(Apologies to John, once again. I need a round tuit...) begin quoting Rodolfo Garc??a Pe??as (kix) as of Sat, Nov 09, 2013 at 05:13:17PM +0000: [chop] >> The edge is based on the aesthetic sensibilities of the curator; better >> arguments are the way to change the edge, not the tyranny of the majority. > > But, if the arguments of the majority are dropped, could we talk about > the tyranny of the curator?
In an open-source world? Not really. The code is all there, if the majority really find the curator to be unacceptable, they'll fork. If the fork attempt fails, the majority isn't as major as they thought, or the curator wasn't as off-base as was believed. So there isn't any tyranny of a curator, as if the majority back the curator, it's not a tyranny, and if they don't, they follow the fork, and again, no tyranny. The more interesting thing to discuss is: "What problem do you have with the aesthetic of the curator?" >> If you're not going to have a small team handle the duties and argue >> among themselves, then you've got to rely on a single curator. > > Perhaps the small team is not invited to handle the duties. How big is the team for *BSDs or the Linux kernel, compared to the contributing base? I don't think WindowMaker has enough players to handle having a small team of curators, without crippling development. >> You can always fork the repo and run it yourself in parallel. That's the >> ultimate open-source solution to "I'm unhappy with the curation of $project." > > Yes, but this is the argument of the children that has the game and > set the rules. My game, my rules. It's also the fundamental argument for security: your systems, your rules. Not all lessons learned as children are bad. :) > There are a lot of project and forks for this reason. Many > applications to do the same. Is bad be a little bit open? It is not quite a bit open? Anyone can suggest a patch. Chances are, changes will make it into the codebase. The problem does not appear to be openness, but with standards and cadence; not all changes are accepted at first acceptance, and may require a couple of iterations before they're deemed be be 'up to snuff' and applied, and the time between the submission and the ultimate acceptance can be slightly longer than some people wish for. I have never seen a patch rejected without a reason, and the reasons that I have read have always seem, well, reasonable. Voting to overrule a reasonable objection in order to speed things up strikes me as unwise. Patience is not well served by the hammer of democracy. [chop] >>> I know what you said. But, why (only) you can say what is accepted or >>> not? Could you accept a patch in the git that you don't agree with? >> >> You're talking about forcing him to discard his aesthetic? > > The question is, is their project or is our project? If wmaker is > their project, he can do everyting. If is our project, perhaps he > should hear us. Please, choose, is our or yours? That is important for > me. And the previous question was not replied, could you accept that > patch? I don't agree with your perspective. Do you want a curator or an integrator? A curator has opinions and sensibilities based on experience and talent, and a big part of the job is applying those. You trust the curator to do the Right Thing. An integrator is just someone who has the task of orderly incorporating changes and somehow making it work well enough to gain acceptance. If you want a curator, delegate the responsibility and either abide by the decisions they make, or find a new curator. If you want an integrator, pay them a salary. I do not see the trust misplaced. I *do* have a problem of talking about a curator being _forced_ (in theory) to accept a change they object to, rather than having a discussion that teases out and clarifies the actual issues, so they can be addressed in the open, and either the curator's mind changed by discussion, or the proposed change modified to address the objections. Committees make poor engineering decisions. -- SJS -- To unsubscribe, send mail to wmaker-dev-unsubscr...@lists.windowmaker.org.
