This is just one example, happens to be at the top of the list at the moment, not picking on anyone.

Twelve balls and a scale

Googling {Twelve balls and a scale} gets "About 37,600,000 results (0.23 seconds)" Yep 37 MILLION, the first few at least are the identical question (unattributed) that we have.

Assuming this site gets out of private beta; in 3 months there is self evaluation, and this question is going to fail.


Needs Improvement if this question is poorly written, not generally useful, or has an answer that is no better than what can be found elsewhere. Alternatively, this question does not turn up on the first page or two in Google despite repeated search attempts.

As the answer is already posted multiple times and all of them are at the top of the search results, the answer will fail.

Do questions like this give any value to the site? If not does that mean questions that can be found on the internet should be deleted?

  • $\begingroup$ Related: meta.puzzling.stackexchange.com/q/51 $\endgroup$
    – msh210
    Commented May 19, 2014 at 8:32
  • $\begingroup$ Right now the question under examination is actually on page 2 of Google (n=12 for me and n=14 for generic results). $\endgroup$ Commented Nov 10, 2014 at 17:46

5 Answers 5


Our site is designed to be a canonical reference for high-quality answers to classic and modern problems. These answers are designed to appeal to the level of question that posted them.

For example, I posted a question about Rubik's cube corner-twist group theory. This answer is one we're unlikely to find anywhere else, because it appeals specifically to the way I wrote the question. This is a classic and canonical problem in Rubik's cube group theory, for which there are dozens of answers of varying complexity; however, they're hard to find, and not well-organized.

If we have a canonical reference for classical puzzles, such as the twelve balls on the scale, when people have these questions, they'll know they can come here to find - or ask. Not every question, especially those which have been answered before, will come up on Google search results. Our aim, however, is to make sure that our answers are high enough quality that they do float to the top.

In essence, we actually can answer these questions better because we:

  • are not a forum environment, in which it is often nigh on impossible to find good answers
  • have people willing to detail answers and their explanations as clearly as possible
  • have a system of voting by which clear and concise answers float to the top
  • have a layout which is clear and easy to read

I therefore think that a canonical reference of questions is a very healthy bank for this site. Because of the nature of the topic of our site, we will inevitably have a hard time getting questions to the top of Google search results. But, when we do, it will draw fantastic amounts of traffic to the site, which grows into a self-reciprocating cycle.

The site evaluation process is a good indicator and metric, but it needs to remain contextualized within the environment and topic of a site.

  • $\begingroup$ Canonical references for common questions are good — but that's not what the private beta is about. During the private beta, we need to seed the site with expert questions. $\endgroup$ Commented May 19, 2014 at 2:59
  • $\begingroup$ I'm trying to provide those expert questions alongside the regular ones (e.g. the minimal magic square one). $\endgroup$
    – user88
    Commented May 19, 2014 at 3:19

No. Nothing should be off topic just because it exists elsewhere. We sure as heck don't want a site only full of things that exist elsewhere (although even that could theoretically be ok, if the grouping here is more useful.)

But that, in and of itself, isn't a reason things need to be off topic. Even if we had a ton of puzzles that do exist elsewhere, but where the top answer here is, on average, a better explanation, etc, that could still mean we've created a better puzzle site, in aggregate, than anything else out there.

  • 1
    $\begingroup$ In addition, getting all the asked-before questions over with now makes room for more interesting questions to be asked in the public beta later on. If we let them just come in naturally, it'll seem like half the site is filled with questions that have already been done a million times. $\endgroup$
    – user88
    Commented May 17, 2014 at 3:57

I don't know whether I agree with others that say that getting all the "been done before" questions out of the way now is better or worse than doing it later or not at all.

However, I can see the concern in answering questions that can be easily answered with a Google search, since that is almost definitely going to lead to more work in the long run, as users who become more familiar with the site, and asking questions, do so.

At the very least, allowing these kinds of questions will lead to a large amount of administrative overhead, with duplicate closures.

Instead, perhaps we should define off-topic questions to be:

  • Questions that can be answered entirely with use of commonly-available internet resources:

    • Google
    • Wikipedia
    • ... and others

    Unless you identify a specific source of concern or concept that you required assistance with. See examples below

This should solve the problem of having answers which are simply links to, or copied straight from the first page of Google results.


An on-topic question:

How does the strategy behind twelve balls and a scale work?

I've been given the following as an answer:

  • Weigh A against B.
  • If A > B, then weigh A1, B1, and B2 against B3, B4, and C1.
    • If the weights are equal, then one of A2...4 is heavier; weigh A2 and A3. If they are equal, A4 is heavier. If one is heavier, then that ball is heaviest.
    • If the first group is heavier, then either A1 is heavier, or B3-4 is lighter. Compare B3 and B4; if they are equal, A1 is heavier; if they are different, the lightest is the lightest ball.
    • If the first group is lighter, then either B3 or B4 is lighter. Weigh them and see.
  • If A < B, renumber all A-balls to B-balls, and perform the above steps.
  • If A = B, weigh A1, A2, A3 against C1, C2, C3
    • If they are equal, then weigh A1 against C4. If A1 is lighter, then C4 is the odd ball and it is heavy. If A1 is heavier, then C4 is the odd ball and it is light.
    • If A is heavier than C, weigh C1 against C2. If they are equal, then C3 is the odd ball and it is lighter. If they are not equal, then the lighter of the two balls is the lightest ball
    • If A is lighter than C, weigh C1 against C2. If they are equal, then C3 is the odd ball and it is heavier. If they are not equal, then the heavier of the two balls is the heaviest ball.

But I want to know how it works. How does doing this determine which ball is heaviest?

An off-topic question:

How do I solve this twelve balls and a scale problem?

(As per original question)

  • 1
    $\begingroup$ Once the site has launched, we do want googlable questions — with the hope that this site will become the first Google hit. It's only the private beta that's different: for the time being, we want expert questions. $\endgroup$ Commented May 19, 2014 at 2:57

I figure that if we get all the "done-before" questions out of the way during the private beta, we can help cut down on people being massively astonished when these puzzles are posted anew by a new user.

It's inevitable that somebody will ask these questions, and I'd rather get that over with sooner than later. We can point to these questions as examples of things that have been already been asked, rather than just things that have cropped up elsewhere on the Internet and are thereby somehow "invalid" to ask here. They are on topic, at any rate.

Not only that, but I think it should definitely be possible to provide an answer to that question that's either better or at least of comparable quality to the other answers out there, especially to the twelve balls and a scale problem. A cursory examination of the first two pages of the Google results for "twelve balls and a scale" shows that some of them show just the answers, others are forum posts where the solutions are disorganized, and the rest are either dead links or don't even have the answer at all. There's a lot of room for improvement on a helpful answer to that question.

  • 1
    $\begingroup$ I can't seem to find meta links to related discussions at the moment, but private beta is the time to fill the site with the best questions and answers not re-print the internet. $\endgroup$ Commented May 17, 2014 at 10:08
  • $\begingroup$ We're not reprinting the Internet. We can't get the best questions and answers in private beta. Private beta is a time for filling in the baseline we set out in the definition stage, and that involves getting all the "done-before" questions out of the way so they don't pop up later, since they're on-topic but ultimately uninteresting. $\endgroup$
    – user88
    Commented May 17, 2014 at 14:14
  • $\begingroup$ Yeah, +1. Private beta is like a better Area 51 definition phase, with emphasis on community-building. $\endgroup$
    – user20
    Commented May 17, 2014 at 21:54
  • 1
    $\begingroup$ It's better to let the low-value, million-hit questions be asked naturally. They will be asked anyway. During the private beta, focus on asking questions where there is some added value. $\endgroup$ Commented May 19, 2014 at 2:58

During the private beta, we want to focus on questions that will attract experts. The questions asked during the private beta set the tone for the site, and given that the private beta is run by a small core of highly-motivated individuals, you can only expect quality to decrease. So you should avoid posting questions that are all over the web. Asking frequently-asked questions isn't forbidden, but it does reduce the value of the site. Focus on seeding the site with highly valuable questions whose answer is hard to find.

The “been done before” questions will occur naturally once the site is open to the public. There's no need to seed them.


You must log in to answer this question.

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