Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

People do work from home, but you wouldn't know it trying to drive anywhere around here.  The metro area adds nearly 100,000 people a year, most are going further and further out.  The city has been growing over the past 20 years or so after the usual declines.  But the metro area population is more than 10x the Atlanta city population.  Sprawl.

I wish we at least had a subway out to the baseball field, it's in a terrible traffic area, brilliant.

Heh. Chicago doesn't have a population growth problem (according to census data, at least). More people are leaving every year. I'm sure that has something to do with ridership too. 

All of the above plus lower maintenance costs, which will decrease the costs of rides from whoever survives between Uber/Lyft/Waymo/etc.

Still, I think autonomous driving is decoupled from EV. We could just as easily go autonomous with ICE engines as EV. The lower maintenance costs of EV only play a factor in comparison to ICE--if EV doesn't take over as quickly as projected, then EV isn't necessarily "better" for autonomous ride-hailing. 
Perhaps this could occur due to scale-up costs of increasing battery production capacity for a huge portion of the automotive fleet. I.e. I see the same thing in the HDD/SSD market. People talk about how wonderful SSDs are--and they're exactly that wonderful. But then they predict that HDDs are going the way of the dinosaur, and they're not. SSDs require NAND, and to build out the NAND fabrication facilities to meet exabyte demand would cost hundreds of billions of dollars, while flooding the market with NAND and dropping costs through the floor. Every NAND producer would lose their shirts if they did it. 
So I don't think it should just be assumed EV takes over. It might be the case, but I think if autonomous ride-hailing is ready before the market can supply EV, there's no reason they need to be coupled. 

The movements toward electrification and automation aren't interdependent, but they're not completely independent, either. All of the auto companies realize that EVs are going to win out eventually and that autonomous vehicles will make individual car ownership largely obsolete. How that all plays out is a big question but how it ends is pretty much understood.

As for mass transit, with automation, I agree that buses will prove to be the better solution. I've used buses in subways when I've lived in Philadelphia, Chicago, and DC. The problem with buses is they still have to deal with traffic, stop signs, stop lights, construction, etc. Subways are much more reliable for the most part, and they're scheduled more frequently, but I agree that expanding existing transit infrastructure is not going to happen for the most part (the only project in the DC metro is to finish building a line that goes to Dulles Airport), nor should it.

The movements toward electrification and automation aren't interdependent, but they're not completely independent, either. All of the auto companies realize that EVs are going to win out eventually and that autonomous vehicles will make individual car ownership largely obsolete. How that all plays out is a big question but how it ends is pretty much understood.

Well, I'm not sure that BEV is going to win out. I was just reading today that China is now betting on fuel cells. But they want 1M fuel cell cars on the road by 2030, so that's clearly a long-term goal. 
I think there's a significant issue trying to scale up battery production for BEV. Right now the cost of a BEV is heavily due to a battery, which is why Toyota is pushing hybrids so much more strongly than BEV. It allows them to achieve higher efficiency without having to have enough battery capacity to achieve range. 
That was my point with SSD vs HDD. It doesn't matter if BEV is technically superior to ICE if you can't produce enough batteries--at the cost necessary to get adoption--to achieve scale. Do we have enough mines worldwide to get the metals for these batteries? How much would it cost to expand mining? Do we have enough mining companies to supply that many metals, or do the mining companies need to expand? If the mining companies expand and flood the market with metals, can they sell them for a high enough price for the mining companies to cover costs? 
These are incredibly difficult questions, and currently BEV is a tiny portion of the worldwide automotive market. How long it takes to get to scale--if it's possible at all--is not a question we can ignore. 

As for mass transit, with automation, I agree that buses will prove to be the better solution. I've used buses in subways when I've lived in Philadelphia, Chicago, and DC. The problem with buses is they still have to deal with traffic, stop signs, stop lights, construction, etc. Subways are much more reliable for the most part, and they're scheduled more frequently, but I agree that expanding existing transit infrastructure is not going to happen for the most part (the only project in the DC metro is to finish building a line that goes to Dulles Airport), nor should it.

Subways make sense, but they require insane population density to be economic. So they're only viable within large dense cities. Which also makes sense, because that's the only place where you can justify tunneling under everything. 
So if you're comparing buses vs subways, it's a much different comparison to buses vs light rail. 
FYI when I lived in San Jose, my employer paid for light rail as their HQ was right next to a station. They were on the north side of San Jose, and I lived on the south side of San Jose, also right near a station. Driving in rush hour traffic (incl. carpool lanes) would take an hour. The light rail would take an hour [regardless of traffic] due to the many stops. A bus would probably have taken longer as there might be route changes in there, but then if my house and my employer weren't *right* on the route, I'd have to use a bus to get to/from the terminal station anyway. 
The main advantage of light rail to car was that you could sit there and do email, check your phone, read a book, etc, rather than concentrating on driving. But if I had to pay a couple hundred dollars a month for a monthly pass (which is what I believe they were at the time), I probably wouldn't have done it. 

Grade separated express bus ways can be an option.  The folks in traffic would see buses flying by unimpeded, delivering patrons to wherever, perhaps a subway stop.  When the bus gets to a congested area it can use normal streets for a few blocks.  

MARTA was built here in the 1980s.  The "good news" is that the areas surrounding many of the stations are exploding with development.  It's a big selling point.  The bad news is there are no plans for extensions of hard rail, everything planned is light rail or bus ways or street cars.  I'm not a far of street cars.  Buses are better, more flexible.

The state spent a huge sum on express lanes for I-75 outside the loop, toll lanes.  They are going to spend $1.8 billion more on express lanes, and some bus ways, for the north freeway (which is a disaster).  Those monies can't be spent on rail anything by law.

We were headed to Costco today to get gas and the wife stopped of the the tile shop, and then I noticed they had Peachtree closed for some reason, and I had to divert to another Costco out near Marietta, the traffic at 1 PM was ridiculous.  I would never live here if I had to face that daily.  I-285 is 6 lanes each way and was packed, but at least it was moving.

Wow--I'm surprised at how down on rail you guys are.

My metropolitan area would be a disaster without rail. Buses are great for short trips, but bad for long ones. Rail is an extremely efficient way to connect high density areas that are further apart than buses can reasonably serve. As much money as we pour into infrastructure for cars, a little more investment in rail would be a great idea, particularly as we continue to urbanize as a country.

Having spent a lot of time in SJ/SF, I agree that it's quite useful there. SF is so expensive that it's nearly impossible to live in the city unless you're rich, and it's so space-constrained that it's almost impossible to justify a car day-to-day. And because it's so centralized, having BART to get right into the heart of downtown is nice--particularly because BART goes to SFO. For people coming into the city that know they won't need a rental car, BART gets them from the airport to downtown in ~25 minutes IIRC. 
Light rail in San Jose and on the Peninsula? Much less useful. Outside of SF, there's a lot more sprawl, where it's unlikely that you not only live close enough to a light rail station and your destination is close enough to a light rail station to justify the additional time and cost of taking light rail. For travelers, if you're there on business, there's no connection from SJC to light rail, so if you wanted to get to a light rail station you'd need to Uber/taxi to the light rail. At which point you might as well either rent a car or just Uber from destination to destination. 

HSR is only viable within certain distances--the idea of a nationwide network is crazy. But an HSR in California connecting the population centers, or connecting the major population centers in the midwest and on the eastern seaboard (and a handful of other places) makes a ton of sense. The bureaucratic issues (and associated costs) with building one are the biggest problems. I've done my share of traveling to and from SoCal from the Bay Area. I live really close to SFO and it's still a five hour experience to get from my house to downtown LA. If I could make that on HSR in even the same amount of time (and it would be substantially less), in a much more comfortable, better serviced environment (train travel is way more amenable to working than plane travel), I would happily pay what I pay for air travel.

Question: what brings you to downtown LA, and do you think your experience is typical? If you live close to SFO, it should be no more than a 2 1/2 hour trip to get you to LAX. It's the getting from LAX to downtown LA that's the hard part. 
I will highlight that if you're trying to get to downtown LA from LAX, it's terrible. Unlike SFO to downtown SF, there is NO mass transit connecting the two. I think that would be an excellent use of mass transit, but right now you have to take a bus from LAX to the Green Line, tranfer to the Blue Line, and then transfer to the Red Line if you want to get to Union Station. That would be a pain in the ass, and take at least an hour plus. 
But I can't imagine why anyone needs to get to downtown LA?
Part of it for me is living in Orange County and absolutely hating to drive into downtown LA, but outside of sporting events, concerts, etc, I can't imagine many reasons to go to downtown LA. And HSR would take you to Union Station. LA also has so much sprawl that you'd find yourself having difficult public transit here anyway, or taking Ubers/taxis everywhere.
Generally business travelers are going to need the freedom of a rental car and can shoulder or expense the cost, and probably aren't going straight to downtown LA anyway because business in LA is all over the place. Which means that getting to LAX (or Burbank, or Ontario, or Long Beach, or John Wayne--we do have 5 major airports here) is probably just as functional as getting to Union Station downtown. If it's about vacationers, I highly doubt that enough vacationers are going to be riding HSR to create enough ridership to make any sense. 

Atlanta's growth has been pretty steady since the 1960s.  I don't think there was some point where it really took off after that point.

1900	419,375		—
1910	522,442		24.6%
1920	622,283		19.1%
1930	715,391		15.0%
1940	820,579		14.7%
1950	997,666		21.6%
1960	1,312,474		31.6%
1970	1,763,626		34.4%
1980	2,233,324		26.6%
1990	2,959,950		32.5%
2000	4,112,198		38.9%
2010	5,268,860		28.1%
Est. 2017	5,884,736		11.7%

The area has no geographic barriers, so folks can move further and further out.