Talk:Welch's t-test: Difference between revisions

: I don't know what sort of official policies there are, but my take for Perl, which is intimately tied with CPAN, is that I like to see a version that uses no modules or only core modules (e.g. List::Util), followed by, if appropriate, versions using modules. The first version is useful in that it can run everywhere with no module installation, and it shows all the work. The versions using modules are one or more of faster, shorter, clearer, more functional, or more idiomatic. Especially if the no-module version is very long, I think it's best to reverse the order -- show the 2-10 line easy solutions, then the long "if you want to write everything by hand:" method.

: It is also not unusual, especially as new tasks get more complex, to see the first code added to a page using a module. This lets us get a Perl solution up quickly and succinctly, and later someone can write a huge everything-by-hand version. I suppose there is another discussion of whether it would make sense to have a "common library" pointer so we don't have to keep pasting in the same helper functions on 10+ tasks, which would let one concentrate more on the task.

 

== New title ==