So the new welcome wagon rolled in and brought us more examples of comments considered "not welcoming". While I did participate in the survey and found a few comments that I found not welcoming, I really wonder about the examples given this time.
“Why do you want to do this? You have conflate...
More precisely - it feels like getting the user to refine the question (and get the answer she needs, rather than getting the answer she thinks she needs) would be more productive than a quick close.
or shog can rust proof the undercarriage and delete it ;p
@ArtOfCode Normally you see a "closed as off-topic" in the revision history, but in the case of meta.stackexchange.com/questions/319470/…, you don't, and the stub doesn't show with "[migrated]" in the title.
@balpha could you please fix notarealquestion.appspot.com? I used to enjoy that game, but now it gives an error message that Python 2.5 is not supported.
95% of comments being fine means that 1 in 20 are not, and that's quite a large amount of comments. And when a single unwelcoming comment has the potential to turn away a potential long-term contributor, well, it should be reduced. — Sonic the Inclusive Hedgehog2 hours ago
@forest I'll tell you something from my teaching experience. If someone is already used to hearing negative feedback for many reasons, and someone tries to give them constructive criticism, they're not going to listen.
@SonictheInclusiveHedgehog Unfortunately, just because someone might take constructive criticism badly does not mean that it is not extremely important to give.
I'm also pretty sure those comments were pulled from live comments, which means many of the really bad ones had already been deleted, which, I'd think would skew the results toward there being more fine comments and fewer rude ones.
@Catija Then they're using examples of good comments as bad comments, which is irresponsible. If problematic comments were deleted and can't be retrieved, just don't publish acceptable comments and label them as problematic.
I'm not asserting that those comments are problematic or not. I'm saying, just because 95+% of comments are fine doesn't mean that not-fine comments aren't a problem.
@SonictheInclusiveHedgehog But it very possibly does. When people are intentionally told to look for things out of context that could possibly be seen as abusive, they'll find far more and far worse things than the actual recipient of the "abusive" comments, when put in context.
@forest Even if you factor that out and say that 1% of comments are not fine, that's still a lot of comments, once you consider the volume of comments.
Sure. So delete the abusive ones. But when it's something like 0.05% of comments (I imagine nearly all of which are quickly deleted), is it really that big a deal?
We're probably seeing a lot more of specific stuff than (some) employees, and we're pretty much the ones who need to decide on acceptability on the ground, unless someone kicks it up or another mod has a polite disagreement.
Sure. What I'm critical of is the fact that so very few people think there's any problem at all wrt abusive comments, but the employees (who don't really use the site to the same level) do. And naturally, the employees, the only ones with real power, decide to push out the welcome wagon and freak out about totally normal comments.
@SonictheInclusiveHedgehog I know you are. Overall, judging by the actual statistics (as well as anecdotal evidence from Meta and the like), you are a minority.
@JourneymanGeek You're right, I was talking about the implementations.
@forest Imagine I'm a new user and ask a question about my code on SO, "this code has a problem, can you please help me find it?" Then I get some comments (paraphrasing from real ones I've seen on Triage) to the effect of "have you tried debugging your code?" To an experienced SO user they may be commenting about SO's policy and whatnot, but to a new user, it comes across as snark.
It's not like there are all these unwritten rules that we shove on new users. When they come and ask something stupid, people will tell them their question can't be answered. If they're upset that they aren't given the answer on a silver platter, that's not our fault. It's their's.
@forest In one famous micro-study, it was shown that new users often aren't even aware of an "FAQ" or "help center". Only few users saw the right sidebar, and most of them only saw it after composing their question, and they thought, "they won't mind, right?"
@forest Let me tell you something. For many years, I had the exact same mentality as you. In fact, I still sort of do when it comes to social communication.
@SonictheInclusiveHedgehog That's a good reason to ask the guys at UX.SE how to avoid the banner blindness and ensure that people actually see the FAQ.
@forest No, the problem is that we have all these written rules that are confusing and unclear and not easy to find... and then we penalize people for not knowing them or where to find them. We do a bad job of helping them succeed, and that's why we're building the AaQ wizard but that doesn't make snarkiness OK.
@forest As does my literal parser. But it's wrong to think that said feeling came entirely from the authors themselves; in many cases, they effect such a reaction.
@SonictheInclusiveHedgehog Well then that's a valid response. It's totally fine to clarify that the question is abstract and not intended for a real system.
Not everyone is as resilient as you are. Assuming that everyone can do something because you did it is... not the way things really are. Right now we're optimizing for the people who are willing to put up with what we dish out... not people who actually know what they're doing. Lots of very talented programmers won't participate in this network because of the culture.
The ones that do, if the asker takes the question as an insult then they're not benefiting from it. Nor are we, since the reaction we're getting doesn't accurately reflect intent.
All I'm seeing here is problems that have solutions, but I fail to see how being "more welcoming" is a solution in cases where they won't read comments anyway.
"more welcoming" is at best a solution to the age-old problem of "the forum mentality" - the idea that everything needs a response and the only way to keep new folks running amok is to hammer the rules into their heads as quickly as possible
that worked great when it was a few dozen people in a newsgroup. It doesn't scale though
There are a LOT more people using SO than there are people who can ask those constructive questions
And the goal isn't to convert the world into SO users
it's to build that ol' library of useful answers
folks asking clarifying questions in comments ain't writing useful answers
they're pissing away their time on folks who probably won't respond
@forest maybe, but that's only a problem if you consider every question to be some precious little gem the loss of which will forever leave the world broken
@forest ideally, folks include enough information for you to read between the lines. So you answer the question they should've asked, explaining your rationale along with that answer.
If they don't... Well, you answer the question asked. Or don't. Either way, someone has that question and your answer does some good or the question disappears and no one has wasted their time on it.
@forest or what about Super User where the wrong a command could destroy all the data on your backups, or SO where you could introduce bugs into some widely-used app?
put the extra effort into explaining the context of your answer. If the folks using it don't read that, they sure as hell ain't gonna read the comments either.
If someone asks how to securely delete data on an SSD, without knowing their threat model, it would be completely impossible to answer it without several pages.
@Shog9 So all the questions I've answered that I could only answer by writing a comment saying "can you describe your threat model?" should not have been answered? VTC often scares newbies away...
But politely telling them that it's unanswerable, and that they need to specify what they mean by deletion and who their adversary is is vital.
Again, if you really care about the question and want to answer it, do whatever you need to - all this goes out the window then. Ask a question. Have a conversation. Write a novel in your answer. Whatever works.
You're advocating against something that every Meta post ever and even the FAQ say is absolutely vital to do, and which anecdotally has not only saved numerous questions, but lead to irritated OPs turning thankful.
But... That's a question you really care about. That's something you want to spend time on. I'm not really worried folks who are yearning to write an in-depth answer on secure data deletion are gonna half-ass these questions.
Sometimes the reason I ask for clarification on more subjective/non-technical sites is, because they could literally find the answer on the first page if they search the title of the question in the first place. Indeed, the question is on-topic, but badly research that would be downvoted to abyss...
The examples you complained about earlier... the ones from the blog post... D'you think the folks writing those were so terse because they were simultaneously composing a mighty work of great depth and insight with an eye toward posting it as an answer to the question (once clarification was provided)?
@forest I'm coming from the perspective of... It seems like an awful lot of folks are doing stuff that makes them unhappy. And makes other people unhappy. And accomplishes nothing. And so... Why bother?
Oh sure, if they don't reply to the comment, yeah skip it.
But you are saying that we shouldn't try and that, if OP doesn't always know how the site works, we should leave them to rot instead of helpfully instructing them.
@forest If teaching folks how the site works makes you happy, knock yourself out. If it becomes drudgery, to the point where you no longer care about learning how to do it more effectively... Stop.
Leaving cookie-cutter comments doesn't exactly sound fun to me. I've tried it. It wasn't. I stopped.
It seemed to work for me. Maybe this is just a difference between the sites I use and Stack Overflow. On Information Security and Cryptography, people very often respond to comments asking them for clarification. When they don't, we VTC.
This is why it saddens me to see folks spending so much time on bad questions. If the question isn't motivating anyone to write... Just get rid of it. Abandon it and let it be deleted, or perhaps motivate someone else.
A question that is intrinsically bad due to no effort going into writing it, or a question which is the wrong question, an XY problem, or just not correct English?
All this forum-style interrogation in the comments, closing, reopening, editing... For a question no one cares about... What good is it, other than to sap the remaining life of so many people to no result?
@forest if a no-effort question prompts a useful answer, it is every bit as good as a question that someone labored over.
Better, perhaps, since the asker didn't waste unnecessary time on it
@andmyself sure, and this does result in some arguments. However... There are far more questions than anyone could answer, even if everyone answered until more than satiated.
Which is the real tragedy on SO as well: folks answer questions that are of little use to anyone, then wonder why they get no love for doing so - even from those that they help. While too many questions which would be useful to many get lost in the noise.
@forest The Russian Mathematics site, which was a former hashcode.ru site, explicitly objected to transitioning to the SE 2.0 engine because they allowed homework questions and didn't want to change their policy. So math.hashcode.ru is technically under SE ownership, but operates entirely separately from the network.
@Shog9 I feel like I might have asked this at some time before, but in analyzing "site health" and emergent community behaviors across the Network, is there any sort of conclusion to be drawn that (pulling numbers semi-randomly, here) 20-100 qpd is a sweet spot for the architecture; sites getting >300 qpd might benefit from breaking apart?
@andmyself Also, an actual claim I've seen to justify that something isn't a homework question: "this was a test question" or "this was an online problem"
Or does it not seem like qpd is the reliable metric of when a site's going to start to fall off the sort of cliff you're describing SO sitting at the bottom of?
@Shog9 And what drives that skill level (on both sides)--are you talking about the underlying content, or the community expectations/reception of posts?
That's why Medical Sciences is trying to pivot into more of a professional-to-professional site: that's a model that actually works. It's what SO tries to be.
So you see where splitting a topic can become very problematic: if it creates this separation from the folks learning to be experts from the folks who don't even realize how much they don't know... The site can't be very healthy.
Imagine if all the folks who ask questions on SO along the lines of, "I want to build a site like Facebook, but for cats" had their own site. The folks who've never written a line of code, but assume it's pretty easy and think they have this great idea for an app...
@Shog9 It might be fun to take all the idiots who ask questions like that, and put them in one big site where the blind can lead the blind, and just watch.
@Shog9 Is that a problem of androidenthusiasts having its own site, or that it peels off some of the SO population while leaving another segment there?
"professional and enthusiast" was that sweet spot where folks could, with some effort, transition from the asker to the answerer, from the supplicant to the provider.
This post, has an accepted answer that has a single issue that I wanted to help note/fix.
"Then Blindman67 suggested the naive approach with this solution:
"function findCommon(arr)..."
This implementation has a small error when given an array with a single high value (e.g. arr = [1,2,33,2,1,1...
The people who really care about improving will improve. The ones who don't will leave (and not click on ads, so obviously SE staff does not want them to leave).
I mean (again: spitballing/gedanken here) what if android had been ruled off-topic on SO and all those Q/A migrated to android.se when it was created? It just seems naively to me that if Network sites that effectively duplicate some subset of SO suffer from the fracturing of their potential userbases and SO is too big... those two problems have a common solution?
That's the real trick behind all this welcoming stuff. To find a way to say that without giving the wrong impression. To encourage folks to learn, not just participate. To grow and replace us all after we're dead and gone.
@forest Saying nothing does neither. That's... sort of the point, if I'm understanding. If you don't comment, you're not going to be unwelcoming... and you're also not being a yes man, whatever that means to you.
@forest I don't. I used SO enough to submit the edits I needed to earn 24 rep to get the downvote privilege. That's it. None of the sites I use are even close to large enough... and on those sites, questions are precious... users spend time trying to cultivate them because there are so few, there's nothing else to do... it's a completely different thing... but, by the same token, there are so few questions, you don't get burned out trying to improve them all.
So the new welcome wagon rolled in and brought us more examples of comments considered "not welcoming". While I did participate in the survey and found a few comments that I found not welcoming, I really wonder about the examples given this time.
“Why do you want to do this? You have conflate...
This post, has an accepted answer that has a single issue that I wanted to help note/fix.
"Then Blindman67 suggested the naive approach with this solution:
"function findCommon(arr)..."
This implementation has a small error when given an array with a single high value (e.g. arr = [1,2,33,2,1,1...
