The Serverfault community seems to treat questions differently than other SE sites... and there definitely seems to be a "culture" or general understanding on how to treat questions that isn't immediately clear.

  • Generally speaking, why is a question upvoted or downvoted here?

  • Should I interpret any meaning based on the value of the upvote / downvote?

  • Is this site just for the "elite" and N00bs can just go their own way.

Friends who I sent to Serverfault got the impression it wasn't too friendly of a place, and stopped using it. In fact I think I remember seeing a similar Meta question & observation within the past 6 months by other users.

PS - I don't care about "rep" but I do care about finding questions that are well written (and responses). When I use the SE DataExplorer to scan through posts, I'd like to know what measurements (upvotes) I should filter by.

Your perspective is welcome.

  • 9
    I'll note that as a site targeted at professionals, we fully expect you to have put some effort into solving your problem before you come here. We also expect you to have done some research before you come here. Personally, I love helping people that have interesting problems or weird edge cases. What I don't like is someone's basic legwork for them (looking through logs when they haven't yet, looking for knowledge base articles for them). If a question is very basic or poorly explained and the OP shows no proof of trying to resolve the issue on their own, then I'm highly likely to downvote it
    – MDMarra
    Commented Mar 2, 2013 at 3:38
  • 2
    What is the expected breadth of expertise of a professional? I don't think it's uncommon for an Exchange admin to ask dumb questions about networking or vice versa. Part of being a "Professional" IMHO is to recognize that each person has something unique to contribute in their subject matter of expertise. Not much "expertise" is universal, and therefore simple questions (and nieve misunderstandings) that aren't interesting shouldn't be downvoted just because the reader is bored. It should closed as a dupe or answered. Commented Mar 3, 2013 at 22:01
  • 5
    A simple question is different than a question where the asker put, literally, no research into solving him or herself.
    – MDMarra
    Commented Mar 3, 2013 at 22:14
  • 6
    Personally, I've found SF one of the less amicable SO sites that I use. I know friendliness isn't the goal, but it helps create a solid community and userbase. I understand from reading this that this is a professional community but I think regular users need to make some effort in steering people toward the type of questions they want, rather than simply down voting. Comments can be provided in a way that helps people ask better questions, rather than just encouraging them to leave the community.
    – glenstorey
    Commented Mar 4, 2013 at 7:55
  • 6
    @glenstorey I'd agree - and I suggest that most of the regulars who are engaged in the community usually do comment. We respect others and want the community to grow. But its a central pillar of the SO sites that one isn't obliged to do so. Respect is a two way street as well, so a question that's especially egregious in its lack of effort might get short shrift here because that's all it deserves. As I've already said, it's not that this site is 'unfriendly', rather that it expects certain standards from its friends.
    – Rob Moir
    Commented Mar 4, 2013 at 8:20
  • 2
    @glenstorey We really try to judge the "Professional IT" quality of a Question on the content of the question. Quickly skimming your profile: You're a Teacher who does some Web Development - you are 100% outside the target audience we're trying to reach. I am not trying to imply that we don't care about your problems - problems on SF have to live up to a Professional Quality that is expected of a System Administrator (et al). Also SF is NOT part of SO; SF and SO are part of Stack Exchange.
    – Chris S
    Commented Mar 4, 2013 at 14:25
  • I just posted a "answer your own question" type of post The intent is to share information in the context of performance troubleshooting. I never heard of fltmc before and discovered it while troubleshooting Exchange and Blackberry servers yesterday. I wonder how the +- votes will go on that. Commented Mar 4, 2013 at 14:39
  • 3
    @ChrisS In my defence I am the network admin of the school - we have between 200 and 500 connected devices at any point. I'm not a Network Professional, but I'm dealing with networking everyday. Thanks for the correction re: SF/SO; my mistake
    – glenstorey
    Commented Mar 4, 2013 at 16:59

5 Answers 5


People are free to vote exactly as their whims take them.

If you float your cursor over the up and down vote arrows the text you see is about all the direction given

  • This question shows research effort; it is useful and clear
  • This question does not show any research effort; it is unclear or not useful.

  1. Serverfault was (uniquely within SE) created for Professionals. I (and many other players) expect that as a professional someone would demonstrate their research as part of their question. Sadly this is often not the case and is dealt with appropriately.

  2. Read into voting what you will. It is though well understood (across SE in general) that detailed well researched and presented answers will often be overlooked in favour of a quick quip.

  3. Serverfault was created for professionals.

  • 3
    You forgot an important point: Serverfault was created for professionals. Commented Mar 3, 2013 at 10:27
  • @MichaelHampton: oooh yes, so I did.
    – user9517
    Commented Mar 3, 2013 at 10:49

To paraphrase a certain comment about Unix, Serverfault is friendly, it just expects certain standards from its friends. It's not that the site is elitest - questions from people new to systems administration are perfectly welcome here.

We expect well asked questions of a professional standard - that means that questions should be on topic, clearly written, clearly express what they're after and show some research. This is central to asking questions here, and is designed to help visitors to the site in the long run.

I personally think this has the following benefits for the question asker:

  • Taking the time to compose a good question and review your own thoughts on the subject will sometimes lead you to discovering the right answer yourself. I've solved more than a few problems by actually thinking through how to express them here or in discussions with colleagues.

  • It will help you get an answer faster: Less ambiguity will mean less time going down dead ends and/or clarifying what you meant and more time answering your central question.

  • It will certainly lead to not just more chance of getting an answer, but also to improving your chances of getting a great answer.


A user who posts a question in an clear, answerable way and who does some independent research is trying to do the right thing and deserves some respect. But we here at ServerFault can't fix every problem. We need users to at least read the man page and plug the question into Google before they come and ask here. They should also phrase the question in a way that others can understand. People who do this are at least trying to be "professional". Systems administration is a very complicated problem and there is often no clear answer. Most of us remember the struggle to get started, and we struggle to fix problems on a daily basis. We know that man pages are hard to read and that Google is full of unclear or old answers. The systems are sometimes so complex or arcane that it is hard to even figure out what is the right question to ask, much less provide an answer

I'm not sure what the other sites are like, but Serverfault has a constant stream of low-quality questions in which it is clear that the posters didn't perform the basic steps required to form a question. The review queues like ("vote to close", "first post") show that many posters don't take the basic steps needed to form a good question.

The bar here may be higher then on other sites, but the answers also tend to be higher quality. I'd rather ask a good question and receive a good answer then ask a bad question and receive a bad answer.


I don't think SF treats questions any differently than other sites. Based on what I see reading and voting on SF and a few other SE sites, the same types of bad questions get downvotes and good questions get upvotes on other sites.

It would be interesting to see actual stats, but it may be that there's a bit more voting overall on SF than some other sites, which would mean bad questions are more likely to get downvotes or close votes than to be ignored.

got the impression it wasn't too friendly of a place,

I can never understand this type of comment...

(Side point: an indication that SF isn't especially unfriendly is that comments like this show up on every SE site that I've read.)

Sure, it's not nice if someone asks an honest but badly crafted question and gets some cranky comments along with the downvotes. But many people come here, post any sort of question, and then - based on the comments they leave - feel hurt when they get downvotes or close votes or lack of comments/edits to help improve their question. We've had people post "why didn't anyone help me improve my question?"

Not everyone is willing to take the additional time it takes time to fix a bad question or to clarify issues in a so-so question that's missing information. It's up to the person who wants their question answered to figure out what constitutes a good question and what the possible responses (votes, comments, answers) mean.

I don't consider it unfriendly to downvote a question and move on (I do it a lot) and it doesn't mean anything different here than any other SE site: just that I thought it was a bad question.

  • I can tell you from experience that SF does treat questions differently from other SF sites. I believe that the emphasis on professional questions means that SF gets proportionally more off-topic and poor quality questions that other sites; given this, SF users are quick to judge good, on-topic questions as poor, off-topic questions. I am an avid Stack Exchange user and my company's syadmin. I take time to research questions and spent time asking them in detail. Nowhere else do my questions get downvotes like they don on SF. Espescially if they're about certain topics.
    – Josh
    Commented Mar 8, 2013 at 13:18
  • Can you point us to some examples of questions you've asked that are on-topic and well-researched that nevertheless garner downvotes?
    – MadHatter
    Commented Mar 13, 2013 at 9:35

Is this site just for the "elite" and N00bs can just go their own way.

"Elite" is called "professional" in these parts.

Basically, it is a site where "regulars" (aka high-rep users) do the heavy lifting(inspired by Cris S's comment here is the result set of another query retrieving the number of answers by Top X answerers last month). Those regulars mostly have jobs where the solution of problems similar to the ones appearing on this site is the daily grind. And they consider an analytical approach to these problems paramount. If a question does not contain enough information to be answered and it seems too bothersome to interact with the asker to get the information required, this question surely will have a tough position here and be ignored or closed.

So basically, the answer to this question is "yes" if you define "elite" and "N00bs" accordingly.

This being said, questions are upvoted significantly less often than answers. Apparently, it is much harder to acknowledge a good question than is the case with a valid answer. Quite likely because a question is rather earning points for performance, while an answer simply has to be valid.

  • +1 This is useful. I should focus my DataExplorer queries on the answer vote count and not so heavily on the question vote count. Commented Mar 2, 2013 at 15:28
  • 3
    To be honest I think we set the bar for professionalism very low, so if people can't even get over it then ...
    – user9517
    Commented Mar 3, 2013 at 19:46
  • 1
    To make it a bit more clear regarding the "regulars" over the last month: 4 Users contributed 11% of all answers. 17 contributed 25%. 91 Users contributed 50%. There were a total of 3610 answers from 1340 users in that time period. Of the 7000+ Users over the last 6 months: one user posted 1 out of every 35. (Stats coaxed out of data.stackexchange.com/serverfault/query/82390/…)
    – Chris S
    Commented Mar 4, 2013 at 14:42

You must log in to answer this question.

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