On 24/07/18 20:49, Vincent Povirk wrote:
My take is that code can be "good enough" to go in, but also not be what the maintainer wants. You're not responsible for predicting that, nor will you generally have the knowledge and experience they have. However, being able to take feedback and adjust your approach is highly valued here.
I guess I have a hard time seeing where the difference falls between the two. Besides, it seemed that Alexandre wanted to make it clear that a sign-off means a patch is good enough in every respect—legal, stylistic, correctness.
I know I've made my share of dumb mistakes that ended up in submitted patches, and occasionally in Wine. You're never going to be 100% certain, just do your best.
I would like to suggest that maybe we as a community should stop talking about "Julliard rank". While I'm sure he considers your history when reviewing patches, it sounds like the way that's discussed creates undue anxiety. And it's overemphasized IMO, the quality of your work is what ultimately matters. (I ALSO would like to continue to suggest that we put in systems to make sure everyone's patches get reviewed.)
I guess "Julliard Rank" is just a dumb code word for "Alexandre's level of trust", and it's observable enough that a name had to be invented for it. And how much a maintainer trusts you affects how much patience they're willing to give to reviewing your patches (or answering your questions, I guess). Which is something that I can't help but feel concerned about. But I guess that concern just boils down to "I submit poor-quality patches and I'm concerned about receiving deferential treatement", which is not really something that can or should be helped.