Bruce Momjian wrote: > Dave Page wrote: >> Bruce Momjian wrote: >>>> Can somebody please explain to me what beta means if you can commit new >>>> stuff after it has been declared? >>> We allow /contrib to be more lax about beta changes. >> Why? When people were complaining about not being able to use TSearch >> because their ISPs wouldn't install contrib modules we couldn't >> understand why they would think that way. If we are going to be less >> stringent about /contrib, maybe they were right to cautious. > > The idea is /contrib isn't installed by default and it isn't tied into > the core code, and can be tested easier because it is stand-alone. We > can rethink that logic but that has been the guide in the past.
I think you just outlined a whole lot of arguments for pgfoundry, and not for contrib. //Magnus ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match