1

So it seems ok for someone to answer a question to use google http://meta.serverfault.com/questions/6124/should-someone-post-an-answer-when-all-they-did-was-google-fu

But if a user does not use google the question is closed (eg)

https://serverfault.com/questions/723754/sql-server-2014-enterprise-edition-is-consuming-too-much-ram-i-need-to-limit-it

I don't disagree that its pretty simple to answer, I disagree that the close reason is (essentially) " your question is poorly worded" when what we need is an answer that says (essentially) " you lazy bug, it took you longer to type this question than for me to search for the answer, go away".

I'm ok with the premise that basic research should be done, but IMHO we should pick a consistent stance, its either ok to google for the user and post or there should be a close option telling the user to search.

| |
4

I'm thinking that question should actually be reopened. I know, I'm as shocked as anyone! Here's what is rattling through my head (in no particularly coherent order):

  • The question does, indeed, have a clear problem statement and an answer that can be expressed in no more than a few paragraphs.
  • It is related to administering computer systems in a business environment, or it is a question that could reasonably asked by someone in such a situation.
  • Practically every question asked here could be answered with a little Google fu. That, alone, should not be a criteria for closing a question.
  • New answerers need something to cut their teeth on. This question seems like a good candidate for that.
  • I was told, many moons ago, by someone far more level-headed than I, that just because I felt it was beneath my skills and experience to answer a particular question, that alone didn't make it a bad question. I don't always manage to live by that, but I try to bear it in mind when I'm waving the closehammer around wildly.
  • I recall one of the founding principles of SO, at least, was to be a repository of all programming knowledge, even if a lot of it is trivial stuff. Whilst SF != SO, I like to think we're trying to build the same sort of thing, within our specific topic area.
  • The major point against reopening, IMO, is our desire for more questions worthy of a professional, and that you're not much of a professional if you can't stick "SQL Server limit memory" into Google and mix-and-match the results to your situation. I refer the honourable member to my earlier point about how just about every question on SF can be answered with Google fu, and extend and revise that by saying I'd prefer to use our Google juice than to hoard it for a rainy day.

I'm not going to stomp on a closing, though. I'll leave it closed, and if others find my arguments persuasive and the question gets reopened, so be it.

| |
  • 3
    Good points indeed. I still have a hard time with the no apparent research bit, but we'll see how this plays out in the next day or so. If people think it should be re-opened, I'm not going to oppose the community. – EEAA Sep 21 '15 at 22:41
  • "Whilst SF != SO, I like to think we're trying to build the same sort of thing" and has been argued ad nauseum we are specifically NOT trying to be a repository, we've made it crystal clear we only want a certain type of explicit question. I'm perfectly OK that its closed but we need to be clearer on why. – Jim B Sep 21 '15 at 22:55
  • The first comment left was the problem there in my opinion. It seem funny when I read it, but the tone can be badly decoded especially if you are not english. – yagmoth555 Sep 22 '15 at 1:06
  • No research is not only a valid close reason, if you can easily find duplicates on server fault itself it's not just no research, it's an easy close as duplicate. – Reaces Sep 22 '15 at 7:17
  • 1
    @reaces Close-as-dupe, sure. Which close reason would you use for "No research"? None of the existing close reasons scream "use me for no research questions!" to me. – womble Sep 22 '15 at 8:13
  • @womble It's so unlikely though that we come across a question that is blatantly no research and does not have a duplicate. And if we do want to forward the aims of StackExchange wanting to be a knowledge repository. Then answering something which could be googled but doesn't have a ServerFault answer is just fine. – Reaces Sep 22 '15 at 9:13
  • A "No Research" close reason is just begging to be abused, I'd be afraid to add one. It's pretty clear from all the site's documentation and the time we've spent explaining how to ask a good question that we expect to be AFTER Google in the help-seeking chain though. A "Custom" close and a comment pointing at any of those pages would be my choice if there's no good duplicate target & it's a REALLY lazy question. – voretaq7 Sep 22 '15 at 19:09
  • @voretaq7 you know as well as I do that a new users doesn't read the sites documentation,. ten post- they just post. Custom Close is not a bad option either but I suspect that the canned close reason will be "you didn't google" , otherwise it's far simpler to just answer the question. – Jim B Sep 23 '15 at 4:40
9

Ignoring the fact that RAM usage is good in database servers, this one failed the "no research effort" sniff test for me. I am not a Windows guy, but was able to find a straightforward answer to the OP's question, direct from Microsoft Documentation, with a single Google Search.

Server Fault should not be a replacement for reading documentation. If a user won't at least consult the freely-available documentation on a simple issue like this, then frankly they don't deserve the free assistance available here. Don't take that as mean-spirited, I don't intend it as such. My position on this is comes from the viewpoint that it's far better to encourage people to find answers on their own than to spoon-feed them one-off answers on demand.

| |
  • 1
    I agree it just has a crappy close reason AND meta said its ok to google for the user – Jim B Sep 21 '15 at 22:57
  • I agree to a point, but I would argue that for outward appearances sake (not just the user asking, but those who see how we treated the question) closing it as a duplicate looks like a vastly more friendly action. – Reaces Sep 22 '15 at 7:19
  • 1
    I agree that closing as a dupe is better, so long as there is a dupe. – Jim B Sep 22 '15 at 23:14
5

Should this question be reopened?
Noooooooo

It's not only basic knowledge for most who use MSSQL.
Which doesn't actually matter all that much.

But primarily!
It is a clear duplicate.
There are so many posts on serverfault / dba about the memory usage of SQL Server, I don't see why it's not just closed as a duplicate.
It would answer the OP's question quite quickly if it was just closed as a duplicate, serve as a pointer to the tremendous amount of answers about SQL Server memory, and push the point please do your own research more than simply closing it does. (not taking into account comments who are generally regarded as non-existent in the long run)

When a question comes up that is clearly common knowledge, chances are you'll find a post on ServerFault about it. For any other questions a well written ServerFault answer won't harm anyone if one does not already exist.

| |
  • 4
    It is generally much easier to write a new answer than it is to find the duplicate that you know exists and frequently that (correctly) translates into let's just close this shit. – user9517 Sep 22 '15 at 9:31
  • Excellent. Thanks for tracking those down. I've re-closed it as dupe. – EEAA Sep 22 '15 at 13:03
  • @iain, if the correct answer doesn't show up on google (from SF) then closing it serves no purpose. – Jim B Sep 23 '15 at 2:02

You must log in to answer this question.

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