It's not a peer-reviewed controlled study, of course, but HN isn't a scientific journal. In my view the post is clearly substantive enough for a general interest site.
There's such a thing as a substantive anecdote, and self-observation and self-experimentation have a long and noble history. It's wrong, of course, to draw hard scientific conclusions from such a piece, but it's not wrong to read one, nor to write one, nor to do this kind of experimentation, nor to find it interesting. Plus there's a hacker angle in the DIY aspect.
That's my take on it, anyway. I thought it was cool, and I hope that HN readers are smart enough not to draw the obvious bogus conclusions.
One of the upsides to such posts is they can trigger comments that reference more rigorous sources. While I don't think this is intentional it feels like a variation on the trick of getting an answer to a question by either saying something can't be done or by posting obviously wrong or otherwise flawed code.
Comments
How substantive is an n=1 anecdote about some biomarker interventions with no hard endpoints?
It's not a peer-reviewed controlled study, of course, but HN isn't a scientific journal. In my view the post is clearly substantive enough for a general interest site.
There's such a thing as a substantive anecdote, and self-observation and self-experimentation have a long and noble history. It's wrong, of course, to draw hard scientific conclusions from such a piece, but it's not wrong to read one, nor to write one, nor to do this kind of experimentation, nor to find it interesting. Plus there's a hacker angle in the DIY aspect.
That's my take on it, anyway. I thought it was cool, and I hope that HN readers are smart enough not to draw the obvious bogus conclusions.
One of the upsides to such posts is they can trigger comments that reference more rigorous sources. While I don't think this is intentional it feels like a variation on the trick of getting an answer to a question by either saying something can't be done or by posting obviously wrong or otherwise flawed code.
Basically, xkcd 386.
statistically powerful stories ⊂ substantive stories
Thomas, in relation to my question and Daniel's answer, could you kindly explain the direction of the subset relation you are talking about there?