For the most part, I agree with Cedric:
One of the main reasons why people like dynamically typed languages is precisely because they don’t have to specify these types. Is it really so shocking that this convenience comes at the price of not being able to perform any operation that requires type information on this code?
This problem has always reminded me of the ‘Help me help you!’ scene in Jerry Maguire – if you don’t tell your IDE your intentions, then there is always going to be limits to what your IDE is going to be able to do for you.
Having said that, it took a couple of years of JBuilder, Netbeans, and Visual Age before we got the great refactoring capabilities of Eclipse and IntelliJ. I would not be at all surprised if someone out there figures out a good solution to this problem.
{ 1 comment… read it below or add one }
And still great projects evolve using dynamic typed languages. Of course the support of an IDE is absolutely necessary (and I still miss the good old Java features in Radrails) but still, it is a different concept. Especially in Ruby, the intention must be seen in the code and not from a type hierarchy.
If you ever had a look at Eclipse Plug-in development then, you will notiice that the availability of interfaces, abstract classes and types makes people crazy. They will introduce new stuff every five minutes and in the end they call it a new programming paradigm.
But one important thing left from my point of view: Good code does not derive from a good IDE.