What about maneuvering fire hoses outside high floors?

You can use the building across the street, ladders on the outside of that building, extendable emergency bridges, something like a really tall maneuverable ladder that attaches to rails on another building, or a helicopter with a platform dangling from it.

What about truck noise and air pollution?

Heavy trucks are about 80 db at 50 ft going 30 mph. I think soundproofing a 60 db reduction isn’t that expensive, maybe $2 / sq ft of wall in the US? Since the air pollution is concentrated in one area, you can use air filters to get rid of most of it. Ventilation is also a lot cheaper than full HVAC.

Drainage system failure effects are multiplied.

Yeah, you either need very good drainage reliability and throughput or this won’t work in certain climates.

Snow

I’m not too sure about this; maybe salt the streets or use airplanes to blanket everything in salt? Maybe heat smaller sections and have people push snow into those sections? Also, you can still have some space on streets or roofs for heavy machinery in exchange for higher construction costs for that part of the roof.

Cities stink and higher density multiplies this.

So, crowding is the same. I don’t think either of us knows the exact causes of city stink, which would be helpful here. Air filters and ventilation should help address this.

Parking lanes are necessary to prevent failures from blocking traffic.

This depends a bit on how close you need parking to be in the case of failure. There’s still some parking, and even a parking lane can be filled with cars, as is usually the case in Manhattan. I imagine the solution here is adding paid parking spots until this is not a huge issue or having low latency towing.

Robo delivery theft

If robots are in a fully enclosed tube, it’s still pretty hard to just take one. The robots can also communicate position and video live with a computer system and the city can respond quickly if something is off.

transport accessibility

This already isn’t an issue with trains. Lightweight electric wheelchairs shouldn’t be an issue on the street. A few ramps at changing street levels can accommodate forced street level transitions.

rooftop HVAC street/park pollution

Air filters and lower surface area help a lot with stink and cooling here. Piping ventilation from buildings under streets through side buildings should help with concentrating this into parks. From there, concentrating HVAC units on rooftops minimizes interaction with walking space. You can also do most of the heating/cooling with ground-based heat pumps to reduce these units entirely to huge fans.

This creates a whole lot of new infrastructure every new building needs to be plugged into, as well as new requirements (ability to support extra weight, noise insulation), which makes new development a lot more expensive.

Definitely more expensive; However, I don’t think it’s that much more: Noise insulation is pretty cheap. The extra weight supported here is the same eventual overall weight. Again, this expense is borne when it becomes worth it anyway.

Seeing Like a State

*Note: I’ve also only read Scott’s post.

This is important and I want to address this further. Note that the design here doesn’t say “here’s exactly how the buildings are laid out and what they’re used for”, nor “we’re designing for this fixed density”. Furthermore, the design is not a whole lot different in scope from what we see today in many suburbs and cities. Grids are pretty common, and Jane Jacobs (sort of the antithesis of Le Corbusier, the father of Brasilia’s design) basically thinks they’re great (and a lot better than suburb culdesacs). A grid is fine in optimal conditions, and if the terrain and environment make that a dumb idea just drop the grid and design around the terrain. It’s not critical to have a perfect grid or anything.

Edward Swernofsky 26 Jul 2020 16:44 UTC
1 point
in reply to: romeostevensit’s comment on: A Scalable Urban Design and the Single Building City
Just want to first note that most of the proposed things require no innovation either.

Multistory buildings (also regulated away in most suburbs) adds ~25% to construction costs. Jane Jacobs also mentions parks being much better if they’re on the way to things, and the ratio of park even today doesn’t approach the scale portrayed. Also want to note that suburbs only exist in a few countries.

poor people create more externalities

Poor countries with cities still exist, and many seem to be getting along well enough (at least getting richer over time). If it’s not that I guess my intuition goes the other way. I imagine police stepping up militance to combat some of these things can work. If discrimination is legitimately a good solution to some urban issues, I’d at least be surprised, but it should still work with the other things here.

suburbs seem really appealing to families who want more space

The housing market should generally address this. There’s will certainly eventually be parts of the city topped out in density, but until then if space becomes valuable the market will eat that value by making more space. Families can always live in less dense (but still city-density) sections a little further out.