Mike Cargal wrote:
> Of course, the example also proves the point that statement 
 > terminators facilitate better reporting of these types of errors.

That they do, but one of the appeals of Ruby at least to me is that the 
parser / compiler doesn't force you to take precautions against errors 
that you think don't bite you.

And this thread has had quite a few people say that they think this 
problem is at least for them a non-issue for various reasons - like 
typing empty parens or def/end pairs and then filling them in, and / or 
adhering to fine-grained modularisation with very short source files 
that make syntax errors like this that slip fairly easy to find.

The fact Ruby allows for syntax that is problematic to analyse when 
faulty might seem newbie-unfriendly, but a lot of those syntax features 
are on the other hand what people with some experience like and prefer. 
Making statement terminators compulsory would be facilitating adoption 
of the language at the expense of taking away a positive quality of the 
language I consider unacceptable.

David Vallner