In this question, the asker is asking how to assign 2 IP addresses to their computer using DHCP. However, they clarified their question, and the reason they want to do this is to bypass an IP-based download limit; unfortunately, the answer to their question won't solve their actual problem (if they did get 2 local IP addresses, it wouldn't bypass the download limit, because of NAT). I was going to explain it in a comment, but the comment would have been right up against the length limit, so I figured that it may not belong there. What should I do?

  • Say it in a comment
  • Say it in an answer
  • Something else
share
15  
I'm a big fan of user education. If someone is trying to do something that is provably wrong or based on bad assumptions then they should be educated about it irrespective of whether it "answers" the question. At the very least people should be able to be told that they are "doing it wrong". If the explanation of why takes some doing then I see no issue with it being in an answer. –  Mokubai Jan 31 '13 at 8:26
add comment

4 Answers

up vote 17 down vote accepted

This happens now and then, and it's known as the XY Problem.

This user is asking for a solution which will not solve his initial problem. Instead, he could have asked "How do I by-pass the download limit?".

As long as you point out that the solution the OP is looking for will not solve OP's real problem, and offer an alternative solution instead, then it should be fine.

Is there a risk for your answer being flagged as "not an answer"? According this answer, (from the discussion "When should the “Not an Answer” flag be used?", no:

A bad or a wrong answer is still technically an answer. Moderators aren't here to judge the correctness of answers. That's what the voting system is for, so the right way to handle those is to downvote, edit, or leave a comment.

share
1  
With XY problems I'd suggest to either rewrite the question to state the actual problem, or close the questions as Not a Real Question if they're (more or less) pointless due to false assumptions. –  slhck Jan 31 '13 at 11:42
    
I agree. Rather than pointing out the problem with the question in an answer, which can be missed, it's much better to point it out directly to the OP as a comment, or by rewriting the question. –  TFM Jan 31 '13 at 17:06
1  
I think on some level that it's human nature to seek things not in your best interest. The question narrows this basic question to technical issues: "I have an approach to solve my problem which you can tell won't help me; please help me implement this approach." In which case I would suggest (1) pointing out that the question as asked doesn't address the user's interest, (2) answering the question as given, perhaps in a cursory manner, and (3) tell how to address the user's real interest. –  Jonathan Hayward Jan 31 '13 at 23:58
    
A competent consultant gives their customers what they need not what they want. If you have a problem client, give them precisely what they asked for. This is such strong medicine that one dose is usually sufficient to cure the problem. –  David Schwartz Feb 6 '13 at 22:34
    
What happens when the user retaliates with "I know what I'm asking", inferring that the XY problem doesn't apply to them? –  Qix Feb 11 '13 at 11:05
    
@Qix: There's actually nothing wrong with the XY problems, they are as technical as the rest of the questions. The issue is that searching for the "wrong" starting point will only delay the solution to the "real" problem. –  TFM Feb 18 '13 at 23:04
add comment

I would state it in an answer for a few reasons:

  • technically, it is an answer. Sometimes the answer to the problem is that there is no answer.
  • Space limit on comments.
  • It will give people the opportunity to up-vote/down-vote your "answer" to strongmen the probability of your answer being correct.
share
add comment

What should I do?

  • Say it in a comment
  • Say it in an answer
  • Something else

It's perfectly appropriate, when someone asks a poorly-framed question, to give them an answer which correctly addresses their underlying need, rather than just give them the answer they want to hear.

I usually try to give both, however -- people are more likely to take your explanation of why the answer they asked for isn't going to fix the problem if you can first demonstrate sufficient understanding of the situation to answer the question they thought they should be asking.

share
add comment

If other users can benefit from the asked question, you should not change it. Answer in the answer. Address pragmatics, the reasons why people ask questions, in the comments.

Regarding this answer, I am sure you can have two IP addresses without NAT. Internet routes packets to your gatewat computer, which may have two IP interfaces. I see no problem to use to IP interfaces to increase bandwidth. It is natural solution.

share
    
While you can use two interfaces, and that can theoretically increase bandwidth, his examples used addresses in the 192.168.0.0/24 subnet, which are private IP addresses. You cannot connect to the Internet with such addresses; the assumption would be that they are behind NAT. It turned out that he was using them as placeholders, but as asked, no, that wouldn't have solved his issue. –  cpast Feb 13 '13 at 0:19
add comment

You must log in to answer this question.

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