Summary

This is a $10k retroactive grant to Karl Yang, for his role in starting VaccinateCA. In January 2021, the Covid vaccines were just starting to become available to the general public… in theory. In practice, there was a lot of confusion over how to get a dose. Demand for vaccines was overwhelming; information about them was scarce. Karl took a simple idea (“locate where the Covid vaccines are, and publish it”), and organized a volunteer effort that would go on to help millions of people get their shots.

Q&A with Karl

Adapted from a 30 min interview between Karl and Austin.

Q: What was Karl’s role in starting VaccinateCA?

A: Back in Jan 2021, I’d heard that the vaccines were coming out. I was paying attention as my family had at-risk members, but nobody knew what was happening. Then, Patrick tweeted out a project proposal: create a public list of hospitals in California.

I read that and thought: “this seems like something I can do”. And I was motivated to do something. Throughout the pandemic, I’d been waiting for some way to help. There’s always a vague “they” implicit in “they should do something”. But I was antsy. Why shouldn’t it be an “us”?

Usually, work like this feels outside of my skillset or capabilities. For example, I’d love to accelerate cancer research, but I have no relevant training. But Patrick packaged it in a way that I could act on. I can set up a spreadsheet, I can coordinate volunteers, I can populate a list of hospitals I can call. I can do this.

So I set up a Google doc, found some phone numbers, and planned to start calling them the next day. I also put out a call for help on Twitter; seeing this, a bunch of people DM’d me on Twitter. I started a Discord server, inspired by the “eye mouth eye” thing (where a bunch of GenZ kids meme’d a fake startup into existence, and managed to get term sheets). Over Discord, I coordinated a campaign to call the next day. I was already working the night shift at the time, so I got up at 6am to start making phone calls.

Next, I set up a website to display what information we had. A bunch of volunteers quickly came in and made the website better; then other volunteers set up a backend on Airtable and got free credits from the Airtable team. By the next day, we had built a sophisticated volunteer dialer system. It would pull a phone number from the queue, and have a list of questions for the volunteer to ask.

We were calling hospitals in the beginning. That was the wrong hypothesis; they got mad at us for calling so often! By Day 2, we’d figured out that pharmacies were the place that had stock. Starting from Days 3-6, we discovered overlooked pockets of vaccines, about to expire. Some at a fire station; others at a county vaccine booth. We were discovering vaccines that would not have been found by other people; moving shots forward, saving lives.

Q: Did you raise any funding before starting this?

A: We didn’t stop to think about raising funding; our mindset was entirely “hey, we’re going to do this”. There was no obvious reasons why we would raise. We were just putting up a volunteer-run website, listing where there were vaccines.

Q: How did leadership of VaccinateCA transition from Karl to Patrick?

A: I had been “running” it for the first few days, but mostly it was a community scramble. I just happened to have keys to a lot of things because I set them up.

A week in, Patrick proposed that the project needed a formal CEO, and I agreed. I thought Patrick was the right person, and handed what remaining cultural power of “being first” to him.

Q: How much time did you spend on VaccinateCA?

A: I was working on this for 12 hours a day during that first week… I slept very little. This ramped down as we brought in more people and had more structure. So maybe 200 hours in total in the first month; then my involvement tapered off quite a bit after, and I returned to my day job at the time.

The impact of VaccinateCA

VaccinateCA went on to help millions of Americans locate COVID vaccines near them, first through their website and then through partnerships with orgs like Google. Patrick chronicles the rest of the tale in The Story of VaccinateCA. (It’s long but good; regrantor Zvi Mowshowitz summarizes the Key Mostly Outward-Facing Facts.) One excerpt that stands out is a back-of-the-envelope calculation of the impact of helping people find vaccines; for the $1.2 million in donations the project raised, VaccinateCA was responsible for the equivalent of saving hundreds of lives:

What is the actual human impact of getting one vaccine searcher a useful result? Early in the vaccination effort, a reasonable approximation was that accelerating a dose by one day saves 0.0001 lives in expectation or, equivalently, that 10,000 dose-days saves a life. There are more formal efforts to quantify this in the literature, but that estimate, the sheer moral weight of it, compelled me to take drastic action once I saw our stats on Day 3 and extrapolated.

One million dose-days? One hundred lives.

I believe our partnership with Google accelerated delivery of the vaccines by many millions of dose-days.

Every day matters. Every dose matters.

On the Effective Altruism Forum, Patrick reflects about the moral weight of numbers on a spreadsheet:

I consider myself fairly well-educated and well-raised in a moral tradition that has spent a lot of brainsweat on questions like "What is one's duty to society and to one's fellow man?" In that moral tradition, presented with the narrow question of "Given that one is in a position of authority and has a course of action available to save hundreds of lives, what is one's duty?", the answer is so straightforward as to be uninteresting.

But nobody, not once in my life, drew out the implication regarding expected value math until you all did.

For this you have my eternal gratitude.

And, should circumstances ever find you or yours looking at an expected value calculation that rhymes with the above, know that you'd have my instant attention and (pending thinking through it, in the words of a well-known articulator of my moral tradition) you have my sword.

One final exchange with Karl:

Q: What conclusions do you want readers of this writeup to take away?

A: That there are things that regular people can do. Especially things that people with tech skills can do, that can be enormously valuable. It took a lot of time, but our volunteers used the same stuff that they used in our day jobs to do meaningful work. And it could be done on the side; most volunteers did not stop their day jobs.

Setting up a website, ops work of coordinating volunteers, marketing; these are all normal, regular-people skills. The people who invented vaccines did amazing work, but there’s lots of work regular people can do too. My time on VaccinateCA really enhanced my belief in what I could do in the world.