Vinay Prasad, the UCSF heretic against Faucism, recently posted the following on Twitter:
Heading back to SF from Wash U St Louis
It was great to give 4 talks over 2 days.I hope to make at least 2 available on YouTube, & 3 on plenary session.
It was heartening to talk to so many sensible doctors. When it comes to COVID19 ppl stopped me in the halls to say…
Masking 2 year olds…
School closure…
Vaccine mandates for college kids…
Lockdowns…were all illogical and horrific public health and they are glad I spoke up
Many agreed but the climate was too hostile for them to speak up
Is it fair to say that this is like the Resistance in France during World War II? Before 1945, nobody was a member and after 1945 it transpired that every French person had been a member of the Resistance?
(I had thought that support for the Islamic Resistance Movement (“Hamas”) was the anti-Resistance in the sense that Palestinians polled prior to October 7 generally said that they supported Hamas, but after October 7 we were informed that no Palestinians supported Hamas. That’s still what one learns from the media and the typical Twitter feed, but, as noted in International Day of Solidarity with the Palestinian People (and a poll result), Palestinians polled in November 2023 supported Hamas (75 percent) and Palestinian Islamic Jihad (85 percent).)
Dr. Prasad is a vaccine believer: “People died by delaying the initial vaccine.” I think he means that people died because it took a few extra months to get the vaccine approved and distributed. That’s probably true for Israelis because if the vaccine been approved prior to the 2020 election, Trump would have been reelected and Hamas, Palestinian Islamic Jihad, and various civilians likely wouldn’t have dared to kidnap, rape, and kill U.S. citizens for fear that the unhinged guy in the White House would push the button on them.
I still can’t figure out why we aren’t able to see the effects of the vaccine on a country-by-country basis, e.g., a lower excess death rate for a country that got vaccinated sooner. See Where is the population-wide evidence that COVID vaccines reduce COVID-tagged death rates? for a question that remains unanswered despite some studies within countries and the faith of otherwise reasonably skeptical people such as Dr. Prasad.
My own skepticism from three years ago… If COVID-19 vaccines weren’t tested on likely COVID-19 victims, how do we know that they will reduce COVID-19 deaths? (the kind of people whom SARS-CoV-2 had been killing were excluded from the vaccine trial):
Let’s have a look at the Moderna FDA paperwork. Only 3 people in the vaccine group, out of 15,208 total, died during the study (approximately 3 months; see pages 17 and 18), which tells you that Moderna picked a much healthier population with a much longer life expectancy than the kinds of people who have been tagged on death with COVID-19 positive test result. (If we assume that a typical COVID-19-tagged death is among those with a life expectancy of 4 years, we would have expected at least hundreds of deaths during a similar study of vaccination among people who really need the vaccination. Note that the Swedish data suggest that 4 years is an overestimate.)
Also three years old, but more fun… #MarkedSafe from Homemade Cookies and Crafts:
Dear Parents and Caregivers,
As we approach the December break, it is a time of year where many families and school staff like to give homemade baked goods and crafts as gifts of appreciation. Due to the ongoing COVID-19 pandemic we are going to respectfully ask to put a hold on this practice as a part of our ongoing efforts to keep everyone safe.
We have all worked hard to keep each other safe and to keep our schools open. We appreciate your willingness to find alternative ways to express your gratitude this year. A letter to the teacher with a specific thanks would be greatly appreciated!
(That was at a school for rich suburbanites. It was open with a mask order in force (later they added vaccine coercion). City school systems ostensibly serving the poor in Maskachusetts were still closed and would stay mostly closed until fall 2021.)
That’s a bit of an odd list that he put together.
Were those the points where he thinks he beat conventional wisdom, or the main points where he disagreed at the time?
In my experience, people tend to remember their correct predictions, forget their wrong predictions, and claim they were always right. (See Scott Adams)
Complaining about college students needing vaccines seems petty (and not even correct, on a risk-adjusted basis). How many required shots does a kid need before reaching college-age?
If you said “We’re adding one more shot to the list, for the contagious disease that killed more people than any other in the last 100 years”, would that seem bizarre and unwarranted?
I have little doubt that when people look back at the oddness of the last 5 years, it will sound something like “We were stuck with the most dangerous virus 100 years, one that ultimately killed more than 1 million Americans, and for some hard-to-explain reason, people from one of our two parties were in denial that any measures would work, and that ultimately increased the death toll by several hundred thousand people. People who routinely took vaccines throughout their life refused to take this one because the vaccine became associated with the other political party, even though it was developed under the president of their own party.”
Crazy times.
* I wonder if there’s a similar post from people at these MAGA conferences where the smart people get together and say “I know the vaccine was saving lives, but I couldn’t say it out loud for fear of losing my talk radio show/fox news slot/florida governership.
Most of the anti-COVID-vaxxers I’ve talked to never bring up politics per se, but they do bring up distrust of the government that admittedly lied to them during COVID management.
Hey Professor – Unrelated, but do you have any opinion on Palantir the company and the approach they’ve taken (ontology plus an RDBMS with a user defined schema)?
It seems like layering their product on top of all of those old SAP & Oracle tables is a winning strategy (vs rip and replace and/or graph databases)…
But then again I don’t know anything, which is why I’m asking an EECS professor!
James: I’m embarrassed to admit this, but I can’t figure out what Palantir does!
https://blog.palantir.com/palantir-is-not-a-data-company-palantir-explained-1-a6fcf8b3e4cb
says “We build digital infrastructure for data-driven operations and decision-making. Our products serve as the connective tissue between an organisation’s data, its analytics capabilities, and operational execution. Palantir’s platforms tie these together by bringing the right data to the people who need it, allowing them to take data-driven decisions, conduct sophisticated analytics, and refine operations through feedback.”
As you note, this is what Oracle says they will sell you with their database and various analytics and integration products layered on top. The Interweb seems to agree that at least some Palantir products have equivalents in the Oracle stack.
https://www.peerspot.com/products/comparisons/oracle-data-integrator-odi_vs_palantir-foundry
for example.
All of Palantir’s early customers were in government, right? It looks as though perhaps Palantir’s products are more tailored to fine-grained access control (though of course that can be done with Oracle OBIEE and similar as well).
https://medium.com/palantir/purpose-based-access-controls-at-palantir-f419faa400b3
makes a big deal of that.
Both companies are in trouble for having too many white males:
https://revealnews.org/article/oracle-and-palantir-said-diversity-figures-were-trade-secrets-the-real-secret-embarrassing-numbers/
https://www.inc.com/guadalupe-gonzalez/palantir-oracle-diversity-report-silicon-valley.html
explains the preceding link a little: “According to the companies’ 2015 Employer Information Report EEO-1, an official form companies must submit to the U.S. Department of Labor, women held precious few top slots,…”
Apparently, a company cannot opt out of the DEI religion.
Oracle perhaps tried to buy Palantir in 2016: https://fortune.com/2017/06/29/oracle-palantir-acquisition/
https://www.reddit.com/r/dataengineering/comments/15r6k9i/is_palantir_as_good_as_they_say_is_it_worth_our/
is very negative on Palantir. Example:
“From experience: don’t do it. Resist the highly G lash sales pitches and flashy demos. It’s a black hole of technical debt and never ending needs for their dedicated support. It’s not as secure as they say and at the end of the day it’s largely just Apache Spark wrapped up with a big vendor lock. Just don’t do it!”
“.., I can’t figure out what Palantir does!”
I had the same question about Autonomy, with which the company I then worked for was trying to partner. It may have had one great idea (https://www.technologyreview.com/2011/08/18/192285/why-hp-wants-autonomy-math-skills/) but litigation has revealed that it got itself into trouble trying to turn that into a money-maker.
There is a class of software system suppliers whose product pitch makes more sense in the golf club bar or the CEO conference than it does to technically savvy staff who have to talk the executives down.
David….You nailed it very eloquently…..Please contribute to more conversations. Great to hear sanity in such a cogent way!