Thread by @doctorow

Last summer, I wrote a six-part series on the history, future and present-day of queue- and crowd-management strategies at Disney themeparks, summarizing my endless reading, rumination, and direct experience on the question.

doctorow.medium.com/disneyland-at-… 1/

If you'd like an unrolled version of this thread to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:

pluralistic.net/2021/11/23/jum… 2/

Who gets to do what and when at a themepark may sound like a trivial question, but it's a perfect little microcosm for the distributional problems that are at the heart of all political economy - questions that the pandemic's shortages and shocks threw into stark relief. 3/

It's a question that's animated my work since my first book, Down and Out in the Magic Kingdom, which is literally about how people lay claim to priority over scarce systems, using methods that aren't driven by spot markets or other faith-tenets of neoclassical economics: 4/

craphound.com/down/

That's still in my work, tangled with psychology, economics, material culture, tech, and distributional ethics, which come together in Walkaway, about fully automated luxury Communist resorts in the climate emergency's ruins:

craphound.com/category/walka… 5/

I don't flatter myself that my six-part series was the final word on the subject of how aggregate demand and distribution can be managed through a mix of computers, networks, psychology and showmanship, but I *do* think I did a pretty good job. 6/

And then I spent 103 minutes watching @Defunctland's latest video, "Disney's FastPass: A Complicated History," and reeled. I slept on it, got up this morning, and I'm still reeling. 7/

It not often that I recommend that you spend nearly two hours watching a video, but that is my firm recommendation here.

8/

As the title suggests, the video depicts the evolution of different crowd-management strategies in the Disney parks, starting with the early ticket-books that bundled together "E-ticket rides" (like the Jungle Boats) with A-ticket rides (like the horse-drawn cart up Main St). 9/

It describes the internal executive struggles within Disney to change this system to some kind of reservations- or virtual-queue based system. 10/

These turned on philosophical questions about equitable access, operational questions of administratability, economic questions of profit maximization, and, of course, the technical question of what was actually *possible*. 11/

The turning point came with the advent of process-engineering simulations, which bolstered the case for Fastpasses (a reservation system that evolved into a kind of hybrid reservation/virtual queue system) by putting it on an empirical, rather than philosophical, footing. 12/

But "you treasure what you measure" and "every measurement becomes a target" and so Fastpasses became a symbol that different stakeholders in the Disney parks projected their own aspirations and fears onto. 13/

Regular visitors - annual passholders, mostly - learned how to "optimize" their Fastpass usage. 14/

Front-line operations people learned how to adapt them (for example, when a ride broke down, operators would dispense and discard several hours' worth of Fastpasses to balance out the promised reservations with the ride's capacity). Marketing sold Fastpasses. 15/

Infrequent guests who didn't understand Fastpasses fumed at the people who "cut in line" ahead of them. Imagineering designed queues to accommodate Fastpasses.

Each of these interventions were grounded in ideology, sometimes explicit and sometimes unspoken. 16/

For example: "Everyone who comes to a Disney park should have a chance to ride the best rides." Or: "People who put in the time to read guidebooks and online forums deserve to have an advantage when they come to a Disney park." 17/

Or: "People who save for a once-in-a-lifetime trip should experience the most impressive rides." Or: "People who are loyal customers and come all the time deserve the advantages that come with experience." 18/

It doesn't take much imagination to map these ideological positions onto higher-stakes distributional questions coming up today, like "People who choose not to get vaccinated should go to the back of the line at the ICU." 19/

But because we're talking about six minutes on Space Mountain and not a week on a ventilator, it's possible to examine multiple sides of the argument without quite so much emotional freight. 20/

New technology opened up new possibilities for queue- and crowd-management. At the same time, the company was discovering that a sizable portion of its audience was price-insensitive. 21/

When they tried to reduce the number of Annual Passholders by raising the price, they discovered that most of those APs hung in there. Raising the price again, and again, and again, had little effect on passholder numbers. 22/

This bolstered the case for resolving shortages by raising prices, and Disney started tinkering with selling priority access to rides, with strategies ranging from premiums for people who stayed in a park hotel to a per-visitor upcharge to unlock various additional Fastpass… 23/

…capabilities. Some of these had little effect, and some had very big ones, like the operational challenge posed by thousands of people crisscrossing the parks to collect Fastpasses from ride entrances like the faithful performing the Stations of the Cross. 24/

All of these currents came together with the advent of Magicband, an RFID bracelet tied into a complex back-end that was supposed to facilitate reserved rides, make it easier to buy things, act as your park tickets, and even your credit card. 25/

This project ballooned to a $1b tech boondoggle that struggles to this day (my own experiences with it have been farcical, even Kafkaesque). 26/

It also spawned a whole cottage industry of vloggers, influencers and forum flamers who offered ways to eke out minor advantages or circumvent the multi-hour waits for phone support when things went horribly, inexplicably wrong. 27/

A couple months ago, Disney announced #GeniePlus, a new initiative, based around a digital personal assistant that would recommend itineraries based on a combination of guest preferences, historical trends, and current conditions. 28/

Critically, this system also includes a way to buy Fastpasses - even as it does away with the ability to reserve Fastpasses months in advance. 29/

Describing this, Defunctland advances the argument that this system has the potential to balance out the different ideological positions embodied in different crowd-management tactics through the years. 30/

By limiting virtual queues and reservations to same-day bookings, the system ensures that people who are unfamiliar with the system will get a crack at the best rides. 31/

By keeping the upcharge on Fastpasses low, the system adds a manageable premium to the cost of a once-in-a-lifetime family visit. 32/

By limiting how many Fastpasses you can buy - you can't get a new Fastpass for a ride until you've used the one you're holding, and then it is subject to availability - the system balances ride capacity between frequent and infrequent riders. 33/

Anyone who spends time on Disney park forums will be familiar with these questions of distributional fairness - they are the source of endless arguments that turn on a mix of anecdote and logical inference, but they're notably shy on data. 34/

That's where the Defunctland video takes a hell of a turn. The team worked with computer scientists and data on real-world Disney park visitor wait times to build a powerful simulation of a hypothetical themepark: Shapeland (slogan: "Acute place to shape memories"). 35/

By running this simulation multiple times under different crowd-management strategies, the team was able to create an empirical account of the distributional effects and efficiency of each. The results are *fascinating*. 36/

For example, systems that let you buy your way into the line produce *some* efficiencies (less slack capacity on all rides, more rides per person), but depending on the size of the charge and its specific contours, the distributional and efficiency effects skew wildly. 37/

Some market-based "solutions" produce fewer rides per person, even as some rides sit empty, with ample surplus capacity. 38/

Whenever we're talking about empirical data on Disney wait-times, we're inevitably talking about @TouringPlans, @LenTesta's Disney park travel-agency/app/analytics firm. 39/

Testa started it as a college student who brought a stop-watch to Disney parks to compare posted wait times to actual wait times. 40/

Today, Touring Plans produces guides, realtime recommendations and planning information grounded in a decades-long, detailed data-set compiled by its customers' own use of its apps. It's the single best adjunct to a Disney visit you can have:

touringplans.com 41/

Unsurprisingly, Testa played a role in the creation of Shapeland and the analysis of its data; he says it "shows how FastPass disadvantaged the guests who had the audacity not to study for their vacation like it was a medical board certification." 42/

He also says that he's working with Defunctland to open-source the python code underpinning the Shapeland modeling tool, producing "the basic infrastructure to simulate things and help the next generation of theme park designers." 43/

Shapeland looks like a hell of a themepark to visit. Defunctland's got Shapeland merch up for pre-sale:

defunctland.myshopify.com 44/