Sunday, February 28, 2021 in portland, Oregon, U.S.A.
Earlier this week marked the formal point at which the United States passed 500,000 deaths due to COVID-19.
The last time I flew on a plane was February 29, 2021, so I’m coming up on my one year never-flying-again anniversary.
I’m painfully aware of how long it’s been since the last episode (January 18, a mere 36 days, for those counting) and the unbearable growing pressure of not having written anything here yet. So, here we are. Let’s put some words down and share some of the things that caught my attention in the past few weeks.
As ever, if you feel like it and are able to, subscribing helps support my writing.
I wrote two tweetstorms [sic] on the state of delivering vaccinations in the United States, the first on January 30, listening 51 Things That Can All Be True [Twitter, Threadreader].
The prompt for that first thread was Cat Ferguson’s What went wrong with America’s $44 million vaccine data system, published by the MIT Technology Review, also published on January 30.
The short version of that thread for the time-constrained and differently executively-functioned is that, well, everything is quite complicated. First, you’ve got a procurement environment that in general sees vendor participation in previous government technology projects as both a requirement for participation and an indicator of success, and other purported indicators like having worked in a similar domain. I.e. you can be a vendor who has a) participated in a previous vaccine technnology project, b) it was a large technology project, c) of which project failed to deliver at its high level outcome, a combination of e.g. actually vaccinating people, being usable by the people who have to use it, and delivering value on some sort of ideal of time and money, and that actually means you’re qualified, and that if you are a vendor who has not done all of those things you are disqualified and thus ineligible to bid.
And that’s just the first part in the procurement environment! That’s without the part about competency and participation in the executive sponsorship, that’s ignoring the part about pockets of government having the skills, capability and experience to deliver parts of the system necessary but they’re spared too thin, and then most of all, on a level I think that many do not grasp, the U.S. system is by design complicated in terms of the Federal -> State -> County/City/Local cascade of responsibility, degrees of freedom and what they have responsibility to deliver. Combined with everyone, everywhere, thinking they’re special snowflakes who have unique requirements when for the most part they don’t, and not realizing that those requirement differences are policy differences which can actually be changed.
Anyway, go read the thread, it’s not very long, and apparently Many People Who Work Or Have Worked In Government see it as a reasonable and accurate representation of the complexity of the situation by pointing out lots of different things that are True and not, say, Aligned For Success, Whether Easily Or Not.
Here are some of the resources and further reading I included at the end of that thread [Twitter, Threadreader]:
What went wrong with America’s $44 million vaccine data system? (mentioned earlier)
A Civic Technologist’s Practice Guide by Cyd Harrell, which at the very least explains the division and delegation of responsibility in the layers of U.S. government down to the local level
‘When can I schedule a COVID-19 vaccine appointment?’ Why the government can’t answer this simple question
As a bonus, the forever useful On Design Thinking
The second was a 40-tweet thread (sorry. No, really, I’m sorry. But I’m probably just about as annoyed at Twitter for enabling this kind of behavior as I am myself, so perhaps that is some consolation) about the New York Times’ reporting on a (deep breath) better/more useful vaccine website for New Yorkers that was built for $50 (Twitter, Threadreader).
I’ve actually been dreading writing this thread up again because it’s been very difficult to resist my temptation to do it properly, which means writing an even longer, expanded, contextualized and simultaneously more nuanced (and stark) version when, come on, the original version was 40 tweets long in the first place. (He says, staring at the unfinished too-long-for-an-op/ed version in Google Docs).
Anyway, here are some of the beats, but I encourage you to read the thread if you care about this sort of thing, or if you’re interested in why things like “getting vaccinated” have been, and are, so difficult and frustrating for people in America right now.
But the main thing that prompted my ire was the NYT’s subtle insinuation along the lines of “Ugh, see how easy it is to improve these things? It only costs $50 and a very smart person did it in their spare time. How terrible is government!” and yes, I’m aware there’s a very strong possibility I’m projecting here. And yet.
And yet: (deep breath, recapping the thread) Ma, the developer, is a software engineer at Airbnb, where average total compensation is around $200k (says Glassdoor), and he tells the NYT it took around two weeks to build. So the claim that the website “cost” $50 to build is pretty specious.
I got shit for this online because people rightly pointed out that volunteer vaccine websites are helping people, as if I somehow resent the existence of them, or think they shouldn’t exist. But I don’t think they should exist, in an ideal world they shouldn’t exist at all. As an armchair, unqualified journalist, there are a couple stories here I would’ve thought were newsworthy:
a) here is a website, which is helpful now, which you can use to get vaccinated in a pandemic; and
b) why are volunteer websites filling a horrific gap, and what might need to happen to ensure critical public health doesn’t rely on volunteers
you’ll note that nowhere in a) or b) is “look at how easy it was to build this volunteer website or how little money it cost”. So that’s where I saw the failing of the NYT, in lazily and harmfully perpetuating a high-level stereotype that government/tech is horrible (which like many stereotypes is rooted in truth, because it is a fact that much government/tech is nowhere near as good as we deserve), and that improving government/tech is something that can easily be done as an amateur in your spare time.
Because the latter is true only on a surface level. I used to not like it when people cited Conway’s law because it feels too abstract and hard for people to understand on a relatable level (“organizations design systems which mirror their own communication structure”), but I feel there’s a good way to use Conway’s law as an opportunity to educate people on how profoundly fucked-up certain aspects of government in the U.S. are and let me say for the avoidance of doubt that the vast majority of people in government are doing their goddamn best in the environment they’re in, and this is not an indictment of them.
In other words: yes, the experience of last-mile vaccine delivery websites is horrible and it could be much better in the low-hanging-fruit sense (copy could be much better, websites should have worked on mobile from the get-go), but those are the paint. What’s horrible, and what the websites reflect are aspects like the federated nature of government in the U.S.
What they illustrate are aspects like a system that in general is more concerned with bureaucratic design and administration of processes (which does need to happen! you are doing things for millions of people!) over outcome and, in a beating-the-drum-about-user-research sense, a disconnect of that outcome to when your policy is complicated, it has an affect on actual people, which of course we know in abstract, but, you know, there’s a more fundamental way of knowing.
I saw a tweet somewhere, and I’m afraid I don’t remember the provenance but it may have been from or via someone like Dana Chisnell that service design is… actually stupendously hard for an individual person to do? In that so much is out of their control? Service design in the sense of “I am involved in the design of a service that helps people get vaccinated” in the U.S. is so amorphous, so spider-tentacled in how it works in practice, from federated policy to operations to implementation that at some one-year-into-COVID-pandemic sense all I can do is laugh manically in response. Do you know the level of coordination that would require?!
Anyway, it is breakfast time. There is bacon and pancakes and I will leave it at that, because why not leave things at a despairing, sisyphean absurdity, and yet what can we do but try?
I am not doing well, how are you doing?