[__jal]

Some of it is historic - Mysql has gotten much better in recent years at supporting the parts of being an RDBMS that matters the most when money is riding on it.

So honestly, at this point I do think some of it is impressions from the past which are no longer valid. But still, Mysqlhas done/does all sorts of things that defy the spec, convention or just common sense (I don't know if this has been fixed, but at least for many years, April 30th was treated as a valid date, and there was some profound weirdness of which I can't quite recall the details involving locale stuff).

Postgres generally takes the position that data should always be safe first and speedy sometime later. It also assumes the operator understands their tools. That second one means in comparison with Mysql, people think it is unfriendly. It isn't (if you want to see unfriendly, go work with Oracle), it just expects that its friends learn about it. Which is of course good advice when you're dealing with complicated software on which a lot of money tends to ride.

And as the PG devs have said for years, they don't compete with Mysql. They compete with Oracle. There's no reason to switch if you're happy with Mysql.

[rimantas]

What's wrong with April 30th? And operator understanding their tools applies to MySQL as well. You can set it be more strict, but default is lax. Not sure if reverse is possible with PG.

[spacemanmatt]

PP probably intends to refer to February 30th (or 31st or 32nd), a popular handle to MySQL's (ahem) surprising date handling.