tailcalled comments on Are there any good, easy-to-understand examples of cases where statistical causal network discovery worked well in practice?

[tailcalled]

This matches my impressions relatively well.