The gist of the post was to correlate technologies listed on companies' AngelList profiles with those companies' Signal scores. Signal scores are a bit of a black box, but I was told by an AngelList employee that they're roughly like PageRank over the graph of startups, founders, and investors. That is, a startup's Signal score will be higher if the people affiliated with that startup have high Signal scores.
There were some correlations between higher Signal scores and lower usage of PHP, higher usage of languages like Go and Scala, etc. That supports PG's point, but probably lacks the statistical rigor that you'd want.
The gist of the post was to correlate technologies listed on companies' AngelList profiles with those companies' Signal scores. Signal scores are a bit of a black box, but I was told by an AngelList employee that they're roughly like PageRank over the graph of startups, founders, and investors. That is, a startup's Signal score will be higher if the people affiliated with that startup have high Signal scores.
There were some correlations between higher Signal scores and lower usage of PHP, higher usage of languages like Go and Scala, etc. That supports PG's point, but probably lacks the statistical rigor that you'd want.