5 ms·
StackOverflow being a wiki is at substantial odds with the way answers are selected. If StackOverflow is a wiki, then the best answer to a question should be d
by kauffj 8y ago
StackOverflow being a wiki is at substantial odds with the way answers are selected.
If StackOverflow is a wiki, then the best answer to a question should be determined by the community, not the asker, and needs to be able to evolve and change over time.
However, StackOverflow simultaneously insists questions can't be duplicated while making answer selection immutable and controlled by the asker. This results in a suboptimal experience.
Here is an example question highlighting this: https://stackoverflow.com/questions/884177/how-can-i-determine-what-font-a-browser-is-actually-using-to-render-some-text https://stackoverflow.com/questions/884177/how-can-i-determi...
I currently have the best answer to the question "How can I determine what font a browser is actually using to render some text?". The answer I gave was the best answer in 2010. Things have changed a lot since then, and my answer is no longer the best one on that page.
_But no one can change the best answer, which is insane!_ As a result, you see best answers to old questions being turned into community wikis edited to include better answers in the thread, rather than those answers being selected.
- coldtea 8y ago>If StackOverflow is a wiki, then the best answer to a question should be determined by the community, not the asker. You're thinking of wikipedia. A wiki doesn't determine what the "best content" in it is. It's just a collaborative way to edit text together.
- mademechuckle 8y agolol "You're thinking of" I dont know what they were thinking.. but in any wiki I have used anyone can make corrections and there is a shared responsibility to keep it up to date.. it seems to me that the statement is therefor valid - the basics of what is or is not a good answer is set by the asker not the community, which is how ticketing / Q&A systems work NOT wikis IMHO.
- Ajedi32 8y agoSolutions to this problem have been suggested multiple times[1][2][3][4] by the community and, unfortunately, were mostly ignored. [1]: https://meta.stackoverflow.com/q/326095/1157054 https://meta.stackoverflow.com/q/326095/1157054 [2]: https://meta.stackoverflow.com/q/283456/1157054 https://meta.stackoverflow.com/q/283456/1157054 [3]: https://meta.stackoverflow.com/q/272651/1157054 https://meta.stackoverflow.com/q/272651/1157054 [4]: https://meta.stackoverflow.com/q/253752/1157054 https://meta.stackoverflow.com/q/253752/1157054
- codinghorror 8y agoThat's mostly a perk for the question asker. Nobody (that I know of) stops reading at the first answer, because there are usually multiple good ways to solve any problem in programming. And if the accepted answer has a fraction of the upvotes of the answer just under it.. that's another strong signal to "keep scrolling"
- dantiberian 8y agoPeople definitely stop reading after the first answer, especially if it looks like a reasonable solution. It is a very common experience for me, where I won't realise that there is a much better rated answer below the accepted one, especially if the top answer is long enough so that the second one isn't visible without scrolling. From a UX perspective, you want to put the most relevant information at the top. The votes of all Stack Overflow users gives a far greater signal than the one 'accepted' flag from the asker. This is one of my biggest pain points with Stack Overflow.
- codinghorror 8y agoThere's zero evidence that people stop reading at the first answer. Otherwise no other answers would have any upvotes, and you can see that's not the case.
- dantiberian 8y agoI'm not saying that every single person stops reading at the first answer, just that people drop off over the page, and in my personal experience (and talking with others), it is common to miss that there are higher ranked answers below the accepted one.
- Flimm 8y agoWhat's the value in pinning the OP's selected answer to the top, even if it has -100 votes? Or even if it 100 votes, but the second has 900? Shouldn't the overwhelming will of the community take precedence over OP's selected answer, who may not have bothered to review the answers for years? Also, I'm sure the claim isn't that everybody stops reading after the first answer, but that the first answer does get a lot more reads than answers further down. I bet it would be possible to prove this with the right analytics.