3 ms·
I'm very glad to see this idea turning into a community mechanism. HN has some great minds, and I love seeing the solutions they come up with to problems. It's
by ignostic 12y ago
I'm very glad to see this idea turning into a community mechanism. HN has some great minds, and I love seeing the solutions they come up with to problems. It's a very technical, analytical, and rational community.
Unfortunately, some of the qualities that make a good technical problem solver can also lead to nitpikcing. It feels like trying to find the bugs in an argument. That can be useful, but only when it ties to the core idea rather than some tangential point. Top comments often sound like, "Well, technically, you're wrong on this one unimportant fact you mentioned briefly."
I'm glad HN is taking steps to ensure that feedback is constructive.
- Kalium 12y agoBear in mind that what is and isn't constructive is not an objective measurement. It's a subjective value judgment. To some, finding bugs in someone's argument is a constructive thing - you are helping them by fixing their argument and showing them how to argue better in the future. Being better able to argue means an increased ability to communicate effectively with others. So my question becomes thus: how do you define constructive? I, personally, have run across multiple instances of one person's constructive and actionable feedback being another's gratuitously negative useless nitpicky feedback.
- panic 12y agoMaybe a good rule is: the smaller the pedantic, nitpicky detail you address in your feedback, the more you should include other details as well. Instead of "this is wrong because detail X" say something like "this handles detail Y well and I think detail Z is clever, but detail X is wrong".
- Kalium 12y agoI think your ideas are good and your intentions are laudible, but I question what purpose is served in asking people to wade through excessive verbiage in order to arrive at what could otherwise be concise and direct criticism.
- panic 12y agoThe idea is to give the reader a more complete picture. Often we skip over what we think is good about something and focus on the problems. I'm not saying you should come up with feel-good bullshit to make something fundamentally flawed seem like a good idea. I'm just saying that, when something is "mostly good but with one problem" we can learn just as much from why it's "mostly good" as we can from what the problem is.
- Kalium 12y agoI was attempting to give a demonstration of where notions of completeness-of-feedback such as this can lead by wrapping what could have been a concise point in needless verbose complexity. I am reminded of "How To Make Friends And Influence People", where it is advised that any criticism be couched in complements because people are easily manipulated by the positive emotional rush of a complement. I'd like to think that such wrapping isn't needed on HN, and people are able to separate ideas or code from themselves for the purpose of accepting useful feedback. It is possible that this assumption is incorrect and should be abandoned, as people often have difficulty separating the two. I find that I do not agree that positive feedback is as useful as negative feedback. Being told you are right certainly feels much better and is much kinder to the ego we all seek to stroke, but it is in being told you are wrong where is opportunity for growth lies. If we're clever, we can even manage to learn from the mistakes of others and avoid them ourselves. I am uncertain that forcing a "more complete" consideration of issues that are in the main not relevant is likely to improve the level of discourse or utility of feedback. I'm going to avoid the issue of one person's fatal flaw being another's useless nitpick - it'll come up next time someone tries to be clever with crypto anyway.