Re: [development] One-to-one tables considered harmful

Robert Douglass wrote:

Not many people are addressing David's actual technical argument about sorting and writing temporary tables and file sorting. So far his argument has made perfect sense to me, and I would not begrudge seeing a comment count column in the node table. I am glad Dries sees this as well.

I spoke with David a little bit about this via IRC. I did offer an alternate solution, but he wasn't too thrilled with it. (Which is, instead of combining the tables, the 1::1 table reflects the data we'll be filtering on. Which in the case of tracker query turns out to be just status = 0). It does have other problems though, and I can see why we might not want to do that.