Why I don't like coffeescript

rant.coffee

#Given a function that accepts a callback...

someFunc = (callback) ->
  #do stuff
  callback(data) if callback

#when calling it...

someFunc (data) ->
  #blah

#it's not clear until you see the '->' symbol that (data)
# is the list of arguments to the anonymous callback function
# rather than an argument to the someFunc() call. And even
# after you see it, there's only a single character difference
# between calling someFunc with a function and declaring a
# function called someFunc which accepts a function.

You've used parens on:

callback(data)

why not

someFunc( (data) -> .. )

Do you often write functions that take a single callback as the only argument? This doesn't read so bad:

someFunc opts, (data) -> ..

There's also the confusion created by picking nondescript names for stuff, which shouldn't happen in practice.

I'm trying to reduce a general difficulty I have with reading (other people's) Coffeescript to its simplest form. I agree that putting parens around the call makes it (slightly) easier to read.

I'm looking at code that is callback heavy (probably too callback-heavy, but that's another story...) and I still find the 'arguments before function symbol' syntax highly confusing when nested inside a call to another function:

```
http.createServer( (req, res) ->
  key = url.parse(req.url, true).query.id
  database.open((err, connected) ->
    retrieveUri(key, connected, (uri, connected) -> 
    #etc...