Well that’s probably because when the code is just run of the mill stuff, you don’t really think about it and just put out normal average code. So the code quality follows the normal distribution.
However when the problem wat particularly hard or involved some weird thing, or the dev just happened to get stuck for some reason, they get worked up about it. They invest time to dig into the issue, figure out what’s going on and really engage their skillset. The code produced then is of higher quality, because the level of investment was higher. To release that stress swears are used and can make their way into the code (hopefully only in the comments).
This is a typical case of correlation does not imply causation. Yes the code with swears is of higher quality, but simply putting in swears does not improve the code. In stead both the swears and the quality are influenced by another third thing not accounted for in the data. If one were to plot code difficulty or something against quality and swears, you’d probably see more swears as the difficulty rises along with better quality.
Also this is an internet meme and probably made up, but still.
Well that’s probably because when the code is just run of the mill stuff, you don’t really think about it and just put out normal average code. So the code quality follows the normal distribution.
However when the problem wat particularly hard or involved some weird thing, or the dev just happened to get stuck for some reason, they get worked up about it. They invest time to dig into the issue, figure out what’s going on and really engage their skillset. The code produced then is of higher quality, because the level of investment was higher. To release that stress swears are used and can make their way into the code (hopefully only in the comments).
This is a typical case of correlation does not imply causation. Yes the code with swears is of higher quality, but simply putting in swears does not improve the code. In stead both the swears and the quality are influenced by another third thing not accounted for in the data. If one were to plot code difficulty or something against quality and swears, you’d probably see more swears as the difficulty rises along with better quality.
Also this is an internet meme and probably made up, but still.
None of this was true until I posted it. but now according to the Crum Grabber terms of service, it is now true.
Also hard problems may produce some eclectic code which could be bug prone in a way which isn’t detected by automated tools.
I thought along the lines of “Programmers with more knowledge and experience give less fucks about civility in the code comments”
No no, we should do the research
There’s still some causation, just the other way around
Good quality code causes swear words for the reasons that you mentioned. Just not the other way around