On 01/04/2014 07:31 PM, Chris Morley wrote: > > We have no formal patch submission process. > Fairly often someone presents a patch to us and it > gets no response.
I agree with both of those statements, but i don't think they're that closely related to each other. Formal processes are not what creates responses: human time and attention are. Human attention is the key resource that the LinuxCNC project is short on. > While thinking of solutions I came to only two ideas: > > - 1) Someone volunteers to follow up patch submissions. > > This would mean: > evaluate or get evaluated the idea and the code. > If accepted, push the code to the right place. > email a response of some kind > > This would require: > Support from the rest of the developers in opinions, > evaluations and freedom to push stuff that is not thrilling > but hurts nothing. > > This is close to what we have other then a volunteer would > make sure this was done for all patches, not just ones he > was interested in. (he could of course delegate if need be) If i understand this idea, you're proposing a person to act as a kind of interface between the person submitting a patch, and the developers who'd review the patch? That'd be useful because the submitter would be acknowledged and not feel ignored, but it would not solve the second part of the problem: someone with deep technical knowledge of linuxcnc needs to evaluate the patch. This means understanding the problem that the submitter is concerned about, understanding their proposed fix to the problem, and iterating the patch itself for correctness and completeness (including tests & documentation). This is not an insignificant investment of someone's time. As the saying goes, "if it was easy, everyone would do it". > - 2) patches are added to the linuxcnc-meeting so at least > there is a small discussion on the matter. > possibly find a person to evaluate and push the code at > this meeting If i understand this idea, it's like #1 except that the interface person is the monthly committee? > I think our prime problem is that our push-capable developer > group is small so time and expertise become scarce. Push access has nothing to do with it. All the work is in examining the patch and deciding what to do with it. Once that's done, pushing takes literally 30 seconds. > I also think that it's tough to evaluate patches that aren't > our interest, when we (as a group) have not really defined > what directions linuxcnc is interested in. > > I hate the thought of losing ideas or code because we seem > uninterested or the barrier is too high. > > Opinions? My wish is that folks who can, contribute in the most meaningful way possible, given the time they're willing to spend volunteering for the project. I think we have a bunch of people who are capable of understanding incoming patches and bug reports, if they take the time to spend on it. It's not fast and easy for any of us, and we all have other demands on our time. IMO it's up to each of us volunteers to step up if/when we can. Processes and management are not applicable or useful in this situation, volunteer hours are. That said... If someone volunteered to do #1 above i think it would be a valuable addition to our project. It would make the situation strictly better, even though it wouldn't fix the problem completely because it wouldn't address the task where someone actually spends the time to evaluate the patch. -- Sebastian Kuzminsky ------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
