Went out last night to Town Hall Seattle (the always-interesting, worthy and thank-god-we-have-it-here-in-Seattle, — thank you founder David Brewster!) to hear Darrin Nordahl and Jarrett Walker: Perspectives on Public Transit. Very stimulating discussion. To boil it down crudely, my take-away in Nordahl v. Walker (and grossly-simplified so I can bring out my point) is sexed-up transit vehicles versus frequency — frequency yields "headways" — how often a vehicle (whether it is car, bus, train or spaceship) passes a stop and thus the amount of time you are waiting to board. 

While I agree completely with Darrin Nordahl's call (and it was a very well-preseented call — I look at technique since I do public presentations and so always interested in how to do it better)  for better customer service and nicer vehicles (clean, safe, attractive -- maybe comfortable?) I believe that Jarret Walker's emphasis on frequency carried the evening. To me it's no issue. I don't use transit because I don't like to plan; my trips are somewhat spontaneous and I like life that way. The buses in Seattle are physically acceptable; they simply don't run often enough.
(Personal example slightly off the point: Last night I went from my house (Maple Leaf) to quick dinner with friends (Montlake) to Townhall (First Hill) to Queen Anne (dropped off friend and a drink) to home again. Elapsed travel time door-to-door (which is the test) on those 4 legs? Maybe 45 minutes. I can't imagine that I could have done it in less that 3 hours because of the numbers of stops, the need for transfers and time of day. Yes, I should go to Metro link and add up the actual legs in real time at the times I used and of course it's not completely fair as without a car I simply wouldn't have made such a routing — and would life have been very much diminished? Probably not really. Just different. Nevertheless, we middle-class North Americans are attuned to going wherever and whenever and that such psychological state has to be considered as a genuine political constraint.)
 
Do bear in mind that it is precisely those off-peaks are especially efficient for cars (no traffic) and especially ineffecient for transit. The last leg of my trip at 10:30 PM from Queen Anne to Cloud City Coffee (nearby) would take 60 minutes by bus while in the empty streets it was less than 15. Driving safely. Huge difference.
My basic problem with American transit is that I don't want to plan nor wait. Ideally, the operational test for good transit, so far as I can see, is that if I have to use a bus schedule, I won't use transit. 

So if frequency/headways are the issue — why doesn't every transit system have high frequency/low headways? Money. Pure and simple. Just not enough money. And here's the out-of-control downward spin. "Enough money" for public transit is purely a political issue. Period. Triple stop. Poorly-funded transit is purely a function of lack of political will.

So why so little political will? Because people don't use transit, so politicians don't feel on-going pulse-pounding pressure to spend more money on transit.

Part of the conventional wisdom about transit which prevents spending real money on quick headways is that "There's no market. People aren't using that route. Why should we be running empty buses?"

My quick answer to that old saw is that just as with roads, with transit, supply induces demand. If you offer a rich 'connection pipe" (to garble Walker's terminolgy) then people will use it and people will get used to the service and presto you have political support. People support things (politically) that they use.

So here is my advice/suggestion to managers of transit systems: experiment. Goose-up the frequency of headways and see what happens. Choose one route (and choose it astutely) and run buses at a level which are not now justified by current traffic. Then make sure people know about it, that they can ride without using a schedule. Is there political risk? Of course. You have to choose your route wisely. You have to understand how long it will take for potential users to realize that they don't have to plan to use transit. You have to budget for running that route at a deep deficit. Your board will look at you and your job is on the line. (Obviously easy for me to say.) But if you are serious about getting people to use your system, people using it as a matter of choice and not because they are poor and without a car, then it's pedal to the metal to show (not cajole through marketing) that the key to system use is frequency.

Show me all the greatest marketing for a transit system etc etc and if you don't have great headways, I won't use it. 
 
Btw, I think that I once spoke to Walker about this political experiment approach and while we agree on frequency, he thought my suggestion wrong, or impossible or something like that.