I'm glad HN has decided anecdotal evidence is more powerful than data. This is exactly what the tech sector needs to thrive.

Data is so easy to manipulate and living in SF I saw it manipulated all the time.

For example during the pandemic one week close to the election suddenly there were posters plastered all over the place for getting tested. I’m talking 100 on every block it was crazy. They doubled the pop up testing areas. People came to our door knocking and telling us to get tested multiple times.

Not a week later I see the NYTimes with a front page top headline “surging cases of Covid in San Francisco” with very scary looking data about the spike in cases.

I looked into it. The spike was entirely due to testing. The article didn’t mention it until the very end and only in the most purposefully vague way. CNN and all the others had similar articles.

This is one of so many cases (Lancet article comes to mind), I’m talking literally hundreds I’ve seen of totally misleading data or very sketchy data backing peoples preconceived priors. The replication crisis is no joke.

Honestly I believe what I see, and I believe first hand anecdotes more than most data.

It is more powerful. Data is often meaningless without interpretation, missing many dimensions, easily and often manipulated, funded by unknown desires and forces, and generally unreliable. Outside of the hardest of sciences it’s worse than worthless.

Meanwhile a single sharp person who lives in a city and interacts with the right intersection of people will tell you closer to the truth than even the better quality studies or media outlets. To think otherwise is imo foolish.

You're right it's much better to rely on information that is expected to be meaningless, has zero dimensions, is guaranteed to be biased, has no backing, and is scientifically proven to be unreliable.

Exactly what I expect from HN's crowd.

If your trust for first hand accounts from people you trust is meaningless, I feel sorry for you. And if you aren’t able to intuit their biases based on your history with them and take that into account, more so.

That you seemingly missed the caveat that I mentioned about “hard sciences” also not a good sign.

I do not trust you, no, because we know as a scientific fact that your perceptions are warped. That's why we do science.

Good, I said people you trust, not me. You shouldn't trust me, but I trust me and my friends.

I don’t care about homelessness in an academic sense. I care about the crackheads on my street. Anecdata is fine.