Are questions allowed on SO/SF/SU if they're strongly programming/sysadmin/computer-related but ask about the future? For example, every few days, someone asks a question on Stack Overflow about closures in Java 7; the most recent one I've seen, and the inspiration for this post, is here. I haven't been tracking Java 7 too closely, but I don't think the official spec for the behavior of closures in the as-yet-unreleased version has been published. Some of the earlier questions on the topic were asked indisputably before the information was available.

On the one hand, the SO-family sites are designed to provide objective answers, and it's just plain impossible to definitively answer "how does X work" when X doesn't exist yet. On the other hand, questions are designed to stick around for a long time and be edited to reflect changes in technology rather than grow stale and wrong over the years.

This topic has been touched upon in the feature request New close reason: "Question asks to predict the future", but despite a medium level of support there, no solid conclusion was reached.


Such questions seem to be invitations to discuss, no?

To the extent that the above is true, they are out of bounds.


No. Questions about the future are by definition speculative e.g. "will such and such happen?" or "if this happens, what will the effect be?" You simply cannot answer that.

I cannot think of an exception which is not speculative.

  • But what happens if we take that to its logical conclusion? Supposing a customer asked "will this code work tomorrow?" or "if my internet connection goes down, what does your code do?". By the rules you've written there, aren't both questions unanswerable? Do you really say "oh, by definition no-one can possibly answer questions about the future" – MarkJ Jul 16 '10 at 13:58
  • 2
    OK, I forgot to add that you need to use common sense... Questions on planning are permitted because plans for the future are known now, not speculative. And the only time people asked, "will this code that works today still work tomorrow?" was on 31st December 1999. – Reinstate Monica - Goodbye SE Jul 16 '10 at 17:44

I think these questions can be fine although I'm genuinely humbled that the community disagrees so strongly.

Questions like "will technology X be supported" can sometimes be answered definitively. For instance it was possible to answer will Windows 7 support the VB6 runtime before the release of Windows 7, because Microsoft made a commitment that it would.

It is possible to make decent predictions of the future. I hope this isn't waxing philosophical, but when you say "see you tomorrow" to a colleague when you leave the office, that's predicting the future.

  • I think your example illustrates the problem perfectly. You can have every intention of seeing your cube neighbor tomorrow, but you could also get hit by an asteroid on the way home in a freak accident. For a more real-world example, game companies can get 90% of the way through development, have a release date and marketing materials prepared, and then find the whole project gets canned with little warning. – Pops Jul 16 '10 at 14:25
  • @Popular yes, you could get hit by an asteroid. Does that mean you should never make plans for tomorrow? Yes, companies sometimes have to change their plans unexpectedly. That doesn't mean they shouldn't plan. – MarkJ Jul 17 '10 at 15:17
  • of course you should plan; I never said you shouldn't. I'm saying that it's possible to tell people with 100% certainty that you intend to see them tomorrow at work, but it's impossible to tell people with 100% certainty that you will see them tomorrow at work. – Pops Jul 17 '10 at 16:43
  • @Popular My point is that I absolutely agree that it is impossible to make 100% certain predictions of the future. But that is not a good argument for banning questions about the future from SO. – MarkJ Jul 17 '10 at 20:28
  • 1
    I think having answers that tells you what is planned to be implemented in the future isn't a real issue. The only problem I can see is that the answers will have to be edited if the plans change unexpectedly. Though I would argue that this issue is identical to talking about current solutions, they also need to be updated over time. Take the Android platform as an example. A lot of the things you could/should do in the beginning of the API release are now obsolete and deprecated. And a lot of the answers that had solutions aren't version specific since they were up-to-date at the time. – Aske B. Oct 4 '12 at 17:10

You must log in to answer this question.

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