On Wed, 2010-04-28 at 12:12 -0400, Valdis.Kletnieks@xxxxxx wrote: > On Tue, 27 Apr 2010 23:00:30 PDT, Joe Perches said: > > On Tue, 2010-04-27 at 22:52 -0700, Greg KH wrote: > > > > I seem to be learning that you little taste. > > > -ENOPARSE > > > > Choosing competing patches based on date received > > order not quality is poor taste. > > So you're saying when Greg gets a somewhat ugly but passable patch 2 weeks ago, > he's supposed to *just know* that you'll be submitting a possibly better one 2 > weeks later and wait for it to show up? No, I'm saying that when Greg gets multiple patches for the same module and doesn't act on any of them for several weeks, (in this case one 6 weeks ago, and two others 4 weeks ago) he should select the better patches, not just apply the first one in chronological order. > How is that supposed to work in reality? Apparently slowly and fitfully. If the cycle time was shorter and the feedback better, it would be less of an issue. Greg has explained he was unavailable to look at patches for an extended period. Unfortunately that extended period was immediately after a talk Greg gave and an article Greg authored designed to generate new contributor patches for staging. The talk and articled worked. Many new individuals followed his template format producing cleanup patches, some good, some less good. Few if any of those patches received feedback from Greg or had notice of patch status for that extended period. It would be better if the backlog of patches were order sifted by quality than chronology. cheers, Joe _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
