Over 54,000 questions with no answers!

Question

We have over 54,000 questions with no answers, this feels somewhat like a concern to me.

Some other ideas than just answering include:

• Give bounties to the highest voted questions, helping them get answered.

• Spit through the negative and zero vote questions at the end, either ...

  • ... get rid of crap by closing questions that don't fit Super User.

  • ... upvote and edit questions to fit Super User more, the latter bumps it so people see the Q.

So, with this meta discussion I would like to call to action to find proper ways to lower this number and to actually get this number down together. It's as easy as clicking Unanswered on the front page, then no answers and working through that list from front to end, or from end to front, or going to a random page number and starting from there.

For those bored or burning out from reviewing, and other enthusiasts...

Let's get some more questions answered!

Answer 1

Very few of the unanswered questions are ones that just need an edit bump or a bounty to get them answered.

• We should certainly do that when it will help, but that won't make a big dent.

Ones that should have been closed for the normal reasons sometimes don't gain closure traction before the action expires. If it's an old question, current standards may now make it more clearly close-worthy.

• Try closing again.
• If you're doing some kind of systematic cleanup, it wouldn't hurt to post a list of candidates on Root Access or Meta to get some concerted attention.

Many old, unanswered questions are ones that, by their nature, have almost certainly long-since been overtaken by events. If the OP is still around:

• post a comment inquiring about the status, and suggesting they post an answer clarifying what their solution or disposition was, or at least closing the loop on the question. Or, if they are still looking for an answer, to update the question, which will bump it.

A substantial portion is questions that require, or to avoid broad answers would benefit from, clarification from the OP. However, they have abandoned the question (posted long ago and they haven't returned since). Variation on the theme: a very localized question not likely to be of general use, and that's been abandoned or likely overtaken by events. Just a close vote now could fail to gain traction if the question is marginal and people don't check whether the OP is still active.

• Vote to close and add a comment, or use a custom close reason, to point out the abandonment issue.

A very substantial portion is questions where a suggestion in the comments was the solution or hint the OP needed. Variation on the theme: the OP posted a comment saying the problem disappeared or had been solved.

• If there is enough information to turn it into an answer, post an answer and mention that the OP indicated in the comments that that was the solution.
• If an actual solution can't be posted as an answer, close it with the custom reason that the OP stated there was no longer a problem to solve (or that it can no longer be replicated).

Answer 2

It's always a fact of life that there are more questions than answers. This is true of all SE sites, and all questions that have ever been asked in general, even questions unrelated to technology long before the Internet was invented.

That's because answers require a significant amount of work for humans (usually) to produce something useful and expedient to solve a problem; but in almost all cases where any "solution" is provided (as a result of someone doing work), the solution is imperfect, leading to yet more questions.

Just think about it: nearly all questions on Super User are about some piece of hardware or software that is supposed to present a solution to a problem. That is, for each "answer" we've already come up with, there are dozens (or hundreds) of questions! And the more complicated and comprehensive the answer (say, an operating system), the more questions there are!

Humans are never satisfied with what they've got, which means they can always conceive of how it'd be better. But in about 110,000 cases, at least, the following series of actions has not taken place, in the right order, to provide an answer:

• Someone asked a good question, or a bad question that nobody who cares about curating the site has bothered to flag or vote to close/delete.
• Someone who's had experience in that area has seen the question.
• That someone was either extraordinarily lucky in their ability to have already gone through the same problem, and can thus very quickly produce a solution; or
• That someone was lucky enough to type the right search terms into a search engine to find someone else on the Internet who has posted the solution to the same problem or a similar enough one that the answerer can put two and two together and write down what he/she thinks is a likely answer.
• The solution is simple enough that the answerer actually has time to do this, considering they probably work for a living (or do something else besides answer questions on Super User during their free time) and spend between 1/5 and 1/3 of their life sleeping.
• The answer wasn't spam that gets voted into oblivion or flagged and nuked.

In short, every time a question gets answered with valuable information, it's a freaking miracle. The scope of questions that can be asked is so broad that there's no way we'll ever have time to answer them all.

And because software becomes obsolete so fast, and peoples' use of software and hardware becomes obsolete so fast, many times the person who asked the question, as well as the rest of the community, will have moved on from the topic of the question, to the point where almost nobody still uses/cares about it. The question then becomes something of a permanent tumbleweed.

The system already has an automatic algorithm in place to delete many such questions, unless they receive enough upvotes that it decides to keep its automated hands off of it.

But for the rest, it's just a fact that we'll permanently accrete more and more unanswered questions. Welcome to humanity.

And no, this isn't a storage problem for Stack Exchange. Last I recall, the entire Q&A database for all of Stack Exchange compresses down to a handful of gigabytes -- it's some number much less than 50 GB. By the time this number becomes 1 TB, we'll have commodity 10 TB SSDs.

The funny part about human text is that it's extremely compressible, meaning that it's very inefficient and doesn't really say a whole lot that hasn't already been said. And compression algorithms are so good that for human text related to a particular domain (in this case, technology), the size of the compressed data might increase slower than linear (linear would be, doubling the input size also doubles the output's size, but I think the compression can do even better than that).