38

The question: Authenticating a user via SMS elicited a very interesting comment from the OP after a very new user (<20 days) suggested the question was off-topic.

I don't mean to kick a hornet's nest, but the incessant arguing on StackExchange about what is or isn't an acceptable post is mind-boggling to me. I get that we don't want people obviously trolling for opinions and instigating flame wars, but should we really be getting up in arms about a serious, productive question where a real person is getting real help about a real issue? Isn't that the whole point of this site? The rules are there to help protect that, not to get in the way.. – John Chrysostom

Whether or not he's right about this specific case, the fact that the network is eliciting this kind of rant from contributing members is not good. You see it several times a week where a new user posts a low-quality first-post, gets torn apart for it, and never comes back. Is that really how we want this community run?

  1. Is it appropriate to refine our Off-Topic policy so that it catches fewer legitimate questions and is more friendly to first-time posters?

  2. Is it appropriate to flag / remove overly harsh comments on first-time posts and replace them with something softer / more encouraging?

  • 7
    .... yeah I see your point, 2/3 questions closed, and a negative overall question score. ... so some crybabying is happening. – Mike Ounsworth Mar 30 '16 at 19:19
  • 5
    just to say that (unsurprisingly to any DMZ denizens) I agree with you that SE is too aggressive about this, bad first time user experience will lead to them not coming back and the point of the site (AFAIK) is to answer people's questions... – Rоry McCune Mar 30 '16 at 19:20
  • 1
    I honestly just assumed it was someone trying to get their "Post 10 Comments" badge. Glad OP was invested enough in getting an answer to post that comment in the first place. – WorseDoughnut Mar 31 '16 at 14:00
  • 4
    I've just come back after a period of silence due to overly harsh judgements of my legitimate attempts to get my work problems solved. I hope the fact that I'm seeing this discussion here means the community is getting wise to the issue. I'm often pretty frustrated BEFORE I ask a question here and NEED the answer to get unstuck! Being lectured isn't helpful, guys. – Sinthia V Apr 11 '16 at 1:46
  • // , @RоryMcCune, the point of answering people's questions is better served by the perspective of curation of a wiki than by the perspective of supporting written questions. – Nathan Basanese Jan 9 at 19:41
22

This questions did strike me as especially hard over on SO, where new and/or low rep users regulary get downvoted and closed very fast.

Over here on Sec.SE, because there is way less traffic, the problem is not as maddening as on SO, but still a problem we as a community should try and take care of.

In fact, on SO there is a "SOCVR" team in an attempt to rather improve the question and be helpful to new users, trying to educate them as to how they may get better answers faster.

(Side note: SOCVR do more than just that, but let's focus on the "help OP"-part right now)

Nontheless, it might be a good approach to

  • leave useful comments when downvoting and/or closevoting (as does schroeder regularly, there is a user script for auto-comments, which makes that easier for recurring problems),
  • have a chat room with experienced users who try to explain the problems with questions to OPs in a more direct way, helping to shape the questions in form.

    This way, new users may still get ranted away (because, hey, it's still the internet), but there are people to help them understand the problems and tell them to not be giving up on their first shot.

As I was a room member on SOCVR and I think the idea is pretty good, I just founded and linked the proposed chat room and am happy to have more people there.

While the team on SO has a broader approach to moderation, I'll quote here some parts of their internal rules that may apply here too:

  • Work with the OP to get their post into shape; most content has some value.
  • De-escalate in case of disagreement.
  • A post is to be actively handled by only one member of the room. (We don't need 4 members all leaving witty statements in the comments or in chat.)
  • Moderate the post, not the user. (keep the discussion on the merits of the post, not on behavior of the user)

If this catches any interest by anyone, I suggest the chat can discuss how things go from there.

  • 7
    +1 for Work with the OP to get their post into shape. It happens somewhat often that a post is closed without any comment, even though a slight reformulation or small additional details could make it a good question. Giving users - especially new users - feedback to improve their question should happen a lot more often when voting to close. – tim Mar 31 '16 at 9:02
  • Exactly my point, @tim :) – Tobi Nary Mar 31 '16 at 9:07
  • This is what we need more of at Sec.SE – Brad Bouchard Apr 13 '16 at 16:31
  • 1
    I used to love contributing to the community by editing ESL posts for grammar until SO started quizzing me for the privilege of donating my time! – Sinthia V Apr 21 '16 at 15:46
21

I agree. I think this is mostly a problem with close votes on questions, but I just noticed this deleted answer: x

The author is the lead of the OWASP Zap Project - one of the more significant open source security tools. He previously answered a few questions on Security Stack Exchange - but has been silent since.

A good example of how this behaviour actively discourages quality contributors.

  • 6
    This legitimately saddens me. I am fairly new user and the harsh reactions like this definitely cause me to both ask and answer less than I otherwise would. – Niels2000 Apr 6 '16 at 8:27
  • 9
    You should evaluate the quality of the answer, not the user who posted it. This person may be a fantastic contributor to open source, but if his answers are bad they should not be kept because of his name. – A.L Apr 6 '16 at 17:34
  • 5
    @A.L - the quality of this answer is fine. You might be thinking it's a bit short, a link only answer. That's all the original poster needs. To post the right link takes a lot of knowledge, which many on here seem to not appreciate at all. – paj28 Apr 7 '16 at 7:05
  • 1
    14 minutes after posting, there was a request for more info. 50 minutes after that, it was converted to a comment. It's not clear how quickly it was downvoted. OTOH, he's posted a very similar answer, but included info from the site before: security.stackexchange.com/questions/32/… I couldn't find the link to the question he was answering... did he really answer the question? – mgjk Apr 7 '16 at 12:15
  • 1
    @mgjk - this is the question – paj28 Apr 7 '16 at 12:20
  • Seems like a very succinct and authoritative answer to the question: "Can the OWASP ZAP check XSS for REST API"... – mgjk Apr 7 '16 at 16:08
  • 4
    @paj28 I think you are reading too much into his silence. He answered 3 times in 2015, and 4 times in 2014. The fact that he's been silent since Dec, and since he doesn't appear to be active in the beginning of the year each year since 2013 might have more bearing. – schroeder Apr 9 '16 at 4:11
  • 4
    @paj28 Also, I do not see a problem with the conversion of this answer to a comment (it was not deleted, but converted). He asks a question in 7.5 words. This is a normal thing to convert to a comment. – schroeder Apr 9 '16 at 4:13
  • 5
    This is not a good example. That's not an answer to the question. The question is "Can the OWASP ZAP detect XSS in REST API's?" Saying "Have you tried...?" is not an answer -- that's a request for clarification, and it belongs in a comment, not an answer. Converting that to a comment was the right outcome. – D.W. Apr 9 '16 at 21:20
10

This is a huge topic, and a legitimate one.

The question at the heart of "topic-ness" is the quality of posts. Too much noise in the questions means that the high contributing people we could be attracting will leave because "there is nothing for them here".

This model was highlighted during the game that StackExchange ran for each site. There is a balance to be struck between culling low-quality, off-topic posts, and maintaining a high enough quality of the site to attract high-quality community members.

It IS a balance, and that balance is part of your question. Maintaining that balance is tricky and messy and is performed by corrections over time.

Perhaps one underlying factor to your question is being nicer to new folks. THAT is something we all can do better, even if we end up closing their posts.

  • Credit where it's due: you usually take the time to write a nice comment explaining why a post is being closed, and how to improve it for re-opening. – Mike Ounsworth Mar 30 '16 at 19:48
  • 2
    @MikeOunsworth Thanks for that encouragement. Sometimes I feel like I get into a rut of swinging the Mod hammer for spam/nonsense/Googlable posts, that it spills over into more legitimate posts. I do try to take time to explain things. – schroeder Mar 30 '16 at 21:10
  • With Googlable posts at least, the user clearly has enough interest and enthusiasm to come here, make an account, and write a post. They're probably a user worth keeping. I try to take a few seconds to make them feel welcome and encouraged - especially if the question is about to get closed. – Mike Ounsworth Mar 30 '16 at 21:20
  • 3
    Is there any evidence that closing fewer questions will cause high-quality contributors to leave? The questions are only put "on hold" and they still appear on the front page. I doubt that leaving more open would discourage ANY high-quality contributors. – paj28 Mar 31 '16 at 14:53
  • 1
    @paj28 evidence would be in the hands of SE devs. As I say, the game they released highlighted the idea (and codified how to accomplish the effect). – schroeder Mar 31 '16 at 15:22
  • I agree with the "tricky and messy" comment. As a recently new professional, I found the SE socialization process unnecessarily rough. When closing questions becomes about censoring and not about question quality, the line has been crossed. There are professionals out there who have different training than the majority of the contributors. We should avoid the know-it-all approach to closing questions, and allow for new answers that have not been previously discussed (even if the question is redundant). – Brent Kirkpatrick Apr 10 '16 at 14:45
  • @BrentKirkpatrick Am I to assume that you feel that you have ben censored? – schroeder Apr 10 '16 at 17:38
  • I have been censored many times in my life. Here on SE, I have had a few questions closed on me that I believe are legitimate and possibly have definitive answers. I realize that I have the option of rethinking my questions from the big-picture and phrasing them a way the community can understand. However, several of my questions have asked for statistical answers or intuitive statistical sense. These types of questions may generate hot debate, but they should be allowed as legitimate questions with multiple definitive answers. – Brent Kirkpatrick Apr 10 '16 at 17:43
  • I realize this whole discussion is messy and the community tries to do its best with these things. – Brent Kirkpatrick Apr 10 '16 at 17:45
  • By the way, when receiving answers to statistical questions, there are rigorous methods of combining statistics from multiple sources. When I ask people to answer based on their experience with intrusions, they are actually contributing to an assessment of the global picture of all intrusions, even ones they did not directly witness. This is statistics. – Brent Kirkpatrick Apr 10 '16 at 17:47
  • @BrentKirkpatrick that's nice that you perceive things that way, but that does not make your question more on-topic or within the format of the site. – schroeder Apr 10 '16 at 17:53
  • Other communities on SE allow questions of this a statistical type. Other communities also allow formal answers of the logical type. If the InfoSec community wants to keep banning those types of questions, you are free to do so. However, you will do so without my participation in the banning, and I will take my rigorous questions to the theory community. – Brent Kirkpatrick Apr 10 '16 at 18:11
  • 1
    Brent - "combining statistics from multiple sources" is not what the majority of SE sites are about, and this definitely not . We have an explicit close reason for "opinion based" - while we do look for experience in answers, we don't ask questions like "how did you deal with X" so yes, we will keep closing those types of questions. There are many other sites, and many online forums that allow discussion, and are setup for gaining consensus answers and debate. SE is not like that. Debate and discussion are a big no-no! – Rory Alsop Apr 11 '16 at 18:30
4

This one bugged me recently, it was not "off-topic" but "Duplicate"

How could online encryption ever work?

The poster seems new, mentioned https once in their question, then got hit with "why not use Google before asking how SSL works..."

It then degenerated into an argument where people debated with the author about how they better understood what they meant than he did... and got closed as a duplicate, when the author clearly stated that the question had nothing to do with SSL.

I would say that off-topic might not be too aggressive, but the quality of the attention to a question seems random, discouraging and arbitrary.

  1. Yes, we should be more friendly. Edit the question to bring it back to life if needed. Never argue with somebody about what they meant, but use what they meant to improve the question.
  2. for comments... like the Google one, I'm not sure... deleting the comment would ruin the chance to get the clarification, but when it degenerates into arguing with the author of the question about what they meant... I'm not sure if the whole thread should just be deleted.

How a conceptual question on public key encryption gets killed off the gate, whereas a complete novice question (on a site for "professionals") gets heaps of positive attention Can a .sh file be malware? is an example of the arbitrariness.

  • 1
    I completely agree. That was a well-intentioned novice grasping for information. The Stack Exchange motto is "Expert answers to your questions", and I can't think of a more fitting question. Shame it got closed, and got such negative comments :-( – Mike Ounsworth Apr 5 '16 at 14:20
  • 1
    I also disagree with closing as dupe of "How does TLS work?". He was asking about the fundamental concepts of exchanging data over an insecure channel. TLS is one technology for doing this, but certainly does not encompass the whole question, at the level of understanding the OP was asking about. – Mike Ounsworth Apr 5 '16 at 14:23
-2

I agree with the question asker and with the other answer writers.

From the algorithmic point of view, the main issue is that new users with low reputations are subjected to more variance in reputation relative to their total reputation than established users. This means that many new users are suffer extinguished reputation on accident. (For the nerdy: the Brownian motion statistical process is a good model for this effect.)

One could fix the down-voting algorithm so that the impact to new user's reputations causes less variance in their reputations. For example, use a simple scaling equation, where the time the account has been open is used as a coefficient for how much the down-votes effect reputation, until a certain threshold of time or reputation is passed.

Similarly the voting to close a question can be scaled so that new user questions take proportionally more votes to close.

I suggest this as a new feature to the reputation algorithm.

  • 1
    I think the question was about closing for being off-topic, not about the effects of downvoting. – schroeder Apr 9 '16 at 4:15
  • My point still stands in relation to down-voting and closing questions as off-topic. The voting for closing can be modified. – Brent Kirkpatrick Apr 9 '16 at 4:22
  • 1. What is an "extinguished reputation"? 2. Note that closing a question as off-topic has no effect on reputation. – D.W. Apr 9 '16 at 21:16
  • An extinguished reputation is one where the user ends up with zero reputation and remains there. – Brent Kirkpatrick Apr 10 '16 at 2:23
-2

I totally agree with all the above posts. I feel that people are too harsh when a new type of question is asked. I feel some people especially on SO take questions that they aren't used to and disregard them.

I believe that no matter how much Logic you have it's almost worthless having it without an imagination or any other sort of creativity. Without any creativity or imagination especially in computer sec/programming we would never be able to come up with anything that was actually our idea. Take most of the famous people from Silicon valley as example most were either helping someone with creative imagination or was the Leader with creative imagination and that was what was most known on the internet!, new ideas. I'm not talking about a scale like that, I'm just wondering why we can't have more options of talking about topics that maybe most won't comprehend. If an average amount of people don't see the bigger picture in a post they may down vote as not useful then the bigger picture wont ever be seen. Maybe a more open community is needed to stop people from killing useful accurate posts... Einstein said without creativity knowledge is nothing!

  • 2
    You got a little too creative with your Einstein quote and misquoted him, and that highlights why this site has boundaries and a format to follow. There are other communities that might allow for free-form thought, but this, and other SE sites, have a unique format that distinguishes it from the others. – schroeder Apr 12 '16 at 4:50
  • That's what you got out of it, that's not what he said. That's a big difference. Again, this is why there are standards for how to ask questions in a community that spans the world. We are not here to serve the individual, but we are here to build a body of knowledge that all can share. If questions get too focused on what's important to the individual in the moment, then we devolve into noise. It's not that people don't understand you, it's that your questions need to contribute as much as the possible answers. And there is a hell of a lot of creativity in that model. – schroeder Apr 12 '16 at 5:07
  • 1
    I agree with Schroeder that a Q&A site should aim to be like an encyclopedia: clear factual questions that are easy to google, with clear factual answers (ideally backed up with references). As you say, creativity is wonderful and lovely and fuels innovation...but isn't the point of this site. If you want to debate philosophy and the history of science, there are other places on the internet for that. – Mike Ounsworth Apr 12 '16 at 14:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .