4 ms·

I understand that this might be specific to me, but I'd actually rather have someone tell me: "You're useless. For god's sake stop contributing until you get y

I understand that this might be specific to me, but I'd actually rather have someone tell me: "You're useless. For god's sake stop contributing until you get your sht together" rather than: “your skills don't seem to be quite at the level where we can accept your contributions; in particular, you don't understand X, Y, and Z very well and it shows in your code. I'm short on time, but if someone else has time to explain why that would be great.” I don't have a good explanation why that is. Maybe it's because I feel some condescension in the second while the first one is just plain raw "I don't have time to deal with your sht. You're an adult: fix it and come back later".

gknoy 13y ago
The more polite response gives at least some shred of a foothold on which the recipient can find out how to better themselves. "Fix your shit" does't explain how your shit is lacking, or the major things you don't understand. e.g., if I were to post some patch that used bubble sort, one could easily say "This is total crap. Don't come back until you fix your shit". If I were a noob, that wouldn't help me contribute better. If one were to rather say something like, "This is very inefficient," and then mention something about O notation or efficient sorting, and point out that this is a performance-critical section of the code, I might be more likely to actually take the time to learn enough to be able to fix it. Consider also the effect on newer programmers. They start, they commit crap, you tell them to die in a fire. They get better, and years later decide that they're not going to bother fixing bugs in your project because "that guy's a jerk". Constructive criticism, or at the very least politeness, helps prevent poisoning the well of talent that is interested in working with you.
- jlgreco 13y ago
  Programming, more so than many other fields, has a strong tradition of self-improvement and education. Tell any "noob" that they are an idiot for using bubble sort and they should have more than enough resources and know-how to figure out why they are wrong and correct it themselves. (Either they independently invented it themselves, in which case they are at least moderately clever and you just told them the name of what they created (what they should google), or they googled it in the first place so they should be able to google it again). Is it best, in a controlled environment with both low signal and noise, to take your time and explain things in detail in a "civil" manner? Sure. That works great in (small) classrooms for example. In a public kernel dev mailing lists though? No, a fuck off should suffice. If the recipient of the fuck off wants more, they can seek it in a more appropriate place. IRC is somewhere in the middle; usually civility or ignoring the person works best. This is particularly true since most IRC channels are fairly explicitly there for people who want to give or receive advice. I've only told someone who was seeking help from me on IRC to fuck off once, and that is when he demanded I talk with him on skype instead of IRC, and pressed the issue. Edit: I suppose I shouldn't expect a civil response if I explain in a civil manner why I don't think civility is always necessary. Hoisted by my own petard perhaps?
pseut 13y ago
Maybe, if they're right. If they're wrong, the second one is a lot better than the first.
aninhumer 13y ago
If that's true I think it is pretty specific to you, and honestly I doubt that you would actually prefer it in practice. It's easy to say you can take the abuse when it's clearly framed as an alternative to a polite comment, not so easy when you get piled on by a bunch of obnoxious commenters. Emotional maturity makes it easier to deal with, but it still sucks. Also you can still give a franker response without being blatantly derisive: "Thanks for your contributions, but I'm afraid they do not meet the quality standards of our project. I think you might need to learn a little more about X, Y and Z, before you do any more work."
6d0debc071 13y ago
Hmm, the first one would basically stop me contributing at all. You go in feeling all good and like you're going to lend a hand, presumably to people you respect if you want to work with them in the first place, and then they turn around and basically tell you they think you're a retard. It'd hurt pretty bad. I'd be interested, would you find something like this to be as patronising: "Thanks for your submission. Could you improve X, Y, Z and resubmit?"
- shadowfiend 13y ago
  Definitely less patronizing, yes. I like it very much, but that works better in a situation where you can see that they already understand X/Y/Z reasonably well. I guess I was trying to present the worst possible situation, where it seems like they are stumbling around with concepts that seem to be over their heads, and then I think the risk of being patronizing is worth the value of being as honest as possible.
emmett 13y ago
A good response here would probably look more like, "While we appreciate you making the effort to contribute to the project, the patches you've been submitting don't meet our quality standards because [reason - poorly structured? no tests? full of bugs?]. I recommend you [take an online learning course, write tests, read our coding guidelines] before you submit any more patches."

gknoy 13y ago

The more polite response gives at least some shred of a foothold on which the recipient can find out how to better themselves. "Fix your shit" does't explain how your shit is lacking, or the major things you don't understand. e.g., if I were to post some patch that used bubble sort, one could easily say "This is total crap. Don't come back until you fix your shit". If I were a noob, that wouldn't help me contribute better. If one were to rather say something like, "This is very inefficient," and then mention something about O notation or efficient sorting, and point out that this is a performance-critical section of the code, I might be more likely to actually take the time to learn enough to be able to fix it. Consider also the effect on newer programmers. They start, they commit crap, you tell them to die in a fire. They get better, and years later decide that they're not going to bother fixing bugs in your project because "that guy's a jerk". Constructive criticism, or at the very least politeness, helps prevent poisoning the well of talent that is interested in working with you.

jlgreco 13y ago
Programming, more so than many other fields, has a strong tradition of self-improvement and education. Tell any "noob" that they are an idiot for using bubble sort and they should have more than enough resources and know-how to figure out why they are wrong and correct it themselves. (Either they independently invented it themselves, in which case they are at least moderately clever and you just told them the name of what they created (what they should google), or they googled it in the first place so they should be able to google it again). Is it best, in a controlled environment with both low signal and noise, to take your time and explain things in detail in a "civil" manner? Sure. That works great in (small) classrooms for example. In a public kernel dev mailing lists though? No, a fuck off should suffice. If the recipient of the fuck off wants more, they can seek it in a more appropriate place. IRC is somewhere in the middle; usually civility or ignoring the person works best. This is particularly true since most IRC channels are fairly explicitly there for people who want to give or receive advice. I've only told someone who was seeking help from me on IRC to fuck off once, and that is when he demanded I talk with him on skype instead of IRC, and pressed the issue. Edit: I suppose I shouldn't expect a civil response if I explain in a civil manner why I don't think civility is always necessary. Hoisted by my own petard perhaps?

pseut 13y ago

Maybe, if they're right. If they're wrong, the second one is a lot better than the first.

aninhumer 13y ago

If that's true I think it is pretty specific to you, and honestly I doubt that you would actually prefer it in practice. It's easy to say you can take the abuse when it's clearly framed as an alternative to a polite comment, not so easy when you get piled on by a bunch of obnoxious commenters. Emotional maturity makes it easier to deal with, but it still sucks. Also you can still give a franker response without being blatantly derisive: "Thanks for your contributions, but I'm afraid they do not meet the quality standards of our project. I think you might need to learn a little more about X, Y and Z, before you do any more work."

6d0debc071 13y ago

Hmm, the first one would basically stop me contributing at all. You go in feeling all good and like you're going to lend a hand, presumably to people you respect if you want to work with them in the first place, and then they turn around and basically tell you they think you're a retard. It'd hurt pretty bad. I'd be interested, would you find something like this to be as patronising: "Thanks for your submission. Could you improve X, Y, Z and resubmit?"

shadowfiend 13y ago
Definitely less patronizing, yes. I like it very much, but that works better in a situation where you can see that they already understand X/Y/Z reasonably well. I guess I was trying to present the worst possible situation, where it seems like they are stumbling around with concepts that seem to be over their heads, and then I think the risk of being patronizing is worth the value of being as honest as possible.

emmett 13y ago

A good response here would probably look more like, "While we appreciate you making the effort to contribute to the project, the patches you've been submitting don't meet our quality standards because [reason - poorly structured? no tests? full of bugs?]. I recommend you [take an online learning course, write tests, read our coding guidelines] before you submit any more patches."