Rich, All,
I think the pseudo code is fine as well, so either way is good for me.
But I don't understand all your objections in your email.
The proposed description is not written in a formal logic, it's written
in English, which I think is fine for this purpose. I don't think there
is any ambiguity in the proposed text. Perhaps it's missing the implied
step "0. Evaluate all rules". The remaining steps refer to the results
from this initial step.
And the pseudo code is not formal logic either, so it is as ambiguous as
the natural language text.
In any case, we need to change the specification in some manner because:
1. There are currently two descriptions for each combining algorithm,
one in English and one in pseudo code, and no statement of which of them
is the normative one.
2. The natural language description is currently not correct. It does
not cover all possible cases.
So we need to fix the natural language description and then choose
either it or the pseudo code as the normative one. I don't mind which
one, as long as they are both right. :-)
Best regards,
Erik
Rich.Levinson wrote:
