It seems that when gauging reader interest about blog posts, the number of times a post is viewed is not actually that useful a statistic, because many of those views will be people who click through quickly from a search engine. A more useful metric of reader interest, I believe, is the number of comments readers write, and whether or not the discussion in the comments continues past the first day. Another useful metric is the number of times readers click links that are embedded in the post, because clicking a link in a post indicates (a) that the reader has made it past the first sentence, and (b) that the reader is interested enough in the post to see yet more material on a similar topic.
Recently, I compared comments and link-clicking (both on the actual blog posts, and on the maps posted on my Flickr account) for the post from last December about a fantasy subway rail network for San Francisco with the later post that discussed a fantasy bus rapid transit network for San Francisco. Some rather pronounced results emerged:
| Dream Subway Map | Dream BRT Map |
|
|---|---|---|
| Comments written within 0-24 hours* | 8 | 2 |
| Additional comments, within 24-48 hours* | 4 | 1 |
| Additional comments, within 48-72 hours* | 1 | 0 |
| Additional comments, after 72 hours (as of 2/12/08)* |
10 | 0 |
| Total comments (as of 2/12/08)* | 23 | 3 |
| Image clicks on Flickr (as of 2/12/08)** |
1,314 | 134 |
| Comments written on the Flickr map posts (as of 2/12/08)* |
5 | 0 |
| People who called the map a favorite on Flickr (as of 2/12/08) |
3 | 0 |
* Note: These numbers only reflect comments written by visitors; I did not count my own comments.
** Note: These numbers combine views from both the citywide map and the more detailed downtown map.
In every category listed above, the subway post was more popular than the BRT post. Okay, so this is not a statistically rigorous discussion, but the numbers are consistent with the generally acknowledged truth that most people find trains to be “cooler” and more impressive than buses. I agree that bus rapid transit will probably never measure up in this respect, but at some point we have to set aside the aesthetics to realistically weigh service improvements and cost-effectiveness. There are good reasons why San Francisco, in particular, should be more interested in bus rapid transit:
- Given the fact that Muni’s latest attempt at expanding the Metro via the T-Third can hardly be deemed to be an unqualified success, we should be eager to embrace alternative methods to improve and speed up service, but on a quicker time scale and with fewer engineering hurdles.
- In terms of population density, San Francisco occupies a slightly difficult middle area. Despite being the second densest city in the United States, after New York, density is not high enough to justify the cost of subway tunnels in every which direction. And yet, density is high enough to ensure that many bus lines are so well used that a standard local bus running in mixed flow with automobiles does not provide the speed, quality, or efficiency of service that is necessary to support the high ridership.
Both these factors point to an ideal plan in which we mix modes: building rail in high usage transit corridors, where there is heavy demand sustained over a substantial distance, and building BRT to fill in the gaps — or as an intermediate step for corridors like Geary, which deserve rail in the long run, but which need improvements as quickly as they can be delivered. Despite the fact that there is no shortage of successful implementations elsewhere that demonstrate how BRT can be done well, local Bay Area experiments with BRT (e.g. AC Transit’s 1R and 72R rapid buses, and VTA’s 522 rapid along El Camino Real) still have buses operating in mixed flow and getting stuck in traffic. And so, for those Bay Area residents who have bothered to learn more about BRT, the real-life local examples they have been provided so far are not really a fair representation.
Ultimately, the test will come when we actually build the projects currently being planned on Geary and Van Ness in San Francisco, and on Telegraph Avenue and East 14th Street in Oakland, Berkeley, and San Leandro. If Muni and AC Transit deliver carefully designed BRT projects with improvements that are immediately palpable to riders, we will go a long way to convincing people that although they are not trains, buses can deliver faster, more comfortable service than people in the Bay Area have been accustomed to. Perhaps then this often under-appreciated transit mode will be given its due.
Transbay Blog 
Let me suggest another possible explanation (although I certainly wouldn’t disagree that “cool” is a factor): a lack of understanding about what BRT really is, even among people who are generally reasonably knowledgeable about transit. You allude to this when you talk about implementations to date here vs. elsewhere. Let me just add to that my perception that some folks might be less interested in discussions about BRT out of …
– suspicion that BRT is a front for an anti-transit conspiracy (The Overhead Wire just had something related to this)
– skepticism that successful international examples are relevant in a North American context
– skepticism that “full” BRT projects will be implemented, and not watered-down, slightly better buses (which is well-founded, although if local experience to date is any indication, this may have as much to do with pro-rail as with anti-transit ideologues)
– backlash against pro-BRT ideologues who think buses can serve as a substitute for rail in all cases (and who make dumb comments like “rubber-tired subway”)
The bottom line is that supporters of BRT still have a lot of winning over to do, and in the meantime, discussions of BRT are likely to be unpopular — either in terms of a lack of interest, or in the controversial sense.
Oh yeah, sound analysis of the applicability of BRT locally … the metaphor I always fall back on is that of the toolbox. Some corridors simply deserve more than buses, but less than rail.
Posted by Steve | 12 February 2008, 5:08 pmTwo comments:
1. Cost Effectiveness means cheap. You get what you pay for, including with BRT.
2. “We Don’t Have the Density” sounds like we’re in Eastern Contra Costa County. We all know that better transit gets better densities. New York didn’t just happen to have all that density when they built the subway.
We need to stop serving the present and start serving the future. If we keep putting together these BRT plans for the city without rapid transit, we’re going to have a deeper service cost hole over time. One driver for 100 people isn’t cheap. It would be nice to think that after these lines are built they will start focusing on Geary or other improvements to Van Ness/Lombard/Mission. But my fear is that when its done, people will say, “but we spent all that money” and nothing will happen for another 20 years. We’ll just keep having to Rescue Muni for greater service costs per rider.
Posted by The Overhead Wire | 12 February 2008, 5:49 pmPS: Eric I’m not skeptical of your good intentions or that BRT will be an improvement to what’s there. Just getting frustrated at our Nation’s general lack of will to think bigger
Posted by The Overhead Wire | 12 February 2008, 5:55 pm“We Don’t Have the Density” sounds like we’re in Eastern Contra Costa County. We all know that better transit gets better densities. New York didn’t just happen to have all that density when they built the subway.
In my opinion, Eastern Contra Costa County should be getting BRT as well. This post is certainly not asserting that suburban riders deserve high-subsidy rail transit, while urban riders deserve more buses. My views are the exact opposite: I would like to see rail expansion where it makes sense, like under Geary. But subways do not always make sense. My point is that we should be open-minded as to what mode will work where. BRT isn’t always the solution, when you need higher capacity — as Steve remarked, the “rubber tired subway” sort of comment is nonsense. On the other hand, would you really want to build a subway tunnel for a bus corridor with just 10K daily riders? Does a capital investment like that really make sense if you’re going to only run a 1-car train every 15 minutes?
PS: Eric I’m not skeptical of your good intentions or that BRT will be an improvement to what’s there. Just getting frustrated at our Nation’s general lack of will to think bigger
You know, I think your point makes sense on a national level. I am also frustrated with the current administration — who will ever forget Mary Peters’ now-classic quote about biking? The administration has not been receptive to the fact that Americans have been thinking more positively about transit in recent years, and in turn, that makes this the ideal time to really make it a priority to fund the construction of metro systems around the country. It seems as though the issue is especially pertinent to low-ridership cities around the country, where the construction of a new light rail line isn’t just an issue of congestion, but also one of civic pride, and the chance to begin a more transit-oriented future.
I also agree there is a worry with, for instance, building BRT on Geary, and then sitting back and being satisfied that in the long run, that will be enough. On the other hand, you know how long it takes to plan and build a new subway tunnel around here. At least a decade, and realistically, much longer. This isn’t to say we shouldn’t have more ambitious plans — ideally, I would like to see BRT and subway plans shaping up about simultaneously. But are you willing to wait decades before if we implement any sort of improvement on Geary?
Posted by Eric | 12 February 2008, 5:58 pmWell of course not. But we’re building these lines on Geary and Van Ness. Arguably two lines that need way more than 1 car trains every 15 minutes or you could argue way more than a 60 foot bus every 5 minutes. In my trip to Budapest, the two ring routes that run the caterpillar trams that had 5 body sections came every minute transporting over 10,000 people per hour comfortably. That would be nice on Geary.
But in reading your post again with your comments in mind, yes it makes sense to educate people about BRT on corridors that could use it where a heavier infrastructure investment is not warranted. But you know i don’t advocate building rail everywhere. I’m just skeptical on the main routes where there is a definite need for higher capacity service and worry that this initial BRT investment will lead to nothing more later.
Posted by The Overhead Wire | 12 February 2008, 6:21 pmYou know, it’s interesting: we’re all basically in agreement here, in terms of the fact that different corridors warrant different modes. The point of contention is that we’re operating under different presumptions. On the one hand, there are concerns that rail projects around the country are being systematically cheapened to the point they are not maximally optimal in the long-term.
On the other hand, there is also a trend, at least here in the Bay Area, that hugely expensive rail projects are pursued not because of a pressing transit need (see the Millbrae and San Jose BART extensions), but for other reasons. In comparison, projects like Geary BRT seem like a breath of fresh air — no pork, but finally, a project that is actually just about transit improvements.
Despite these different presumptions, in the end, I think we probably arrive at very similar visions.
Posted by Eric | 12 February 2008, 7:07 pmHey an interesting question, how much of our federal gas tax money never comes back because we don’t build freeways in the city of San Francisco? We have 750,000 people, many of whom buy gas and pay a tax to the state and feds. How much do you think we get back? How much is going to go into projects in the region, but not to the core?
Posted by The Overhead Wire | 12 February 2008, 7:29 pmAn interesting fact: Muni’s per-rider operating costs are higher for rail than for buses. So in San Francisco at least, it’s not just the capital cost of BRT that’s less. And as we all well know, Muni is in a long-term, structural revenue hole.
Of course cost-effective isn’t all there is to it … but let’s talk about benefits for riders. Whatever you may think about the rightness of our current funding situation, it’s not just that it is what it is–it’s that it will be for some time, clearly, no matter what we may say or do. And you know what? Geary and Van Ness BRT will be up and running 4-5 years before the Central Subway, even with their much later starts. I don’t know about you, but I don’t want to wait 20 years for someone’s ideal.
Now this argument that a busway once built is a busway forever, I get that. But we ripped out the rails in the ’50s, and we started putting them back in the ’60s. And at least on Van Ness and Geary, the key thing, it seems to me, is the right-of-way. Both in terms of immediate improvement of the rider experience, and in terms of securing the space.
Ideally, Geary should be a rail corridor–but I’m not so sure it should be Muni Metro. If the ultimate configuration is express service via BART and high-quality limited-stops service via Muni, I for one live with that.
Posted by Steve | 12 February 2008, 8:14 pm^^^ If you add up the 38, 38L, and parallel bus routes 1, 2, 3, 4, 5, and 31, and their peak hour express routes, we’ve got 110K+ daily riders, or roughly 1/3 of BART’s daily ridership. And then think of the new riders we’d gain by having such a quick link between downtown and the Richmond District. Geary seems to me the epitome of a corridor that should receive BART service — though I’d guess it probably wouldn’t be built all the way out to Point Lobos, as I had drawn in on that dream subway map. That was a bit of a “Subway to the Sea”-style indulging.
Posted by Eric | 12 February 2008, 8:31 pmInteresting conversation. I’ve always found this density map (obviously missing some data in SOMA and other areas) very interesting:
http://www.uncanny.net/~wetzel/sfdensity.htm
According to this, density directly along Geary is actually quite low from Van Ness to Arguello, where it increases and stays pretty steady through 28thish or so. To me, this is perfect – we have high density out in the Avenues where development opportunities are pretty low, and incredible infill opportunities for a couple miles in the middle of the City where upzoning should be a bit more of a viable option.
I’m with Steve though – getting the right of way NOW is the key.
Posted by Chris | 12 February 2008, 9:00 pm1. echoing above Geary already has more riders than several cities’ entire rail operations. No matter what gets built west of Masonic, east of Van Ness cries out for a tunnel.
2. Why you think Muni will be more frugal building BRT? I expect BRT to be just as bungled as rail w/ all the same delays, job demands from untrained local youth, rebuilding the sewer and water first etc. The only advantage is no water crossings. In fact a Tunnel Boring Maschine project might be less disruptive
3 I live on AC’s 1R, I’m not impressed. AND ridership has NOT blossomed. I am sometimes nearly alone @ 8PM on a weekday. coming north from “downtown” Oakland An artic is a lot of bus for four people.
4. Yes I see BRT as the latest stingy response from a governmen that is too stupid to get the need to build real transit. 5. Yes BRT could be a decent short term service until the tunnels get built, but that won’t happen. If we settle for mediocre BRT on Geary that’s all we will get.
6 Van Ness might be a better candidate if only because surface rail could have buses from GGT sharing a car free ROW.
Potrero is wide enough for the same w/ SamTrans.
Posted by david vartanoff | 12 February 2008, 9:47 pmI wonder where Wetzel got that map … most of it’s right, but Duboce Triangle at 55K/sq. mi.? What? In any case, though–your point about Geary is largely correct. I’m not sure I’d call Richmond densities “high,” necessarily, but infill is coming to the Laguna-Masonic segment of the corridor (see the J-Town Better Neighborhoods effort). And regardless of land use, the Richmond has clearly generated some ridership. If you look at the Mission corridor as a whole, including BART, it’s a bit higher — but not that much, which is why I’m still holding out hope for BART below Geary one of these days. I probably won’t live to see it, but I am sort of surprised that so many railfans have just assumed that Muni is the end game. It’s one thing to talk about what’s practical in the near future; quite another to consider the real long-term potential.
Year 2000 densities by census tract (darker is denser):
You can really see there why BART works, though like in East Oakland, it would’ve worked better a bit farther east (of course, it would have cost a lot more).
Posted by Steve | 12 February 2008, 9:56 pmNice density map, Steve.
I have no idea where the Duboce Triangle density comes from, as it seems quite clearly incorrect.
I think that the “end game” of Muni Metro for Geary might be an artifact of the old Central Subway alignment, which alternated between stacked and side-by-side tunnels exactly to allow space for a Geary tunnel in the same stretch. Under the current “preferred alternative”, though, a transfer from the T-Third to a Geary subway would not be convenient. And Rescue Muni’s Post Street tunnel, with a terminus at Montgomery, is even further from the Central Subway’s Union Square/Market Street platform.
Since the Metro transfer at Union Square doesn’t really make sense, I’d rather see the BART service, which would make even more sense with some serious infill on that middle section of Geary between Japantown and the inner Richmond.
Posted by Eric | 12 February 2008, 10:13 pmA little better numbers this time, Eric? ;-)
Posted by Steve | 12 February 2008, 10:28 pmBetter numbers?
EDIT: Doh, I must be sleepy. You must mean comment numbers.
Yes, they are higher this time. But notice how we’ve moved onto discussing rail! ;-)
Posted by Eric | 12 February 2008, 10:34 pmWhile BRT is in principle appealing as a low-cost form of transit, we need to make sure that it’s costs really are lower than rail.
Muni’s projected costs for BRT on Van Ness are $20 million/mile, pretty close to what Portland pays for streetcar tracks.
I’m all for BRT, especially if it really is cheap–put in signal prioritization, spraypaint BUS ONLY on the centerl lane, put in some passenger islands–and that’s it. No massive overhaul of the whole street!
Posted by Nick/295bus | 12 February 2008, 11:27 pmActually the Van Ness budget is quite a bit higher, but then the F-Market cost quite a bit more than the Portland Streetcar when it was built years ago. If you were putting a streetcar on Van Ness, all else being equal, the costs for the streetcar would be significantly higher. You could save money by leaving vehicles in traffic like Portland has done, but I’m not sure what that would get you on Van Ness other than a smoother ride.
Posted by Steve | 12 February 2008, 11:31 pmVan Ness BRT is projected to be about $45-50 million/mile. Telegraph/East 14th BRT is closer to $20 million/mile, but they don’t have quite the same plans for streetscape improvements that are in store for Van Ness. Incidentally, landscaping could be used to clearly separate auto lanes from bus lanes.
Posted by Eric | 12 February 2008, 11:38 pmLet’s try to be realistic about the limitations of unenforced transit lanes, though: for whatever reason, we haven’t been able to really enforce them. I’m all for signal priority and queue jumps, but where there are truly substantial congestion issues, like on Van Ness, it seems to me that we have to spend more to get much of anything.
This is the key issue, IMO: right-of-way. The kind of material beneath the box seems somehow secondary to practical concerns such as speed and reliability.
As for costs, it’s important to be specific in the local context. S.F. BRT, estimated, as Eric has said: $45M/mi. The T-Third: $125M/mi. Of course those are up-front capital costs, and if we’re taking the long view we should also consider operating and life cycle costs. It’s interesting that folks rarely mention the latter; those Bredas of Muni’s we’re going to have around for awhile. But as I mentioned before, in San Francisco you increase your operating costs when you go to rail. And we don’t always have the luxury of the long view, however much we might wish we might.
Posted by Steve | 13 February 2008, 12:04 amOh, I should’ve been clearer about that. When I mentioned landscaping to separate bus lanes, I was assuming the right of way, with center lanes reserved for buses. Considering VN’s congestion and the volume of buses (GGT buses, too), I don’t see how anything less than the full treatment will really do the job. BRT has the chance to turn VN into a special place, and it will be disheartening if the powers that be try to settle for less.
Given that the 47R and 49R will still run in minimally-improved conditions south of VN (assuming that some of those plans brewing for Mission Street have been implemented by the time BRT gets going), it will be very interesting to see how far dedicated lanes on only VN go towards reducing operating costs on those lines.
Posted by Eric | 13 February 2008, 12:28 amCount me in as a BRT skeptic too – considering that just about every implementation on a corridor like yours has ended up as a joke (the only things approaching the promises of BRT are in ‘new’ corridors or in old rail corridors).
A combination of surface light rail and subterranean light rail seems obvious for San Francisco. To the rest of the country, even fairly anti-transit regions, you seem plenty dense enough to justify such an investment – and if you’re going for the “full treatment” vis-a-vis BRT, it’s really not much more expensive to do it right and put down the rails, even with tunnels involved.
Posted by M1EK | 13 February 2008, 8:58 amWhat’s the scoop on vehicles to be used on Van Ness and Geary? Has that ever been decided?
Posted by Chris | 13 February 2008, 10:06 amI wonder where Wetzel got that map … most of it’s right, but Duboce Triangle at 55K/sq. mi.? What? In any case, though–your point about Geary is largely correct. I’m not sure I’d call Richmond densities “high,” necessarily, but infill is coming to the Laguna-Masonic segment of the corridor (see the J-Town Better Neighborhoods effort).
Yeah, Duboce Triangle is definitely wrong. I guess for Richmond densities I’m thinking in regional or national terms. Maybe eventually some of the merchants on Outer Geary will croak…
Speaking of the Laguna/Masonic stretch – wouldn’t it be nice to doze the Mervyns/Best Buy/Office Depot disaster and rebuild some nice mixed use in that area? That place has always struck me as one of the worst retail complexes anywhere. It’s not easy to find your way around by foot OR car.
Posted by Chris | 13 February 2008, 10:40 amSpeaking of the Laguna/Masonic stretch – wouldn’t it be nice to doze the Mervyns/Best Buy/Office Depot disaster and rebuild some nice mixed use in that area? That place has always struck me as one of the worst retail complexes anywhere. It’s not easy to find your way around by foot OR car.
I’ve been hoping for years that this would happen. The whole area is very haphazardly designed, and not at all conducive to walking.
Careful thought should be given to improving the area around all the future BRT stations, though some are better than others. The Geary expressway makes it easy to have dedicated ROW, but it leads to a few counterintuitive street configurations.
Posted by Eric | 13 February 2008, 11:28 amSeems to me painting the center lanes and planting enforcement cameras is a low intensity move well within “police powers” of a county and NOT requiring an EIR. Same preempts. Thus VN should be possible soon and on the cheap. The increased throughput should be instructive.
Posted by david vartanoff | 13 February 2008, 12:18 pm‘Seems to me painting the center lanes and planting enforcement cameras is a low intensity move well within “police powers” of a county and NOT requiring an EIR.’
You’d think.
In London, this is how they got their efficient, fast buses running on uncongested, exclusive lanes; they just started repainting the streets to create bus lanes, and fining people who blocked the bus lanes.
(Then, they found there wasn’t enough room for bus lanes in the very narrow streets of the old downtown area, so they introduced the Congestion Charge to chase the cars out of the area.)
If *that’s* what you mean by BRT, by all means, go for it. lightrailnow.org simply calls that “Better Bus” or “Quality Bus”.
Unfortunately, BRT seems to mean different things to everyone who uses the term, which is probably why it’s gotten a really bad rap. If by BRT you mean massively expensive busways like Pittsburgh has — well, if you’re spending that much money, put in the rails already, the payback will be well worth it.
Frankly, I’d stay away from the hyped “BRT” term. Make posts about “real express buses” and “exclusive bus lanes” and you’ll probably get *more* attention than if you talk about “BRT”.
Posted by Nathanael Nerode | 23 February 2008, 8:44 am“in San Francisco you increase your operating costs when you go to rail.”
This is bizarre in the extreme, since it’s in contradiction to the experience of all other cities which have entrainable railcars. Is it due to bad deals with the unions or what?
And I noticed this:
“a lack of understanding about what BRT really is,”
Well, since no two people seem to agree on what it is, and since its definition from federal and local authorities varies every time they make a new announcement, this isn’t really surprising.
Posted by Nathanael Nerode | 23 February 2008, 8:47 am> This is bizarre in the extreme,
It has mostly to do with the lesser capacity of trains here compared to other places.
San Francisco has many miles of striped transit lanes, and we’re starting to get cameras (on the vehicles). Hopefully they will make some difference.
Posted by Steve | 24 February 2008, 10:39 amHopefully they will make some difference.
Only if we can catch more than one double parker!
Posted by Eric | 24 February 2008, 10:52 amEric, thanks for sparking this great debate.
The condensed issue: Muni vehicles need to stop less (reduce stops / reduce dwell) and go faster with an absolute increase in roadway priority.
Whether it’s heavy/light rail, bus, BRT or small capacity vans, the above statement is true.
Rail forces roadway priority owing to the physics of massive objects. If Muni can force roadway priority via other routes for non-railway vehicles — proper roadway planning, signal preemption, politics, BRT, whatever — than so be it.
As long as Muni stops less and goes faster with higher roadway priority I don’t give a hoot whether it’s an ugly high-platform bus, a piece of Breda junk, or a sleek hybrid low-platform bus. Nor will other City residents care as long as Muni stops less and goes faster.
So, then the real question is, how can we accomplish the above goal (stopping less, going faster) the cheapest? That answer is probably not rail in the short term.
Posted by kfarr | 29 March 2008, 12:52 pm