5 ms·
The problem for me is that generally the highest ranked answers are the oldest, and in a fast moving tech stack this leads to really outdated advice. I do a lot
by nickserv 5y ago
The problem for me is that generally the highest ranked answers are the oldest, and in a fast moving tech stack this leads to really outdated advice.
I do a lot of Python programming and some of the answers only work in 2.7 for example.
Same as you, these days I'll just look at the official docs.
- armchairhacker 5y agoYeah I see that too. Back when python 2.7 was out, the top-rated answer was a really great, python 2.7-specific solution. Now that python 3.9 is out, the top-rated answer is still a really great, python 2.7-specific solution.
- LanceH 5y agoThe painful thing is there seems to be a large body of people on SO who are trying their hardest to fit any question into the scope of a previously asked question and then marking it duplicate/downvoting it. Some tags are worse than others on this about this, but I have one or two which are painfully policed by people who seem to want no question to be answered at all, ever.
- jayd16 5y agoJust take the extra 20 seconds and read through all the positive answers and commentary. Its not like you need to only look at the top ranked answer.
- codegeek 5y agoThat depends. If it is a popular answer, people sometimes edit with latest version answer. For example, I answered something about Python/Flask many years ago which became a very popular answer at the time (simple stuff). Now in Flask's latest version, you don't need to do it that way necessarily and the answer was edited accordingly.
- dotancohen 5y agoGit is the poster child for this. Ignore any accepted answer to a Git question on Stack Overflow if it is older than two years old. This may also be a testament to Git's terrific development, or to Git's obtuse early versions.