I just caught myself doing what I do a lot, and wanted to summarize it, express it, share it and see who else is following this general practice to find other examples of situations where this may be relevant.
In general, the practice is to do something wrong first, in order to establish that everything else is right before the current task.
What I was trying to do, in particular, was to find examples in our code base where the dojo TextArea widget was used. I knew (because I had it in front of me - proof of existence) that the TextBox widget was present in at least one file. So I first looked at what I knew:
grep -r digit.form.TextBox | grep -v CBN
This was wrong: I made a general mistake (for me) by leaving a star, so I fixed this:
grep -r digit.form.TextBox * | Grep -v svn
who did not find any results! A quick comparison with the file I was looking at showed me that I mistakenly wrote "dijit":
grep -r dijit.form.TextBox * | Grep -v svn
And now I got the results. Cool; doing it wrong at first for the purpose meant that my request was right, except for finding the wrong thing, so now I could build the correct request:
grep -r dijit.form.TextArea * | Grep -v svn
and be sure that when this did not give me any results, it was because there were no such files, and not because I distorted the request.
I will add three other examples as answers; add others you know about.