Re: async I/O API: why strings?

[Krux02]

I would like to throw in my two cents. I think that passing one block of memory 
(one argument) as two incomplete arguments (pointer and size, begin and end 
iterator, ) is bad API design. It has been introduces by C because the 
languages with it's pointer arithmetic encourages to do so. C++ is now going 
some hard road to introduce a range type into a language that inherently always 
used begin/end iterators, because they learned. We should not adopt the bad 
parts that came from C. So as long as we don't have a proper range type in Nim, 
I would suggest to just support the `seq[byte]` type.