I had the privilege of getting an advance draft of Patrick McKenzie’s very long story founding and running VaccinateCA, an organization dedicated to providing Americans information on where and how to get vaccinated against Covid-19.

It is an amazing document. Despite its length, I am going to flat out say to Read the Whole Thing if you have the capability of doing that. It is full of things worth knowing and understanding. I benefited from reading it twice, once as a draft and again as a finished document.

Since the VaccinateCA post remains very long and difficult to navigate or draw reference from, this post is an attempt to pull out the most long-term-important passages and facts and give an outline of the events and insights that reflect the world outside VaccinateCA, to keep this bounded. I have a short bullet point section on the inside lessons as well, but to keep this bounded am focusing elsewhere.

A central theme is the author, Patrick McKenzie, being surprised and confused by how dysfunctional were government operations, and how deprioritized was the cause of having less citizens get sick and die. As you read, notice these repeated surprises and confuses, and notice that the correct model would not be so reliably surprised in the same directions.

Similarly, he notes that he believes some of his statements present the situation as so dysfunctional that they sound crazy. The statements do not sound crazy to me at all. They sound accurate, and like the type of thing one would expect. It seems important, if we want things to improve, to realize that such statements are accurate, normal and expected.

Again, seriously, read the original if you can, at least the parts about how government systems (largely didn’t) track the vaccine and ensure it was distributed to citizens.

The Logistical Problems With Distribution Were Unnecessary

For some reason we decided that our standardized cold chain, sufficient to keep the vaccines good for ten weeks, was not good enough, despite that obviously being good enough. So for no real physical reason, we didn’t use our existing infrastructure.

The Pfizer vaccine keeps for ten weeks when transported or stored unopened between two degrees Celsius and eight degrees Celsius (36°F and 46°F), which is the exact same range as the flu vaccine and overlaps that of milk.

Transportation was easily achievable via cold-chain logistics, abundantly commercially available in the US and already used pervasively in the pharmaceutical industry.

The product rollout for the vaccine should have been approximately as logistically straightforward as any product rollout in the United States. It was not, principally because it did not use the US’s formidable advantages in coordinating product rollouts.

The Hyperlocal Information Problem

The core problem that VaccinateCA was created to solve was that our vaccine distribution system:

  1. Took some vaccine doses.
  2. Distributed them to various places.
  3. Did not track how much was distributed to which places when.
  4. Did not track how much vaccine was available at which places.
  5. In practice, let each place decide what subset of the people legally eligible was eligible to get their particular vaccine shots. Some restricted by geography, some were more strict about age or condition, others less so.
  6. Individuals were left to call various places to see who had vaccine shots.
  7. Good news is calling pharmacists and asking ‘do you have the vaccine?’ works.
  8. In California this system managed, in the face of overwhelming demand, to administer only 27% of delivered vaccine shots into arms by mid-January.
  9. The remaining 73% were sitting there. In a freezer. Waiting.
  10. Had no plan to remedy the lack of information. Did not much care.

Patrick McKenzie posted a Tweet suggesting building a website and offering to pay server costs, Karl Yang took up the gauntlet and set up a discord server, and a group of ten worked through the night, forming the nucleus of the working group. Things went from there, the full post has lots of details.

The core plan was…

  1. Gather a bunch of people who can make phone calls.
  2. Call a bunch of places that might have the vaccine.
  3. Ask those places if they have the vaccine and who they would give the vaccine to.
  4. Post this information on a website, unless the provider requested that they didn’t.
  5. Keep calling to update the data, prioritizing whoever likely had the vaccine.
  6. That’s it. It worked. Lots of people used the website to find vaccine shots.

It was still a ton of work and this leaves out a ton of detail – see the full post.

The government was initially skeptical, but offered increasing levels of support once it was clear that VaccinateCA was getting the job done where others didn’t. At first this was informal. They got a Bat-Phone to California and volunteers started getting pointed in the project’s direction.

As opposed to the official California website My Turn, which had limited functionality and accuracy, and which failed to efficiently get shots into arms. The official system failed to understand, for example, that when it gave shots to the University of California at San Diego hospital system they then gave those shots to various different locations. The government posted where their records said their department’s share of the shipped doses should in theory be, as opposed to VaccinateCA which asked those on the ground where the doses were. Only one of these methods remotely matched reality. My Turn was, of course, deemed a huge success.

Or the official Federal attempt, VaccineFinder. This was supposed to aggregate data from the pharmacies. The problems included having a severe deficit of engineers, state politicians having zero interest in letting Biden take credit for anything they’d managed to put together, the pharmacies only getting half the doses to begin with, and the Federal Government having previously agreed to not write a huge percentage of the inventory information down or scrub it of information vital to things like knowing who had how many doses were where.

Oh, and this:

And thus many governors were not in on the plan. But it gets worse. Since the optics would be terrible if America appeared to serve some states much better than others on the official website that everyone would assume must show all the doses, no state doses, not even from states that would opt in, would be shown on it, at least not at the moment of maximum publicity. Got that? Great. (This theory is a theory, from a person with much more experience in DC than me.) 

Also as opposed to much better resourced public health teams and others at major tech companies. Those people had the skills and resources needed to handle our needs, and did not do so. Patrick explains that this was because government was unwilling to work with Big Tech, and made it abundantly clear especially after January 6 that Big Tech Will Pay, so everyone involved was told they must keep their heads down and not do things like go around asking people for medical information.

It is often said that such effects – the ‘if politicians treat an industry or class as unfavored and threaten to treat that group badly in various ways, including not letting them keep their profits or potentially going after them, they will largely cease the Doing of Things, including and especially exactly that you would like them to do’ – are not something worth worrying about. This is a good example of why to worry.

The Low Tech Way versus The High Tech Way

There exists a narrative that engineers love technological solutionism too much, believing complex societal problems could be solved if we just had software that worked. This critique is accurate sometimes. But it did not describe the dynamics this time.

Against my prior expectations, governments wanted magical IT systems that would paper over their bureaucratic divisions. We technologists were willing to ship low-tech approaches that actually worked.

The core unique insight VaccinateCA had was that America has access to a reliable technology for getting information from the healthcare system. It is called a telephone.

Recently I asked around about why voice assistants so often fail to respond appropriately to scarily common requests – for example if you say ‘cancel my 5pm timer’ while having a 5pm reminder or 5pm alarm instead, it won’t notice. A majority of failed requests could be solved by a modest engineering team looking at failed requests and hardcoding if-then statements, and this does not happen, with the explanation being in part that no one wants to be on that team rather than figure out how to use fancy scalable machine learning techniques.

Here we have the opposite. The tech people are trying to get a physical result and notice that they have a workable solution. The government despairs of or dislikes such solutions and wants magical IT integration instead, while the tech people make it all work in a matter of hours or days.

We have essentially zero fast-moving state capacity in such a spot. If the state tried to do things that way, it would require endless approvals and bids and specifications and other steps and take quite a long time, even in the best case scenario.

On Being Legally Forbidden To Administer Lifesaving Healthcare

This section is brutal. I am going to quote a large portion of it and the following two.

It is legal by default to save a life, unless it is through healthcare, in which case it is illegal by default to save a life, unless one has a license to practice healthcare, in which case it is legal by default to save a life, unless it is early 2021 and your intervention of choice is a Covid vaccine, in which case it is illegal by default to save a life and don’t you forget it.

If anything I think the situation is slightly worse than that in the general case. Saving a life even with a medical license is not so much legal by default as legal if you are engaging in standard pre-approved medical lifesaving practices using standard pre-approved medicines and devices, and illegal otherwise. But I digress.

This is about the prioritization system that determined who got access to lifesaving medicine first, since we decided to allocate via power. The elderly are at vastly more risk for Covid than everyone else, enough to overwhelm any and all other risk factors. Speed was of the essence. So what did we do?

Political subdivisions of the United States then conducted politics briskly, and made choices about who would benefit from lifesaving medical care under conditions of extreme scarcity and who would not (yet).

Sometimes these decisions were based on medical reality. Sometimes they were based on political expediency. Very rarely were they effectively administered.

California, not to mince words, prioritized the appearance of equity over saving lives, over and over and over again, as part of an explicitly documented strategy, at all levels of the government. You can read the sanitized version of the rationale, by putative medical ethics experts, in numerous official documents. The less sanitized version came out frequently in meetings.

This was the official strategy.

The unofficial strategy, the result the system actually obtained, was that early access to the vaccine was preferentially awarded based on proximity to power and to the professional-managerial class.

The California prioritization tiering list is abundantly available on the historical record, though it doesn’t record a hundredth of the horse-trading that was involved in writing and implementing it. (As a note to the reader, your state or county may have used different labels or used the same labels but had them mean different things, or changed the meaning of the labels on a weekly basis. Public communication regarding prioritization was a confused mess.)

Consider ‘essential workers’, a concept that predated the availability of the vaccines. If you weren’t on the Essential Critical Infrastructure Workers list, you were at risk of being locked indoors and suffering economic damage, to say nothing of being deprioritized for the vaccine, and so clearly your lobbyists were very bad at their jobs. You should have had better lobbyists. The essential workers list heavily informed the vaccination prioritization schedule. Lobbyists used it as procedural leverage to prioritize their clients for vaccines. The veterinary lobby was unusually candid, in writing, about how it achieved maximum priority (1A) for veterinarians due to them being ‘healthcare workers’.

Teachers’ unions worked tirelessly and landed teachers a 1B. They were ahead of 1C, which included (among others) non-elderly people for whom preexisting severe disability meant that ‘a covid-19 infection is likely to result in severe life-threatening illness or death’. The public rationale was that teachers were at elevated risk of exposure through their occupation. Schools were, of course, mostly closed at the time, and teachers were Zooming along with the rest of the professional-managerial class, but teachers’ unions have power and so 1B it was. Young, healthy teachers quarantining at home were offered the vaccine before people who doctors thought would probably die if they caught Covid.

Now repeat this exercise up and down the social structure and economy of the United States.

There were literal and metaphorical passwords to get priority access to the vaccine.

Healthcare providers were fired for administering doses that were destined to expire uselessly. The public health sector devoted substantial attention to the problem of vaccinating too many people during a pandemic. Administration of the formal spoils system became farcically complicated and frequently outcompeted administration of the vaccine as a goal.

The process of registering for the vaccine inherited the complexity of the negotiation over the prioritization, and so vulnerable people were asked to parse rules that routinely befuddled healthy professional software engineers and healthcare administrators – the state of New York subjected senior citizens to a ‘51 step online questionnaire that include[d] uploading multiple attachments’! 

That isn’t hyperbole! New York meant to do that! On purpose!

Lives were sacrificed by the thousands and tens of thousands for political reasons. Many more were lost because institutions failed to execute with the competence and vigor the United States is abundantly capable of.

And in the following section:

Why did California go head over heels for equity? Aside from the political valence of it and the point at which American society was less than a year after George Floyd’s death, it is classic bikeshedding. Most people in civil society cannot develop, manufacture, distribute, or administer a vaccine. Decrying systemic racism, on the other hand, is quite accessible. We exhaustively train the entire professional-managerial class in doing it. Accordingly, official discussions of strategy for the vaccination effort quickly bent toward systemic racism. Lacking any ability to contribute regarding one pressing problem, many individuals of good will focused on the other.

I believe this routinely was counterproductive for both goals. I feel it necessary to acknowledge that this sounds like the rantings of an unhinged mind and, my mind being hinged, to acknowledge that I know what it sounds like.

I do not think this sounds unhinged at all. Complexifying systems will by default favor power and the types of people who specialize in navigating complex systems. Allocating resources on the basis of power and patronage and expecting the resources to advance ‘equity’ is a category error. That is not how any of this usually works.

Residency restrictions were pervasively enforced at the county level and frequently finer-grained than that. A pop-up clinic, for example, might have been restricted to residents of a single zip code or small group of zip codes.

All people are equal in the eyes of the law in California, but some people are . . . let’s politely say ‘administratively disfavored’.

The theory was, and you could write down this part of it, disfavored potential patients might use social advantages like better access to information and transportation to present themselves for treatment at locations that had doses allocated for favored potential patients. This part of the theory was extremely well-founded. Many people were willing to drive the length and breadth of California for their dose and did so.

What many wanted to do, and this is the part that they couldn’t write down, is deny healthcare to disfavored patients. Since healthcare providers are public accommodations in the state of California, they are legally forbidden from discriminating on the basis of characteristics that some people wanted to discriminate on. So that was laundered through residency restrictions.

Unfortunately for this plan, another advantage disfavored patients have is ability to prove their residency to a bureaucratic process.

We were willing to spend more lives in total in the cause of reducing disparity in death rates along axes that are politically salient in California, by allocating resources in a manner we knew to be inefficient.

Did we execute competently on this plan? No.

The development of vaccines for Covid-19 was a world historical success. The rollout of the vaccines in the United States was plagued with logistical and communication issues.

When the residents of those neighborhoods dear to the state showed up to receive healthcare, they had to hope that the person providing it knew that the residency restriction was only supposed to apply if you were a disfavored patient. But that policy could not be written down, and the actual written policy said that the rules applied to everyone. And the rule was frequently that you could not get treatment without proving residence.

When residency restrictions were implemented in neighborhoods that generally have excellent access to healthcare, they were frequently executed with enthusiasm and ruthless efficiency.

Organizations throughout the United States devoted a huge portion of their efforts early in the vaccine rollout to make sure they were reserving ‘their’ doses for ‘their’ people.

Our society stranded doses in freezers, for justice. We instituted and enforced policies to deny healthcare during a pandemic. We called men with guns into places healthcare is delivered, to throw out people whose crime was seeking healthcare. And we now sleep the sleep of the righteous.

Who is to blame for this? Blame is an irrelevant concept; only our actions and results matter. Our actions created an internal command economy; our results were consistent with those usually experienced by command economies.

This happened. This really happened. This really resulted in a majority of doses sitting idling in a freezer, some of which expired and were thrown out. Those who gave out those doses faced threats and retaliation for doing so.

From the post, here’s Governor Newsome at a press conference on 28 December 2020.

I just want to make this crystal clear. If you skip the line or you intend to skip the line, you will be sanctioned, you will lose your license. You will not only lose your license. We will be very aggressive in terms of highlighting the reputational impacts as well. We are going to be aggressive here.

This was meant to apply to those giving out such shots, even to random people to prevent them from being thrown out unused.

Focus only on what matters… to you.

This is from a later section:

Israel had a very sensible policy: At the end of the day, give literally anyone leftover shots. Run into the street and ask passersby if they’d like the vaccine if you have to.

Not to keep banging this drum but it is an important one: We fired and stripped licenses from professionals who gave away end-of-day shots to disfavored potential patients, including in several well-publicized cases. In some cases, healthcare providers adopted policies to prohibit end-of-day shots because their lawyers told them that if they violated the tier list then the state would revoke the pharmacy’s permission to do business at all. I wonder where they got that cockamamie notion.

Do not let this get memory holed.

People Were Made To Feel Bad About Getting Vaccinated ‘Too Early’

This was a constant problem. I dealt with it constantly with my own family, and also with lots of readers, concerned that they were ‘jumping the line’ or otherwise doing something unfair to get vaccinated when they were first able to do so. There was also vague FUD spread about potential retaliation for ‘skipping the line’ or getting wrong what phase things or someone was in.

So let me add to the chorus of people who can verify that this was a substantial practical barrier.

Noticing Confusion

Why did this happen? Patrick cared about and wanted it to be one way. Those with power, and the systems with power, cared about and wanted it to be the other.

An important confusion is what people who are interested in ‘equity’ care about.

Another simple result from queuing theory: Not letting people change lines is guaranteed to lengthen the time it takes Bad Luck Bob to check out. People changing lanes is an important mechanism for keeping all cashiers active, which maximizes the throughput of the store as a unit. That seems like a result people interested in equity should pay more attention to.

This is exactly backwards. Equity concerns stand directly in opposition to productivity concerns or to combined utility concerns. Caring about equity is to say that you care about (the appearance of caring about) the relationships between the outcomes for favored groups versus unfavored groups rather than caring about getting better outcomes in general.

Patrick notices repeatedly that the system chose these equity concerns, of stopping the wrong people from not getting sick and dying, over the concern of less people getting sick and dying. He also noticed that this failed to achieve equity on its own terms, because it created complexity and awarded the ‘not dying’ prize to those with and favored by power and those with the ability to navigate opaque and complex official systems.

What if only half of people can switch lanes? What if switching lanes requires privilege? Then you definitely want that half to move out from in front of less privileged people in the slow lanes and into the lanes that have less waiting people! It will get those people with less privilege checked out sooner! That is your real goal, right, not simply avoiding line switching for its own sake?

Residency restrictions are policy bans on switching lanes.

In terms of modeling logistics, fully accurate. Can confirm.

In terms of what these people care about: Again, no. This is backwards and I want to ensure that this fact is text. Note this, about some potential funders:

I didn’t engage with debates about how, and this was made absolutely explicit in some conversations – perhaps saving lives but failing to save lives in preferred demographic ratios would be considered worse than not engaging in the project at all.

Such folks do not much care about getting the less privileged people checked out sooner. They mostly care about the system not doing things that get the more privileged people checked out faster, and they care about not giving those people additional privileges, such as the unique ability to switch lanes. Oh no.

This generalizes.

The Chains Lied About Vaccine Availability Because We Didn’t Pay Them

To be clear, when you pay so little to those providing a vital lifesaving service that they do their best to avoid providing it, such folks are not exactly covering themselves in glory, but this is on you, the person who is not paying them enough money. The amounts required would have been trivial.

An information source that made the whole process less time-consuming would also have helped a lot. You lie in a recording or on a website and say you don’t have information in order to not have to answer the same question over and over again, in addition to all the other reasons you might do that, and because you do not want a lot of people getting the idea that they should call you and then getting mad when told no.

A less fun operational wrinkle of using pharmacies to enact healthcare policy: The pharmacies were not necessarily thrilled to be doing it. The reimbursement the pharmacies got for administering the vaccine (approximately $40 a shot) is not a lot of money to a pharmacy. They did not optimize their operations to capture it. (A maximally charitable view of this is: ‘Look, we’re in charge of almost all routine drug delivery in the United States and, in aggregate, that is staggeringly more important than one drug for one disease. Of course we shouldn’t drop what we’re doing to optimize for it. Do you have any idea how important the day-to-day work of pharmacies is and how many lives hang in the balance?’) Particularly early in the vaccination effort, the amount of pharmacist time that was spent fielding calls from the general public was substantial. The Covid vaccine was a distraction from the business of running a well-managed pharmacy and was institutionally treated as such.

So the chains made some choices. One was to lie to patients about vaccine availability.

That sounds like a shocking allegation, and it is so gobsmacking that we had to train our callers about it. ‘When you call XYZ Pharmacy, you will get a chain-wide recording that says that the pharmacist has no additional information about the Covid vaccine and that you should visit their website instead. Their website has no useful information. The recording is lying. Ignore it and press 4 to speak to the pharmacist, then continue to ask your scripted questions as normal.’

Regardless, the result is the same: Pharmacies failed to tell the truth (and said some things that were the opposite of the truth) about the availability of the vaccine to patients, pervasively.

The pharmacist was frequently the only person alive who knew whether their pharmacy had the vaccine, and it would have stayed that way, had we not ignored the lie and asked the pharmacist anyway.

The chains also intentionally had broken websites, as rationing mechanisms, to avoid having customers come to them to try and get vaccines and get mad when told no.

Another somewhat desultory consequence of using the pharmacies as distribution partners was that this exposed many patients to their systems for scheduling appointments. These systems were some of the most important software written in 2021 and they were broken by design.

Why? Because there was a true systemic supply/demand mismatch early in the vaccination effort. If, e.g., BigCo had allowed one to easily sign up for an appointment, BigCo would quickly become known to vaccine seekers and the media as the place to get an appointment. BigCo would tell the overwhelming majority of appointment seekers that there was no appointment available. BigCo would then take a hit in the press and plausibly get called into Congress to answer for why they had no appointments despite being part of the FRPP. BigCo’s competitors would not have this problem because their websites were falling over during crushing demand.

A website that falls over doesn’t disappoint millions of people daily, not for long anyway.

So instead BigCo sometimes intentionally and sometimes through considered inaction engages in rationing through mediocrity. BigCo has good vaccination scheduling systems in 2022, and BigCo could have shipped them in early 2021, but BigCo chose not to. They slow-rolled their development and kept decision-making several layers from the top of the company, deep in their technical orgs, where no one who mattered would ever be called in front of Congress.

Was this strategy successful? Well, do you remember the time their CEOs were called before Congress to account for their performance during the pandemic? *crickets chirp*

Whereas the mid-tier pharmacies had the opposite problem, where not enough people would seek them out and they wouldn’t be able to use all their doses. Nor would they be able, of course, to transfer the doses across the street, Patrick doesn’t even mention that possibility it is so absurd on its face, I mean come on.

Not all broken systems, of course, were broken on purpose. One advantage of being broken by accident is that when people notice, they can fix it.

A Rite Aid pharmacy in San Bernardino asked our caller to sign up for an appointment at the county health department’s website. Our caller, who had been calling into San Bernardino frequently and had seen that website frequently, remarked that he had seen no Rite Aid listed as a possible vaccination location.

The pharmacist then swore into the telephone, hung up, and immediately called the county health department.

There were 13 Rite-Aids in San Bernardino county. None of them, despite being in possession of the most desirable object in the world, had received a single appointment. No pharmacist, with years of training in healthcare, noticed this before we told them.

Each of these organizations wants someone else to be responsible for catching errors like this, and they want them to be effective at doing so. They want, and the nation wants, an organization to be accountable for delivering the vaccine.

VaccinateCA considered this bug, and anything else that kept vaccines in freezers while patients were still waiting, to be our problem.

We kept looking for logistical bugs like this and fixing them for our entire run. I hoped we’d find examples this dramatic at scale; we didn’t. Instead, our largest impact was likely through leveraging the largest organizations in the world.

America Still Pretty Great, All Things Considered

VaccinateCA ended up expanding across the country, vouching for data, connecting their map into places like Google and greatly enhancing our ability to get shots into arms.

The thing that VaccinateCA didn’t end up doing was going international.

The effort wasn’t designed to be sustained long enough to do that, and going international took away many comparative advantages.

We didn’t do that.

The team had signed up for a short, sharp sprint for California and was now months into a project for the entire United States. Partners were getting antsy, children missed their parents, and employers were starting to ask, ‘So now that the pandemic seems basically done, when are you coming back to work?’ And despite all that, if we had had any reasonable path to international impact, the math remains the math and the moral case remains the moral case.

We just couldn’t find a path.

There were parts of the VaccinateCA model that took advantage of relatively unique features of US healthcare infrastructure, like widespread distribution of privately operated pharmacies that had been turned by the government into a primary distribution channel for the vaccine. We didn’t think we’d be able to take advantage of that in most nations. We would instead be back to understanding virtually nothing about relevant healthcare infrastructure while facing even more disadvantages than we had faced on Day 1.

More than that, America still has one very large advantage.

We also benefited from another major strength of America: You cannot get arrested, jailed, or shot for publishing true facts, even if those facts happen to embarrass people in positions of power. Many funders wanted us to expand the model to a particular nation. In early talks with contacts there in civil society, it was explained repeatedly and at length that a local team that embarrassed the government’s vaccination rollout would be arrested and beaten by people carrying guns. This made it ethically challenging to take charitable donations and try to recruit that team.

On Vaccine Hesitancy

At the same time that everyone was making maximalist claims about the vaccine with impunity, they didn’t do this where it would matter most, which was affirming that the vaccine was free.

I cannot underscore enough how much ‘vaccine hesitancy’ was frequently a solvable problem invoked as an excuse for institutional incompetency or unwillingness to do hard work. Or easy work.

We do not call it ‘free-money hesitancy’ when you need to explain to students how Pell Grants work, patiently walk them through the application process for financial aid, and reassure them that the offer really is as good as it sounds, even if they come from a socioeconomic background where nobody, and certainly nobody with a graduate education, has ever given them anything with no strings attached.

Most publicity about the vaccine failed to mention it was totally free, because lawyers said, ‘Well, actually if you have insurance we’ll bill the insurance and by law you will have no out-of-pocket cost but that is not the same thing as free’. Lawyers, man. When I was six years old I learned ‘free*’ is how you spell ‘not free’ in English, because they wanted shipping and handling for the G.I. Joe even though I had collected all the Cheerios box tops those lying liars said to. Many people in society vividly remember similar experiences that had much worse consequences.

People died over that asterisk. Why?

Safe? Sure. Effective? Absolutely. Free? Well, let’s be careful about that one. WTF?

I knew the vaccine was all three of these things, so I wasn’t automatically tracking how often reminders included each of them, and I don’t have a good memory here.

Quick Notes on How To Actually Do a Thing

  • Most impact was via feeding the data set to other publishers such as Google. This worked because what mattered was shots in arms, not the credit. Whereas who got the credit was a question that crippled many public efforts.
  • Including people who made phone calls to someone else, who then used Google.
  • Where did credit still matter? Funding and volunteers.
  • Consider everything your problem. Every issue, every bug, every failure.
  • Getting funding for tech companies is much, much easier than getting funding for non-profits, even if both are exactly the same good-cause-that-will-never-make-money operation. In this case they still had to go non-profit for other optics reasons.
  • A lot of funders want to Create Institutions rather than Do Thing. The more paperwork they have you fill out, the more likely they are to have this preference.
  • Even with extraordinarily good conditions, initial raise was only $1.2 million and Patrick spent half his time seeking money.
  • Even with outrageously great success, by the end, funding was unavailable and the project had to shut down a bit earlier than desired.
  • Formal organization is needed to shield liability and interact with funders and the government, but things got off the ground first by simply Doing Thing.
  • Urgency and a united sense of purpose were keys to success. Everyone involved cared about getting shots into arms, no one cared much how they got there.
  • Lots of very long hours. Important.
  • Numbering the days, then always asking ‘could this be done on an earlier day?’
  • Lots of eager volunteers due to the obviously noble and important cause.
  • Volunteer callers could act as nimble scouts, experiment, find the best strategies. Whereas paid people at call centers need to be executing a script, so you need others to figure out how to do it.
  • Business development is Definitely a Thing as are the deals it enables. Getting onto Google supercharged the impact of the project.
  • Remember that almost everything is ‘a combination of engineering marvel and terrible jank.’
  • Being in a neutral position and having no political agenda was vital.
  • Dealing with tons (3k+) different things by having humans find and parse them one at a time doesn’t scale well but it does scale, damn it, if you want it to.
  • Generalize that. It is hard work. It also kept working.
  • Anyone could coordinate these efforts if they go and do it. Observe:

We saw peer projects sprout up in many states, with varying levels of effort and success. Many credited us as an inspiration. One peer project was ILVaccine.org, a project of Eli Coustan. We were working with him for a while before I learned he was in middle school. When I later blanked on his name and asked someone about the public health infrastructure coordinator who was a middle school student I was asked to be more specific.

A lot of this seems highly relevant to my current tasks at Balsa. I have already been reminded of some of these lessons, including that for-profit companies have a vastly easier job acquiring funding. Even in explicitly EA spaces that are supposed to be non-profits, compare investment in OpenAI, Anthropic and Conjecture (or only Anthropic!) to the amounts spent on, well, everything else.

The sense of urgency and energy and long hours hits hard, both in its presence and its absence. As Balsa is in its current form a long term project without any hard deadlines, there is no ‘could this be done on Day 6 instead of Day 8?’ attached to everything, and I am pacing myself in a sustainable way rather than doing everything possible because every minute counts. I don’t have the ability to focus in totally on a thing for long periods the way I used to, as I am no longer as young. There is a huge difference.

On the flip side, the volunteer enthusiasm hits hard. When Balsa was announced, there was a giant outpouring of people eager to spend their time making things work. For several weeks I filled my calendar with phone calls, thereby getting lots of new info some of which was learning exactly how many phone calls with new people on a variety of topics I could handle before my brain stopped working right, and that I need to not be the person doing lots of volunteer coordination.

Overall Impressions and Takeaways

There are two central stories, the internal and the external. This post focused on the external.

The internal story is at heart a simple story of the power of Doing Thing. When people come together with the intention of working hard to damn well Doing the Thing without regard to credit or compensation or how it looks, they can do quite a lot. When they have experience in such matters, they can do even more.

The external story of America’s vaccination efforts is damning. A lot of this information is in danger of falling into a memory hole.

In particular: America chose to care about the appearance of favoring approved groups over disapproved groups, rather than to care about saving lives, getting shots into arms, people not getting sick and being able to resume their lives and other neat stuff like that. We retaliated against doctors who gave to ‘the wrong people’ doses that would otherwise expire. In doing all this, we created a complex web that favored only and exactly the same managerial class that was signaling their concern for ‘equity’ at the expense of saving lives, while those the policy claimed to benefit lacked the skills or resources to navigate the systems and were left behind even more. Funny that.

Remember how this prioritization process worked. Expect it to happen again.

Various other political concerns were also allowed to override the ability to do a variety of things that would have allowed shots to go into arms. Thanks to ridiculous government-imposed requirements, our existing vast logistical infrastructure that commonly delivers vaccines with exactly the same shelf-life conditions was not used, and instead doses were left untracked. Tech companies were shunned rather than embraced and felt the need to not be visible. Pharmacies were paid so little that some actively sought to avoid giving out doses and intentionally crippled their own websites. The government preferred not sharing vaccine info anywhere to giving the impression that some states were served better than other states. Local and state politicians did not want to share their information with the federal government lest they lose the credit.

That is only some of the things in this summary, let alone the original post.

It is vital that we not let any or all of this fall into a memory hole.

I will conclude with a question and answer from Patrick’s AMA on the Progress Forum. A lot of lowering of status, not much raising, the real government heroes are people whose names we never hear. Be sure to note the contrast between pharmacies and pharmacists, which reflects sharp contrasts throughout the OP where pharmacies consider vaccines an annoyance and pharmacists want to prevent people from dying.

New Comment
2 comments, sorted by Click to highlight new comments since:

What should we do about equity?

When social systems systematically deny some people access to goods, the net badness of that is more than would be expected just by summing over how bad it is for each person not to get the thing. If we both have a dollar, it is a better world overall than one where I have two dollars and you have one cent. Fairness is valuable, and systemic racism is icky.

It also has a way of falling down that "memory hole". People who can like to forget it still isn't a solved problem.

It seems like there was the appearance of an attempt to apply a constraint, to prevent people with systemic power from getting the scarce good unless people without that power had gotten it. That constraint seems to have been both costly to try to impose (because, for example, it resulted in a lot of wasted doses), and also not very effective (because, for example, it involved residency requirement hoops that everyone should have known would have the opposite effect). It's entirely possible that the whole project was actually a scam, and not a genuine attempt to apply an effective equity constraint at a manageable cost.

But the correct alternative is not to just not apply any equity constraints. If you apply a constraint that is a net benefit to the people it is supposed to help, and a larger net cost to the people it is constraining, it is justified if it evens out a large enough systemic difference in well-being.

We should look at why California's equity constraints seem to have failed to produce any equity, why they were so costly, and why they were able to masquerade as a real attempt to produce equity if they were not.

We should not abandon the notion that equity is a good thing that is worth paying a net cost to get.

The correct alternative was absolutely to not apply such constraints, but that's because supply should have been a non-issue. Paying $500/shot, for a course of vaccination begun in the first month, would have cost much less than 0.3 taken off of the 1.9 trillion dollar COVID relief bill of early 2021. This should have been literally free.

Zvi spends a lot of time talking about the problems of choosing scarcity - and dishonestly ignoring the evidence that African-Americans were more likely to die if they got infected - but the actual question should be why this wasn't trivially solved in the run-up to January 20, 2021, aside from that spot of bother concerning the changeover.