Correction: "That is, is commons-beanutils managed by a different PMC than commons-*logging*?"
On Thu, Jul 17, 2008 at 3:36 PM, Les Hazlewood <[EMAIL PROTECTED]> wrote: > > > On Thu, Jul 17, 2008 at 3:10 PM, Niklas Gustavsson <[EMAIL PROTECTED]> > wrote: > >> > Why even worry about that while we still have beanutils as a dependency >> and >> > commons-logging works fine, or if there are CL issues, >> jcl-over-slf4J.jar >> > works fine. >> >> As for beanutils, if you haven't already, feel free to open an issue >> with beanutils and let the same discussion start over there. There is >> at least one commons commiter on this thread (me) so we can carry the >> discussion over. >> >> As a future user of JSecurity, we're in the works of throwing out our >> homebrewed solution in FtpServer for JSecurity, I'm in the opposite >> situation with regards to download size. With the inclusion of >> JSecurity, we will now have to ship jcl-over-slf4j as we use SLF4J. We >> already do for Spring but that's an optional dependency and hopefully >> one that goes away with time. That said, that's just one sample out of >> many, there are of course others in the reverse situation. > > > You have to do that anyway because of the beanutils dependency (unless you > won't use JSecurity's text or web.xml configuration - I don't know your > usage scenario). If the next stable release of beanutils had SLF4J as a > dependency instead of commons logging, it would make much more sense to me > to switch to SLF4J. But I don't know it is worth doing so until this > happens. > > Are commons projects managed by different PMCs? That is, is > commons-beanutils managed by a different PMC than commons-beanutils? I know > they report to the Jakarta PMC, but I was wondering if there were sub-PMCs > that have respective decision making ability. If not, it would be easiest > if all commons projects used SLF4J, and then we wouldn't have to worry about > it anymore for almost any Apache project. > > Or, maybe a better question is that why can't the next version of > commons-logging (2.0?) just do what SLF4J does today with static binding? > If everyone is so eager to move towards that model, why can't they make the > change and then SLF4J becomes reduntant? The benefit of this is that we > would be able to continue to use an internal Apache project, not an external > one. > > I have a strong suspicion that this request has been beaten to death by the > commons-logging respective PMC (it seems to be a somewhat frequent request) > so I don't know that it'd be worth joining the user list and debating it > further. Maybe it would, but my main focus is on our upcoming release > candidate, and I don't know that I could take more logging debates... > > Les > >
