I attened PGCon around 2007. There was a session on why it was so difficult to get submissions accepted into the code base. Postgres had a reputation for being

I attened PGCon around 2007. There was a session on why it was so difficult to get submissions accepted into the code base. Postgres had a reputation for being difficult in this way. There was even a t-shirt from the conference, which read something like "-1 from me, Regards, Tom Lane". If I remember correctly, part of the answer was that Postgres has extremely high standards for the right way to do things, including coding style. And that if you just jumped over a bar this high, you too can get your patch accepted. They discussed the age of the code base, and the fact that if they didn't maintain extremely strict standards, that the code would just decay. And it hasn't. I am occasionally on some of the PG mail lists, going back eleven years. One thing that has always struck me is the incredibly high level of competence displayed in pretty much every discussion. And this is on every question, from newbies, and from experienced people asking difficult questions. Often (usually?) some of the most senior members of the Postgres team will answer. I rejoined pgsql-general again recently, and still, Tom Lane is a frequent contributor. It really is an amazing community, and this is due to the model set by the long-serving PG brain trust.