I was talking to someone today about 'accents' in programming, where you write code in one language in an unidiomatic style strongly influenced by being more familiar with another language.

If the language you're writing is the more powerful or featureful of the two, a common case of this is that you (mostly) stick to the subset that matches features in your old language, because you write FooLang by imagining a BarLang program in your head and just superficially changing the syntax. Like writing a loop in Python when you could use a list comprehension. Or open-coding some standard algorithm in C++ where you could just say std::do_it_for_me, because you're used to having to do it yourself in C.

Another case is where you contort the language itself to look more like the familiar one. Like writing BarLang-ishly named wrappers on the FooLang library functions, or outright reimplementing your favourite BarLang function. (I often write a 'chomp' function in languages that aren't Perl.) Or once I saw someone start all their Python programs with 'from sys import argv, exit, stdout' so they could refer to those names the same way you would in C. A very extreme case of this is to contort the _syntax_, where you can get away with it – the notorious Bournegol.

Anyone got any other cases of this phenomenon which don't fall into those categories? I'm particularly interested to know if there are any recognisable tendencies you'd observe from someone coming to a _less_ powerful language from a more powerful one. Do Haskellers forced to write C (say) have any common habits that look like "best approximation to a monad I can bodge together", for example?