SparkR naming convention

[Dmitriy Selivanov]

Hi, SparkR developers. Thank for all your work. I'm newbie here, but google a little bit, before opening new thread. I'm surprised about zoo of name conventions used in SparkR - 

1. dot-separated function names (read.df, etc)
2. underscore-separated function names (group_by, etc)
3. camel-case-separated function names (groupBy, etc)

Of course, we can use all of these, but I suppose mixing of all of the above is not generally good idea. May be we should create general rules for that? I like Hadley's guide - http://r-pkgs.had.co.nz/style.html, but also I realise, that SparkR build on top of S4 classes and they usually use camelCase naming conventions.