3
$\begingroup$

Check this question with over a hundred upvotes: If squaring a number means multiplying that number with itself then shouldn't taking square root of a number mean to divide a number by itself?

Notice that the asker simply asked the question, without writing a paragraph about the philosophy they had why this might be right, wrong, etc. The user showed no work, yet is extremely upvoted.

But, then I see questions all the time, where someone has written up a question in a similarly simple form, maybe saying something like "I tried method A and B, but those did not give a tight enough bound. Does anyone know how I might show this?" And there will inevitbly be some 10K+ gold member in the comments saying "Downvoted as you didn't show any work", Show work, Give context etc.

I understand that the standards say to show some work, but I would rather a clean question that doesn't have shitty work under it. Imagine the square root guy from above wrote paragraphs on why it could or could not be that thing. That would make for a much worse question, but I think it is actually more in line with the "rules".

So should we actually show work if it makes the question worse?

$\endgroup$
15
  • 7
    $\begingroup$ That's from $2016$. Site standards have changed over time. $\endgroup$
    – lulu
    Commented Mar 27, 2023 at 23:29
  • 2
    $\begingroup$ Gotcha, not sure who made that change then. Is it not the case that some questions are simply better without a laundry list of potentially wrong methods the OP tried? $\endgroup$
    – Snared
    Commented Mar 27, 2023 at 23:30
  • $\begingroup$ I like how you wrote $2016$ instead of 2016 lol. @lulu $\endgroup$
    – Snared
    Commented Mar 27, 2023 at 23:43
  • $\begingroup$ Force of habit...didn't even realize I had done that. $\endgroup$
    – lulu
    Commented Mar 27, 2023 at 23:44
  • 2
    $\begingroup$ To your question, I can certainly imagine questions that have value but which wouldn't benefit from "work". Translation questions, say. "I came across this mathematical phrase in German and I can't sort out exactly what it means". No need to see bad translations, surely. But context still matters. Provide the citation, describe the subject matter and the historical period (if relevant). $\endgroup$
    – lulu
    Commented Mar 27, 2023 at 23:46
  • 3
    $\begingroup$ Note also that the question was tweeted by SE (and possibly also a hot network question) so it likely received many votes from users not very familiar with this site and its standards. $\endgroup$ Commented Mar 28, 2023 at 0:00
  • 2
    $\begingroup$ That post (as of now) has 156 upvotes and 38 downvotes. There are obviously some disagreement on the quality of that post. $\endgroup$ Commented Mar 28, 2023 at 0:00
  • 1
    $\begingroup$ If you have no clue how to proceed then it is much better to include other context rather than attempts that may well lead nowhere (and possibly mislead answerers into believing that you wish to consider only methods based on your attempt). But always be sure to search for related questions first (these can help you supply context, or possibly already answer your question). $\endgroup$ Commented Mar 28, 2023 at 0:11
  • 7
    $\begingroup$ Note that "showing work" has never been a requirement (at least, not if "showing work" is understood to be the high school notion of demonstrating what steps have been performed in a computation). Rather, the requirement is that questions must include relevant context. There are other (better?) ways of including context---"showing work" is only one form of context. That being said, the standards have evolved over time, and there are plenty of old questions which wouldn't stand up today. $\endgroup$
    – Xander Henderson Mod
    Commented Mar 28, 2023 at 0:46
  • 3
    $\begingroup$ Note, also, that this is kind of expected, as most of the basic, foundational questions have already been asked-and-answered, hence a novel question is likely going to require much more context in order to justify its inclusion in the database. $\endgroup$
    – Xander Henderson Mod
    Commented Mar 28, 2023 at 0:47
  • 1
    $\begingroup$ " would rather a clean question that doesn't have shitty work under it." -- To address this bit in particular: there is no reason to assume that the work cannot be salvaged, and I feel like this much should be obvious to anyone who has done some amount of math. A lot of "I tried X" posts that don't include the X always come off as half-hearted attempts with no serious effort or thought behind them, or worse, lying. But maybe I'm jaded. As for whether it improves the aesthetics of the question, I don't know; that I imagine is context-dependent. $\endgroup$ Commented Mar 28, 2023 at 7:31
  • 1
    $\begingroup$ Context (work, for example) helps to give better answers, and it is good in any case. Context is necessary for basic questions, otherwise it is not clear what answer is expected and highly likely without context the question will be downvoted and closed. $\endgroup$
    – kludg
    Commented Mar 28, 2023 at 10:50
  • $\begingroup$ There are at least two situations (other than solution-verification questions) where showing work is the best form of context (1) When the OP was instructed to use a certain proof method (e.g. use Gaussian elimination to find solution to matrix equation) (2) When one can use OP effort to probe the question (for example, a well-drawn diagram for a geometry question, or well-written code for a conjecture on the convergence of some sequence). In many other cases, showing work will be inferior to some other form of context and we should find a way of encouraging those other forms. $\endgroup$ Commented Mar 29, 2023 at 3:52
  • $\begingroup$ @SarveshRavichandranIyer you should post this comment as an answer $\endgroup$
    – Snared
    Commented Mar 29, 2023 at 4:20
  • $\begingroup$ I will expand on the comment in an answer then. $\endgroup$ Commented Mar 29, 2023 at 5:42

1 Answer 1

7
$\begingroup$

Like was said in the comments, what really helps a question prosper is context. At least on MSE, context serves two purposes :

  • It localizes or generalizes the question at hand appropriately by placing it in a scenario where it is useful to future visitors as well.

  • It provides tools that answerers can use to attack the questions, and helps them fit into the shoes of the asker (and future askers) so that they can understand the problem in a more nuanced way and provide a more high-quality answer.


Let's think of what the various forms of context present in the help page do. A source is an excellent reference for all the background material that an answerer might need. Background and motivation are localizers/generalizers of the question, because knowing where it naturally came from would mean that we'd need to tackle it from that point of view, or with the tools presented from that point of view. Ditto definitions which provide basic tools to understand and then attack the question.

What does effort , or work do? Ideally, effort/work can do the following : OP follows a particular line of reasoning, and expects the others to finish off with the tools they used : that is context, even if it isn't explicitly stated. Similarly, seeing the nature of the work, one can glean the OP's background , at least somewhat.

The problem comes when effort doesn't come off as any of these : it appears to be a last-resort attempt hoping that someone can feed off what's been written. It comes off as directionless and a way to circumvent the context requirement. That doesn't contribute to the question, which is then effectively devoid of context.

There are two very clear-cut cases (other than solution-verification questions) in which I believe that effort is the best form of context.

  • When the OP was instructed to use a certain proof method, and uses it in their effort. Once the OP makes it clear that the proof method was part of instruction, then the effort can be seen as part of the tool needed to attack the question.

  • When one can use OP effort to probe a question (for example, a well-drawn diagram for a geometry question, or well-written code for a conjecture on the convergence/divergence of some sequence). This kind of effort is extremely helpful because experts may not be able to code or draw the best diagram in one go.


What about the question given in the original post? It's really awkward because it doesn't seem like one can add any context. There are arguments that it results from a thought process that is rudimentary, yet natural : hence the up votes. On the other hand, one can also argue that the topic isn't worth debating or that the thinking preceding the question is shallow (see the first comment under the question) , hence the down votes. There's a reason the question split so much opinion, with 39 downvotes to 156 up votes. It did not receive enough close votes within three days of posting it (it may have got some which aged away) and was eventually protected because someone felt it had useful answers.

There is some context that could have been added but I'm not sure that it would have made the question any better. This one is a hair-splitter as far as I'm concerned, I'd probably leave it alone in a review queue.


Having said that, there is an overemphasis on effort/work since day one and that occurs when answerers just want to make sure that the poster isn't making others do their work and has done some work of their own. I call this an overemphasis because, in my opinion, looking merely for effort as a honesty certificate lowers the quality of questions. One should aim for a slightly higher ceiling, namely an effort to advertise the question and make it interesting and attractive to answer, which is one of the roles that context should play.

There are two problems with overemphasis :

  • Effort is presented when , in fact, something better should have been presented as context.

  • Effort is demanded even when it cannot visibly be presented, and the demand is irrationally forced in the comments.

An example of the first question is when people attempt to solve contest questions using token amounts of effort. Most of that effort could be in vain, just give me a list of what you think are similar questions and a source to ensure that it isn't from an ongoing contest, and that's good enough.

An example of the second kind is the situation where one does exactly what I say above for a contest question , only to be greeted with "what have you tried?" which isn't required because there is alternate context. Another example is when people are asking for clarifications (what does author mean in ... part of ... book?), asking for effort is clearly not possible.

I would seriously ask users to consider de-emphasizing effort as context and asking them to mention actually what is useful : a source, a list of similar questions from a search, a motivation of any sort, background level (e.g. for the question in the original post, "I'm a grade 6 student" may have been helpful) are all useful and very much so. In fact, I see users that don't mention these because they believe they could be extraneous, but they are far more useful than effort on many occasions.

My conclusion would be : there are some clear situations where effort is the best form of context, but I sincerely believe that it is overemphasized as context over the other forms, and would like to reduce its demand and instead increase the demand for other forms of context. On the question in the original post, I find it difficult to comment on because it is a real outlier, and would skip it if it came in my review queue tomorrow.

$\endgroup$

You must log in to answer this question.

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