The thing I think worth remembering is that questions can be improved. Some of the worst questions I've seen have actually gotten improved and made really useful in the end. But you need to give the asker the chance to improve, even if thousands of users before them haven't.
Yeah, I think one of the issues Stack Overflow has is that a lot of long-term users have gotten burnt out on giving people such chances due to the high probability that trying to point out what's wrong with someone's question will result in you being told to go fuck yourself
That's how I felt on M&TV with ID questions... dozens of them, daily, horrible quality, lacking all detail... I had an auto comment for a while and then I just got tired of it and started downvoting and close voting... but I made the choice not to comment rather than post snarky comments... I let someone else comment because I knew I wasn't able to do the asker right.
I still do usually point out the problems in cases where someone seems to at least be vaguely capable of communicating what they want. But questions like the couple I linked to above? I can tell they're hopeless, tbh.
It's a struggle and I understand it 100%. You love these sites - whichever site it is - and you want the best for it and you're tired of the day-to-day scut work.
@MarkAmery So, are the answers here actually wrong or are they downvoted because they're answers to bad questions?
A dump of 200 lines of code accompanied by "It doesn't work, plz fix" is a bad question but at least one that somebody sufficiently motivated could puzzle out and solve... but there are plenty of questions that don't even reach that threshold
What i hate is when i know for a fact that the code in question will throw an error that explains what is wrong, and the user doesn't provide it and/or acts like no error is occuring.
but people are more than happy to provide an answer pointing out the correct syntax
@Catija the 200 line code dump? Yes, definitely off-topic. But at least I feel it may be worthwhile to explain that, and why, because it's an asker who is capable of forming a coherent question, and may understand what I'm telling them.
@KevinB Sometimes it would seem like the first comment there should be something along the lines of "Are you getting any errors when you run this? What are they?"
If SO (corporate) is looking into the future for the Network, they're going to be assuming that a lot of these new users showing up today and yesterday and last week... they're going to be the avid users in two, three years... but only if they stick around.
Every person who gets scared off could be the next Skeet.
... though, realistically, there's probably a heavy anti-correlation between being likely to be scared off by how people treat you here and being likely to be the next Skeet
Not that I'm in the "deliberate dickery is good for quality control" camp, but I at least concede that much to them
Having the knowledge and having the tenacity to put up with crap aren't always hand in hand, no... but why should someone who has the knowledge need the tenacity?
I guess the point I meant is more that the "crap" is a lot less likely to come your way - or at least there'll be less of it coming your way - if you have the knowledge
Yeah, regardless of what the issue is or what side of it you're on, you just don't have the emotional robustness to accept my insightful criticism! is a complaint you can always throw at your most stubborn opponents :P
I'm cautious about going too far up into that sort of meta-analysis, because, like, there's sound criticism of questions and there's terrible wrongheaded criticism of questions (and likewise of comment tone, to take @Catija's point). The actual concrete details matter.
Putting in the work to be cautious with your tone is, to some extent, the same as asking users to put in the work to write good questions... If you're going to do a half-assed job of either, go do something else.
And, some might say, being cautious with your tone should be simpler since you don't have to find help documentation or use SE's search features.
To riff off both the much-disputed point I've been making about different politeness standards and the convo I was just having with Shog about contradictory question standards, another way in which they're the same is that different groups of users have wildly different ideas of what getting it right looks like, and no matter what you do, somebody may shout at you for it
Well, things aren't easy... right? The important thing is how you respond to someone shouting at you. We want people to be willing to step back and rephrase... or step away entirely. We don't want people who are going to aggravate the issue and get into a screaming match...
Sure. "step away entirely" is almost always the proper solution once you've made the point you want to and somebody is non-constructively pissed at you about it
@MarkAmery I'm never going to wrap an OP in a warm blanket and rock them gently just so I can deliver, "The code you've provided doesn't produce the error you've described," without fear of their negative reaction... but comments exactly like that are often perceived as unwelcoming. And like you said, there's no standard. Everyone has a different reaction. Some people are fine with getting down to brass tacks. Others want some hand-holding. Good luck standardizing. :/
@Catija so, a couple of things strike me after reading that
One is that I personally think those sort of comments are unhelpful because they inspire doubt in the reader about precisely why they're getting linked to a big page of guidance
@Catija My second point, regarding tone, is that I think a bunch of users (quite possibly including you) have a blind spot about how layering on signals of politeness in a post (things like "Welcome!" and saying "please" and "thank you") can make the tone sound more formal, and that in turn can make it easier to read in some sort of adversarial subtext
If my boss says "Yo Mark, swing by my office for a sec" then I will be pretty sure that everything is fine. If he says "Dear Mark, please could you stop by my office when it's convenient for you. Thank you." then I'm gonna worry I'm getting fired.
@MarkAmery so... I don't want to speak for my colleagues... But my take on this is that they're essentially trying to arrive at a definition for problems that we can solve by working from an unassailable foundation and showing their work at each step of the process.
IOW, they're very likely going to end up with some conclusions that... Most folks will roll their eyes at as being completely obvious and hardly worth the effort. BUT. It won't be purely anecdotal; they'll have an understanding of what leads folks to feel unwelcomed, or to behave in an unwelcoming manner, that's more than just speculation and strawmen.
@MarkAmery good stuff. I ended up sitting through two. Can't really talk about either, but... Heard a lot of the same stuff I've been hearing in this chatroom.
@Shog9 I think I'm cautious about SE data analysis because I've tended to be frustrated by it spectacularly missing the point in the past. Two cases come to mind:
1. The +1/-1 title filter. (Sorry, I know I always bring this issue up. Also, I'm suddenly doubting myself about whether this was actually analysis SE did or some community member who posted an answer.) As I recall, y'all did some classifying of comments with a +1 or -1 in front and found that they tended to be non-constructive. Except y'all did it on a sample where +1s were much more common, and looking over the same raw data I found that -1 comments were disproportionately likely to be useful.
2. The "problem" title filter. The community objected primarily to it on the basis that it would prevent the formal names of named Problems from being included in titles. Then y'all did an analysis of title edits that had resulted from the filter which didn't include looking for cases where formal names of problems had been removed, and declared that the filter was doing good.
I'm generally in favor of trying to quantify stuff, but there's a track record of the quantitative questions to ask not being guided by the community's (potentially flawed!) common sense, and so ultimately just seeming to miss the point and not make the case that the staff want it to make
@MarkAmery jut to be clear, the +/-1 thing was all me, based entirely on my growing irritation with folks who emailed me about getting downvoted. The data analysis came afterwards, and resulted in me adjusting the filter to allow more comments to be posted.
@MarkAmery This one I don't remember clearly, but for all the complaints I'd have little trouble believing that most titles containing "problem" are... problematic.
I say this because I've since been asked to implement similar filters on several other sites
IIRC, the filters in current use don't disallow "problem" entirely, but rather block titles ending with it and/or very short titles with it.
@Shog9 Me too. But 1) there's a tradeoff; is it worth slightly damaging a few bits of good content in order to maybe slightly improve lots of bad content? It's not obvious to me. And 2) whether or not the filter was a good idea, a quantitative analysis of the impact of the filter that by design didn't even look at the kind of impact that most concerned the complainers was never going to satisfy them.
We could probably revisit it. But... Maybe we should try & get some other things implemented first.
Tim had an idea once for a "question title goodness meter" that we could display for askers; something like those password complexity meters you see now and then, but based on heuristics known to match up with useful titles (length, distinct keywords, etc)
That'd be a much more elegant solution to the... heh... problem
@TimPost sadly no. I have had no reason to upgrade from my classic blue black widow. Some folks have had terrible luck with razers and you need their config software (which is cloud connected) for many neat things.
Been happy with more recent purchases so build quality hasn't been an issue.
Razer's using their own switches so no opinion there but I have been happy with the general layout on those things