OpenACS 4.x is absolutely moving forward. If/When ACS/Tcl stops, OpenACS/Tcl will continue. We have no reason to switch to Java at this point. More importantly, if OpenACS/Java does come along, I don't foresee OpenACS/Tcl disappearing. There is a large community behind (Open)ACS/Tcl, and we intend to keep that community strong.
OpenACS 4.x core is meant to be a solid base. We don't *want* to change the architecture too often, but we will certainly maintain it, improve it, patch it. The great thing about 4.x is that anyone can develop packages. We will not only allow this, we will encourage it. Maybe we'll host the CVS repositories for these packages (a-la SourceForge). Maybe we'll provide rotating ads on OpenACS.org for these user-contributed packages. Whatever we come up with to grow the amount of available software on OpenACS, we will do. Because our goal is not to put out a monolithic package. Our goal is to define a solid architecture on which the community's creative ideas can build.
Rest assured. OpenACS/Tcl is here to stay.