This is so far the best and the most concise comment in favor of code comments I've read so far.
Thanks man. teisipäev, 30. aprill 2013 21:16.39 UTC+3 kirjutas Vince O'Sullivan: > > On Monday, 15 April 2013 07:38:03 UTC+1, brucechapman wrote: > >> If what we write first is "the simplist thing that might work", then >> I'd suggest comments should explain code that is not apparently the "the >> simplist thing that might work". or "comments should explain why the >> simplistic thing that might have worked, didn't" >> > > Unfortunately, code that the developer has stripped bare, in order to make > it more "simple", is rarely code in a form that is most useful to someone > who has to maintain it in the future (including when that someone is the > same person and "the future" might only be days later). > > Comments are (or, at least, ought to be) good precisely because they are > redundant. Error checking is only possible when redundancy is present. If > the comments and the code match, our confidence in it increases; if not, it > decreases. Unit tests work in the same way. They are redundant code - > often bigger and more complex than the code under test - in that they are > written but not shipped (just as comments are written but not executed). > Nevertheless, no one would suggest that removing tests would be a good > thing because it would make the overall code base simpler. > -- You received this message because you are subscribed to the Google Groups "Java Posse" group. To unsubscribe from this group and stop receiving emails from it, send an email to javaposse+unsubscr...@googlegroups.com. To post to this group, send email to javaposse@googlegroups.com. Visit this group at http://groups.google.com/group/javaposse?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
