One of the hazy issues hanging over the head of the EGL language is its long-term goal. Is EGL intended to be a replacement for RPG, or a complement to it? Like many things, I think it depends upon whom you talk to at IBM. One of the mistakes I've made in the past is to consider IBM as a monolithic entity with consistent goals, but that's far from the truth. In reality, there are factions inside of IBM just as there are in any large corporation, and sometimes the left hand and the right hand aren't quite in sync.
Personally, I can tell you that the EGL team has made significant investment in making sure that EGL works with the IBM i, and specifically with existing ILE code. The biggest example of that is the persistent connection, which allows us to take advantage of all the power of the ILE environment while still presenting multiple interfaces to the outside world.
But I think it's important that we talk about this. For your company, would it be more important that IBM focus on providing tools that allow EGL to act as a thin interface to your existing ILE, or are you more interested in using EGL to write brand new platform-independent applications?
Or is it a little bit of both?