33

I'm interested in soliciting opinions on the meaning of the phrase from the FAQ in the question title. I've just returned from an extended hiatus from SF, and I'm finding that a lot of the things that frustrated me last time are still here (I'd hoped that a bit of time and distance might help my perspective, but perhaps not).

The main thing that bugs me is the fact that so many questions (and, to a smaller degree, answers) aren't what I would consider appropriate of professional system administrators. The vast majority of questions are of the form "I don't know what I'm doing, please do my job for me".

As a "poll", I'm going to throw two answers into this question myself:

  • "Serverfault is for professional system administrators to discuss matters relating to their job"; and
  • "Serverfault is for people to come to ask questions of professional system administrators"

But I'm also interested in more nuanced views from other people.

(And if you want to point out the irony that someone who is, technically, no longer a professional system administrator is asking this question, feel free).

  • 6
    well, in all honesty, it is a bit of both. The important part is enforcing question quality, that is, it's OK to ask questions of professional sysadmins provided you have done your homework first. – Jeff Atwood Jul 17 '11 at 3:23
  • I can relate, and asked something similar a few months ago. meta.serverfault.com/questions/973/… – Mark Henderson Jul 17 '11 at 8:10
  • 7
    FWIW, my interpretation of "professional" or "professionalism" is related to how the person does the job, not whether they are being paid to do it. Conversely, just because someone is being paid to be a sysadmin doesn't in and of itself make make that person a professional sysadmin, as is amply demonstrated by some of the questions and answers being posted on SF. – John Gardeniers Jul 17 '11 at 22:24
  • @Mark: Yeah, I found Joel's answer in a link-clicking fiesta from sysadmin1138's post. I'm not sure it answers the question I posed, but it certainly answers the larger issue of my frustration. – womble Jul 17 '11 at 22:45
20

We've had a few go-rounds on the "what is professional?" topic already. The biggest one:

And every couple months someone will ask why a question of theirs was closed, and has to have explained to them why they should have read the FAQ or understood it better:

As I see it, "professional" in the FAQ is a very key filter between us and SuperUser (the other one being the word "server"). We've already lost some active users because questions here aren't professional enough, and they saw the continual stream of un-clued askers as proof the SF concept had failed; they'd had enough of the "I'm totally under water, just give me a fix," attitude on other sites and fora. Having Professional in the FAQ gives us reason to close the more egregious questions like that.

One of our key problems is that System Administration is a very broad field. Linux professionals with ten years of experience suddenly thrust into an environment where they have to maintain Exchange and SharePoint (or Network Engineers suddenly working on servers) are going to be asking some very basic questions in spite of their otherwise deep experience. Sometimes you can tell that someone has that kind of deep experience and is in a new-to-them environment, other times they just can't ask good questions anyway and when doing so in another language it just comes out a barely coherent muddle.

The proposed wording expands the scope of ServerFault to a degree I disagree with. It may be 'revising to reflect reality', but the new wording puts an explicit endorsement on the behavior that many of our active users explicitly do not want. We have an active group of 3K+ users who vote-to-close these sorts of 'off topic' questions.

  • I get your point, and I'm thankful for the links you provided (Kyle Brandt's post in "Perception of Purpose" is a nice answer towards my question). I'm not sure what you're referring to by "proposed wording" -- I'm not looking to change the FAQ or anything, just trying to provide a couple of different ideas about what Server Fault might mean, and seeing which one more people agree with. – womble Jul 17 '11 at 4:11
  • It seems that 5 votes to close aren't hard to come by for the majority of the bad & misplaced questions. Sometimes it's just a matter of me, as a user, needing to filter well enough to see what I'm interested in & cull the rest. Maybe you want to filter out the "low quality rating" questions? (Not sure whether that's possible?) – Kara Marfia Jul 17 '11 at 4:32
  • 2
    "5 votes to close aren't hard to come by" -- that hasn't been my experience. There's plenty of dross that floats around in the "voted to close" section of the 10k tools that never gets to 5. – womble Jul 17 '11 at 5:17
  • 5
    3K rep is required to vote to close. We have 178 out of 60K people who can vote to close. Only 36 of the 178 can see the 10K page. In the last week 42 (including mods) used a vote on 55 questions. Mods tend to react to other votes and close before the 5 required is reached but they closed 8 without prompting. So we had 35 non mods vote on 47 questions. We need to increase the visibility of questions that need closing. We need to get more people who can vote and we need to get them to do so. – Iain Jul 17 '11 at 9:51
  • @Iain Mod-hammering tends to happen after users flag votes in lieu of close-voting. Sometimes questions get flags w/o close votes which is where some no-vote-closes happen. Othertimes, we run across 'em naturally. Sometimes a 3K+ user will get impatient and will both close-vote and flag, but this is happily rare. – sysadmin1138 Jul 17 '11 at 15:03
  • 3
    @sysadmin1138: That's what I noted. We just need to get more people voting more often. The real core of voters is something like 10-15 people. – Iain Jul 17 '11 at 16:35
  • 2
    That is a really interesting analysis. I had kind of suspected that there was a pretty small cadre of people casting most of the close votes. There is even a chat room in which they can hang out and plot where to strike next... It would be interesting to see a similar analysis of who votes up questions - I've seen it commented frequently here that unlike other SO sites, hardly anyone votes questions up on Server Fault. I would be comforted to find that the few who vote up good questions had a strong correlation with those who vote to close most. – dunxd Apr 19 '12 at 15:44
7

First, I want to apologize for posting this if it goes against some standards for decorum around here, especially since not only my rep but also my general knowledge level is very low. However, I feel the need to ask about the attitudes of the prevailing members regarding people in my situation.

I'm still in college, and currently working at a company on their IT staff, but my background knowledge is fairly limited. I'm hoping to learn as much as I can, and I feel Server Fault is a great thing for me to be a part of...but I feel like it would be improper if i did not try and at least find out what the general feelings were towards people who want to learn.

I guess what I'm saying is, is it considered acceptable to use this site as a place to interact, learn, and gain experience until i can legitimately consider myself one of your peers, or will I be chased out for not knowing very much?

note: i would have posted this as a comment if i were allowed to.

  • 1
    Please post questions as such. This is a Q&A site, not a regular discussion forum, so asking a question as an answer to another question doesn't fit in with how the site works. – John Gardeniers May 2 '12 at 21:35
  • 2
    Good questions that fit in with the site's aims are welcome from any source, acolyte, which generally means questions about actual problems rather than hypothetical "what if I did this..." discussions. If you want to discuss stuff, it might be welcome in chat, though that tends to be rather free-wheeling and off topic (think "in the bar after work" more than "talking shop in your lunchbreak", if that makes sense). Its certainly not a case of "not knowing enough" - anyone can have a good question and anyone can have a good answer to that question. – Rob Moir May 2 '12 at 22:21
  • ok, that's good to know. +1 on the username mate. vinyl scratch is teh best pony. also, i'm assuming upvotes on meta topics don't give you rep. – acolyte May 3 '12 at 2:21
5

Whenever this topic comes up, it is a bit abstract. I do think we have a problem with the "expertness" of Server Fault. That being said, in an attempt to start to pin this down I came up with the following expertness ranks for a question:

  1. Person asking doesn't even know where to start, probably not an SA, likely to get a close vote or two but question stays open
  2. Basic Question -- i.e. a junior SA or someone new to the particular technology. Answer is very likely elsewhere on the net or in a man page -- but that is okay
  3. Normal SA question, understand the technology, but wondering how do non-obvious X in particular -- maybe to save time because someone has done it before. Possibly answered elsewhere, but not likely
  4. Will likely require an expert in the technology to answer it
  5. This is going to take a master in the technology to solve. Likely a bug, or a question particular a very small subset of users of that technology. i.e. high performance or issues with scalability.

Given a healthy site, and a healthy profession, I would expect to see a normal distribution of frequencies centered on Rank 3. This isn't based on any particular evidence, just what I personally feel the site and profession should be. Some might like it to be all 4s and 5s with maybe some 3s but this just results in comp.lang.c syndrome (Sick of "unprofessional" questions)

Given what I know about system administration as field, I would still expect 3 to be the highest frequency, but with less weight on 4 and 5 and more on 1 and 2 if the site is performing ideally.

So what I did is going through the newest questions and rank them with quick snap judgments until I got bored. I plan on repeating this a couple times this week. This is, of course, pretty shabby sampling and highly subjective. On the other hand, it is at least some data.

Round 1:

enter image description here

Rank, QId
3, 291406
2, 291400
3, 291396
2, 291396
2, 291394
2, 291390
1, 291388
3, 291387
2, 291385
1, 291393
2, 291381
2, 291378
1, 291374
4, 291373
3, 291372
4, 291371
3, 291367
4, 291365
2, 291364
4, 291356
4, 291354
2, 291353
2, 291349
3, 291348
3, 291347
4, 291345
2, 291343
4, 291342
2, 291337
2, 291336
2, 291333
  • 1
    I tend to agree but believe that if people actually took a few minutes to acquaint (or re-acquaint) themselves with the site (AKA read the FAQ and pay attention to other questions and answers) there should be almost nothing in level 1 and a whole lot less in level 2 than what we do see. Bear in mind that most level 2 questions indicate a lack of professionalism because the poster has generally failed to do any real research prior to posting. – John Gardeniers Jul 18 '11 at 21:42
1

Problem I have with the word "professional" is I'm usually using SF in "a professional capacity" (ie, to solve problems that arise as part of my paid employment), but I'm not a "professional system administrator". Working with Linux servers constitutes perhaps 10% of my job.

The thing that distinguishes a "professional" approach vs an "amateur" one is probably how much time you can devote to finding a solution, and the trade-offs you're willing to make.

I'd be interested to know what actual problems are caused by having people like me ask "easy" questions, and whether raising the bar to participation wouldn't do more harm than good.

  • 2
    If you have a question please post it as such. – John Gardeniers Apr 11 '12 at 4:11
  • 1
    Taking select words out of context does not an argument make. You are either doing System Admin work in a Professional Capacity, or your Professional Capacity is another job. Time has nothing to do with finding a solution; see despair.com/incompetence.html. "Easy" questions drive away actual knowledgeable professionals and garners an atmostphere of "help desk" (or hell desk as most of us think of it). No pro wants to work for 8 hours, then pour more time into helping people who didn't read our FAQ. There are plenty of forums around that will "help" you. We do not aim to be one. – Chris S Apr 11 '12 at 12:40
  • 2
    Steve, the problem isn't with people asking "easy" questions as such, its with people asking questions that don't really fit the site's aim and purpose. We welcome on-topic questions as one of the aims of stack exchange sites is to become a reference library for their topic. In my opinion, the only problem this site has with "easy" is when its a synonym for "Do my work for me" or a blatantly off-topic question. – Rob Moir Apr 11 '12 at 13:46
  • 2
    It's worth noting that a someone switching to a new OS, device, etc. may well ask questions which would appear to be "easy" to those with greater experience in that area. e.g. A Windows admin trying to get his head around a Linux system may well ask questions which appear very easy to answer for a Linux admin. This can be compounded by not really knowing what search terms to use while researching the issue. – John Gardeniers Apr 11 '12 at 22:43
  • An "easy question" is one that someone with reasonable experience in a professional role would be able to solve on their own using the tiniest bit of effort. Things like "How do I allow SSH through iptables" is an easy question, for example. It doesn't belong here. You'd hope that someone that manages Linux servers in a professional capacity would at least know where to find that answer if they didn't know it off the top of their head. We're not here to do people's basic research/troubleshooting. We're here to google things for people. We expect real effort to solve the issue from the OP. – MDMarra Apr 12 '12 at 15:11
  • 1
    Ok, here's my take: the way I solve "easy" problems is usually through Google. If I'm searching for it, it's likely someone else will want to search for it. If I can't find a good, simple answer, then why not ask here, so the next person will find that answer? It's not that I want someone to "do my work for me" - it's that I want to contribute to the body of knowledge that is Q&A, including "easy" questions. (And I self-answer a few of those easy questions, too...) – Steve Bennett Apr 13 '12 at 0:09
  • 1
    Incidentally, my most popular question here: serverfault.com/questions/367921/… Do you consider this "on topic"? It seems pretty "easy" to me, but clearly it interested people. – Steve Bennett Apr 13 '12 at 0:12
  • 1
    There's a fine line between an easy question and an easy answer. Your question I would say is on-topic, though by a narrow margin, and though the answers were "easy" for some people, most non-unix SysAdmins would never guess any of them. – Chris S Apr 13 '12 at 0:46
  • My opinion is that if you can get the answer clearly by reading the man page, you didn't put enough effort in. I find it insulting for someone to ask a question that is already answered in the documentation. It is like saying "my time is more valuable than yours. I'm not reading the docs, but you can for me." That said, I think your example question is probably fine, though maybe better suited for unix.se. – MDMarra Apr 13 '12 at 10:44
  • 1
    Interesting. In response, I find it annoying that man pages are so hard to read that it's usually quicker just to google the specific issue you're having. It can take a long time to read through all the subtle distinctions between similar command line switches to work out exactly what combination you need. Frequently, googling (and ending up on a site like this one) gives a much better answer: "here are the most common switches, you might also need this, and make sure you don't do X". – Steve Bennett Apr 13 '12 at 14:04
  • 3
    @SteveBennett There's a difference between asking a question that says "I read the man page for rsync and I've tried $x $y and $z, but I can't for the life of me figure out why it's not producing the desired results." and a question that says "I need to copy files. I heard rsync was good. Someone tell me what I need to do." – MDMarra Apr 13 '12 at 23:12
  • Indeed. What I'm arguing is that the second is not necessarily bad, as long as the question and answer describe a common scenario - and as long as there aren't already similar questions on here. – Steve Bennett Apr 14 '12 at 1:54
  • 1
    @Steve, the second example IS bad because it shows the poster hasn't put any effort in for him/herself. Such a question is almost certain to get closed. – John Gardeniers May 3 '12 at 21:26
  • You're starting from the existing axioms, that include the requirement to demonstrate effort. I'm not. – Steve Bennett May 3 '12 at 23:51

You must log in to answer this question.

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