- "Coffee reduces risk of developing Parkinson's"
- "Wine helps reduce heart disease"
- "Nicotine stimulates cognitive function"
Like many other drugs, caffeine has some upsides, but also some pretty significant downsides (its dependency-forming properties being one of the big ones).
And since this is HN, adding new dependencies to your life seems analogous to code: Introducing new dependencies to your repo should be done thoughtfully and carefully. Sometimes the pros outweigh the cons (the dependency does something that would save you a ton of work), but there are usually downsides to taking on dependencies as well (increased security liability, relinquished control over part of your stack, more build complexity, slower builds, etc)
Is that why the analogy is bad? In other words, if a line of reasoning leads you away from the covid shot, then that line of reasoning must be flawed (horrifically, even)?
The main downside is it effecting your sleep, which for most people can be controlled by not having it after a certain time
So there is a correlation between coffee and being healthy, but causation is very messy.
Again, I'm very very comfortable with. Wait -- millions of people have been drinking coffee daily FOR YEARS, with VERY FEW visibly super-harmful side effects.
That data is as good as, or perhaps better than, modern medical data -- which ITSELF cuts both ways.