Re:Code Length Measured in 14 Languages
by Locke1689 on Wednesday November 14, 2012 @01:08PM
(#4784553)
No, he was right, I stopped reading when the author started discussing the mechanics of crawling web pages. I assumed that the blog post would be structured sensibly (analysis, then implementation).
You can be as terse or as verbose as you want in many languages (think forth, perl, apl).
If you want to do a meaningful comparison pick a non-toy problem and write a full sized application, something that in the worst language is a few thousand lines with a proper UI, a web app, a desktop application and a large simulation of something or other. You will still have to account for language gifts such as libraries or built-ins somehow.
And you tell the devs they get points for producing short code.
Then you can compare along several axis, performance, code maintainability (closely related to how easy it is to read) and so on.
That's useful data. This isn't, regardless of whether mathematica looks good.
edit: ok. point taken.