10

Possible Duplicate:
Is “Don't do it” a valid answer?

I'm writing in regards to this question on Stack Overflow: Wordpress stop bookmarking

OP is asking how to redirect his website visitors so if they hit an internal page, they are redirected back to the homepage. The purpose of this appears to be to increase revenue and/or lower bounce rate.

I know that there are plenty of sites that do something similar to this, like splitting their articles across multiple pages to force you to clickthrough to read the whole article (and show you more ads). I've just never seen it stated so plainly in public.

My question is:

Sometimes "you shouldn't do that" is the correct answer to a question. We shouldn't simply take the OP's assumptions for granted. Questioning the specific implementation sometimes leads to a better solution.

But do you think that applies in this case?

In this case, I'm thinking "don't do that" not because I think there's a better way to implement it, but because I think the whole idea is dumb in the first place. Is that my business and is it my place to speak up?

What is the appropriate response here?

  • Just answer the question
  • Tell the OP "you shouldn't do that"
  • Something else?

I'm familiar with questions such as Is "Don't do it" a valid answer? which generally tell you to be helpful while at the same time presenting other options, however in this case I have a feeling the OP doesn't give a darn about anyone's opinion of what he's trying to do. He will simply take the solution and run off to build his annoying website. Is it still recommended to provide a solution anyway?

5
  • 1
    While the question is technically valid, the OP is being a massive dick, and he won't care about a "don't do it" answer. Just downvoting and walking away is also an option, and the most time-efficient one at that.
    – Pekka
    Jun 11, 2012 at 0:49
  • @Pekka I flagged it to be moved to Webmasters. Do you think that makes sense? He isn't asking a specific programming question, just a general question about how to implement a specific marketing strategy.
    – JimmyPena
    Jun 11, 2012 at 1:08
  • It looks like someone already tried the "don't do it answer" – and that answer has got 11 upvotes already. The OP may not care, but such feedback may help other neophytes who peruse these boards for sound information and advice. Jun 11, 2012 at 1:09
  • @J.R. yeah, that's a strong argument in favour of posting a "don't do it" answer, regardless of the OP's character flaws...
    – Pekka
    Jun 11, 2012 at 1:15
  • I'm not following why the OP's attitude changes the answer to «Is "don't do it" a valid answer?». Agreed with Mrozek that this is a dupe. It should also be noted that the question under discussion is a bad SO question and should just be nuked.
    – jscs
    Jun 11, 2012 at 4:48

6 Answers 6

8

I think the perfect answer on a question like the one you linked to contains:

  • Good arguments why what he's trying to do is a really bad idea.
  • A solution for his problem since there's a good chance that he either doesn't care about the fact that it's bad or already knows it but has a boss who wants it anyway.
3
  • 2
    I see these pretty often; a simple (hopefully) explanation of how you can do X, with a note of "hey, you shouldn't do X because it might Y." Extra credit for including information about a non-stupid way to get the same effect
    – Ben Brocka
    Jun 11, 2012 at 0:49
  • Do you think it should be in that order? Argument first, then solution?
    – JimmyPena
    Jun 11, 2012 at 1:09
  • 1
    In such a case yes - so it'll be read first. Jun 11, 2012 at 1:50
3

I think it often is appropriate; I see examples of this all the time. The O.P. may not accept the answer, but other folks very well may vote it up. If it gets voted down, well, then people don't agree with you. But it's worth a try. And someone else who is closer to being on the fence may search, find your answer, and think twice about implementing the questionable practice.

Remember that StackExchange is supposed to be a living repository of questions and good answers. It's actually not all about helping the person who posts the question; it's about helping anybody who ever reads your answer.

3

The ultimate goal of anyone asking a question on Stack Overflow is to get help.

As long as you're truly try to be helpful – explaining clearly why the OP shouldn't do what he's doing and, of course, doing it with a nice tone – I don't think there's anything wrong with saying don't do that!.

In this specific case, the OP doesn't seem to eager to take good intentioned advice. I quote:

I was not looking for your honest opinion, i was looking for a solution to my problem. Content is king, and the only place on the web they can get this content is my site, so if it takes 40 clicks to get to it they will do it.

If that's his opinion, fine! But he won't be able to say afterwards that nobody warned him.

What is the appropriate response here?

  • Just answer the question
  • Tell the OP "you shouldn't do that"
  • Something else?

As long as there's nothing crucially wrong with trying to help the OP achieve his goal, I would go with something else: Do both!

Actually, unless you're actually going to answer the question or debate the technical wrongs of the OP's ways, saying don't do that! in a comment, would be more appropriate.

2

If the OP wants to treat his users this way or in fact thinks his content has a high enough value that users will in fact jump through hoops to get at it, then who are we to question his judgement?

However you all just hoodwinked yourselves into a pointless discussion with the OP and missed what should be the main concern about the question - it's not very good, so why aren't you closing it as low quality or not a real question instead of arguing the toss?

*I'm not looking for opinions on ethical web design, i'm looking for a method to a solution to solve a problem.

ideas: Session ID, Complex redirect based on previous url?

Did you all miss this memo?:

Stack Overflow is not a research assistant

6
  • Good point, good point, Sir
    – Pekka
    Jun 11, 2012 at 1:26
  • I flagged it to be moved to Webmasters, since it isn't asking for help with a specific programming problem.
    – JimmyPena
    Jun 11, 2012 at 1:26
  • @JP. I'd decline that migration request, and I reckon the WebMasters mods would probably say no as well due to quality and the baggage it now carries.
    – Kev
    Jun 11, 2012 at 1:29
  • OK. Is there a way to change how I've flagged it?
    – JimmyPena
    Jun 11, 2012 at 1:34
  • @JP. - I declined your flag, you should be able to flag to close as the reason of your choosing.
    – Kev
    Jun 11, 2012 at 1:36
  • Thank you Kev, in the end I flagged it as NARQ although it seems "not constructive" also fits the bill.
    – JimmyPena
    Jun 11, 2012 at 1:42
1

This is a primarily technical site. A "don't do that" answer is appropriate if you think that the OP, or whoever will have to maintain his code, will regret the stated goal/decision.

In your case of a website, that might be the case due to the site users ultimately going away, but the further you get from technical arguments and start competing with whoever is paying the OP for defining the task, the less you should consider writing an answer.

It is your choice what questions you answer and what questions you downvote with an explanation.

1

however in this case I have a feeling the OP doesn't give a darn about anyone's opinion of what he's trying to do.

So?

Stack Overflow is not about just helping the person asking the question. The Stack Exchange model is a way of building a database of information by using a question/answer format. That's what the "too localized" close reason is all about; questions that would never be useful for anyone besides the OP.

You should always provide the answer that you feel best resolves the issue. If the best resolution is "not that way," then say so. If the OP refuses to accept that solution, fine; the next person who has the same problem might be more reasonable. Let the community up/down vote your answer appropriately.

Yes, you might not get the accept check. But ultimately, you've done your part in providing real, useful knowledge. Not necessarily to the OP, but to someone.

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