I think in general Julia achieves a good (and difficult) balance between 
terseness, clarity and consistency.

That said, here is an example where she doesn't:

function_name
function_module
functionloc

One could say that "loc" is a very short word (and that it's clear enough 
to avoid the longer "location", that's fine), so it would be valid to 
remove the underscore and juxtapose it to "function", but consistency ends 
up completely broken then. It's disturbing to see this happening for one of 
three functions that form a cohesive set and are even listed consecutively 
in the official documentation. Please strive to avoid the R (and to some 
lesser extent, python) naming chaos. Maybe the coding convention rules are 
not clear enough. True, there will always be a place for subjetivity, but 
then let's try to find a rationale for degenerate cases like the above and 
codify it, or alternatively fix the violations while there is time, in 
order not to set a bad precedent. I don't want to exaggerate the importance 
of naming conventions, but consistency and terseness do reduce cognitive 
burden and, additionaly, make the language more attractive to newcomers 
initially affected by the Parkinson law of triviality, and more pleasant 
for the everday worker that focuses in getting the job done but 
nevertheless is a still a sensitive being ;) (a fortiori, this is specially 
true for Julia, considering that -by now- people are mostly courting her 
because of how beautiful she is).

I propose to scan the documentation and spotlight another cases like the 
above, and try to explain why they are what they are, or what they should 
be instead. Then we could add the list to an issue and also revisit the 
naming rules in the face of evidence.

Cheers
--
Carlos


Reply via email to