TYPOL-Properties

[kuehne@isa.informatik.th-darmstadt.de (Thomas Kuehne)]

Hello Typol specialists,

I am aiming to write "linear" typol programs.
A typol program is linear if it is both
"link-less" and
"constraint-less"

These terms are defined in the article

"Functional evaluation of natural semantics specifications"
from
Isabelle ATTALI and Jacques CHAZARAIN
It is Deliverable D3.2 of Work package 1, Third Review Report, Janary 1991.

Now my questions:

Am i right to assume, that the rule

MapRest |- Pid -< Paccess : Result
--------------------------------------------------------------------------
pmap[paccess(pdef(proc_def(proc_header(Pid', _, _), _, _)), _).MapRest] |- 
  Pid -< Paccess : Result;
  provided diff(Pid', Pid);

does not violate the link-less condition, although Pid and Pid' are linked 
via the predicate "diff".
But Pid is not in the set of the input attribute and that is why the rule
is still link-less to my understanding.
Is that true?

However the same rule is not constraint-less because the Hypotheses is a 
structured term.
Do i get a linear typol program by transforming the rule into a rule with a
Variable as Hypotheses and adding another SET in which the Variable becomes
the subject?

In case I need stuctured terms in the Hypotheses and subject as well, like

false() |- op_name(A, B) -> A;
true() |- op_name(A, B) -> B;

is it a successful transformation to write

OTHER_SET(B,op_name(A, B))
----
B |- op_name(A, B);

OTHER_SET:

|- false(), op_name(A, B) -> A;
|- true(), op_name(A, B) -> B;

As far as i can see, the rules are now constrained-less.

Is there a particular reason for the strict formulation of 
constrained-lessness, if the transformation is so trivial?

My main subject is to claim my typol specification to be evalueable by 
a functional implementation of typol.
I guess there are no tools which can verify those properties?

Thank you

        Thomas

PS: I am sorry for my very short and dizzy questions, but i am in a terrible
    hurry. Next date i can write is the 5th of September and i wanted you to
    have that time in answering. Sorry again!