I have heard there will be a new appendix [S] in the next B31.3 issue and that it relates to computer analysis.
Does anyone have some advance information on it's likely content and implications?

Task Group B of the B31.3 Section Committee has been working on this Appendix S for quite a while. The goal is to put together several piping models that address analysis situations that are not clearly defined by the code. Two that come to mind are 1) sustained stresses with nonlinear supports (e.g., +Y) and 2) the proper expansion stress range calculation arising from moment reversal in a system with multiple thermal sets (e.g., hot leg/cold leg reversal at a tee).
In our attempt to put together these examples, we found we had no base line to start with. One obvious issue is the formulation of sustained stress. It is difficult to get committee approval on these "more advanced" issues when the basics are not obvious.
The first pass, then, of Appendix S is a simple "Z bend" that passes both the expansion and sustained stress checks. It also touches on the issue of design vs. operating conditions as it affects wall thickness calculations and system analysis. It also includes a minimum temperature case which increases the expansion stress range.
I don't see any reason why this Appendix will not go out in the 2004 Edition of B31.3. The more interesting work will appear in later editions.

Does this mean that B31.3 is going to start evolving into a design manual? I'm surprised the code committee wants to get anywhere near making actual design recommendations like that.