Why precision pays off

The person reading your question cannot see your notebook, your textbook or the last twenty minutes of your thinking. Everything they need has to be in the question. A vague question forces them to guess, and the first reply is usually a request for the information you could have included.

There is a second, less obvious benefit. Writing a precise question is itself a problem-solving step. Stating exactly where an argument breaks often reveals why it breaks, and carefully written questions are often answered by their own authors before anyone else reads them.

The five parts of a good question

  1. ContextSay what course or text the problem comes from, which methods you are expected to use, and what level of rigor is wanted. A first-year integral and a measure-theory exercise can look identical on the page.
  2. The goalState the problem exactly, with every condition: domains, initial or boundary values, parameter ranges. Copy it rather than paraphrasing it from memory.
  3. What you triedShow the method you used and the actual lines you wrote, not a description of them. If you tried two methods, show both.
  4. Where it breaksPoint to the first line you are not sure of, or say what result you expected and what you got instead. This is the most valuable sentence in the question.
  5. NotationDefine any symbol that is not completely standard and write formulas so they cannot be misread.

Reduce to a minimal example

If your difficulty appears inside a long computation, try to reproduce it in the smallest case that still fails. If an error appears when solving a fourth-order equation with three forcing terms, check whether it appears with one forcing term, then with a second-order equation. If a matrix calculation goes wrong for a 5×55\times 5 matrix, try 2×22\times 2.

A minimal example does three things. It makes the question short enough to read. It separates the real difficulty from arithmetic noise. And it often shows you the answer: if the small case works, the problem is somewhere in what you removed.

Before and after

Vague: "My Laplace transform of tsin⁡2tt\sin 2t is wrong. Can someone help?"

Precise: "I want L{tsin⁡2t}(s)\mathcal{L}\{t\sin 2t\}(s). I used the rule L{tf(t)}(s)=−F′(s)\mathcal{L}\{t f(t)\}(s) = -F'(s) with f(t)=sin⁡2tf(t) = \sin 2t, so F(s)=2/(s2+4)F(s) = 2/(s^2+4). I computed F′(s)=−4s/(s2+4)2F'(s) = -4s/(s^2+4)^2 and wrote that as the answer, but my table gives 4s/(s2+4)24s/(s^2+4)^2. Is the rule correct, and where does the sign come from?"

The precise version answers itself. The rule has a minus sign, and it was not applied: −F′(s)=4s/(s2+4)2-F'(s) = 4s/(s^2+4)^2. Writing out the rule next to the computation exposed the missing step. As a check, numerical integration of e−sttsin⁡2te^{-st}t\sin 2t at s=1s = 1 gives 0.16=4/250.16 = 4/25, which matches the table.

A question that pinpoints a method failure

Vague: "Undetermined coefficients doesn't work for y′′+y=sin⁡xy'' + y = \sin x." Precise: "For y′′+y=sin⁡xy'' + y = \sin x I tried yp=Asin⁡x+Bcos⁡xy_p = A\sin x + B\cos x. Substituting gives 0=sin⁡x0 = \sin x, so no AA and BB work. I notice that sin⁡x\sin x also solves y′′+y=0y'' + y = 0. Is that the reason, and what trial function should I use?"

  1. The precise question already contains the key observation: the forcing sin⁡x\sin x is a solution of the homogeneous equation, so the standard trial is annihilated by D2+1D^2 + 1.
  2. The standard fix multiplies the trial by xx: yp=x(Asin⁡x+Bcos⁡x)y_p = x(A\sin x + B\cos x).
  3. Substituting gives yp′′+yp=2Acos⁡x−2Bsin⁡xy_p'' + y_p = 2A\cos x - 2B\sin x, which equals sin⁡x\sin x when A=0A = 0 and B=−12B = -\tfrac{1}{2}.
  4. Check by substitution: with yp=−x2cos⁡xy_p = -\tfrac{x}{2}\cos x, yp′′=sin⁡x+x2cos⁡xy_p'' = \sin x + \tfrac{x}{2}\cos x, so yp′′+yp=sin⁡xy_p'' + y_p = \sin x.
yp(x)=−x2cos⁡xy_p(x) = -\frac{x}{2}\cos x

Compare what each version allows. The vague one invites a general lecture on undetermined coefficients. The precise one can be answered in two lines, and the reply will address exactly the point the asker had reached.

If you want a template, this one covers the five parts in four sentences: "I am working on [the problem, with all conditions] from [course or text]. I tried [method], and my steps are below. At step [n] I get [result], but I expected [result] because [reason]. My question is [one sentence]." Fill every bracket; if one cannot be filled, that is usually the first thing to work out.

Write the mathematics so it can be read

  • Use LaTeX-style notation where the platform supports it, or unambiguous plain text: write sin⁡(2t)/(s2+4)\sin(2t)/(s^2+4), not "sin2t/s^2+4".
  • Define every symbol the first time it appears, including the variable of differentiation and the domain.
  • State parameter ranges. Many disagreements about an answer are really disagreements about whether a parameter is positive.
  • If you include a photo of handwritten work, also type the line where it breaks, so that the question is searchable and quotable.
  • Ask one question at a time. If you have three, post them separately or number them.

Bringing a question to a session or the community

The same structure makes every RIDA MATH channel more useful to you.

  • Booking a session. When you request private sessions or problem-solving support, include one or two questions written in this form. It lets the first session start at your actual sticking point instead of at a general review.
  • Posting in the community. The Problem desk channel works best with questions that state context, attempt and breaking point. Others can then answer the question you asked, and later readers with the same difficulty can find it.
  • Reading the book or research. For questions about a text, give the section and equation number and quote the exact line. The Reading room is the place for these; reports of possible errors in the book can go through the contact form.

References

  1. George Pólya. How to Solve It. Princeton University Press, 1945. On restating the unknown, the data and the condition..
  2. Kevin Houston. How to Think Like a Mathematician. Cambridge University Press, 2009.