Re: "dict"ionary typing?

I'm eager for dictionaries to be a great feature in SWI,

including highly usable syntax.  Thus I'm pressing for 
improvements before things get locked in.  Is anyone else 
concerned about the proposed syntax?

I suppose things like this are the primary motive for 
the new syntax.  

M.polar_radius() := D :-
           D is sqrt(M.x*M.x + M.y*M.y).

?- P = point{x:3, y:4}, X = P.x, D = P.polar_radius().



However, I'm struggling to accept this.  I feel it confuses 
the standard meaning of =/2 as "unify terms".  Here the right 
side feels like it ought to be a term but I gather it isn't.
(And if it isn't a term, it can't be meta-programmed?)
Newcomers will think functional notation and be baffled when 
P = 2+3 doesn't work.

I might prefer P.polar_radius(D), and if that's sugar 
for polar_radius(P,D) this  is okay because I can still use 
X = P.polar_radius, call(X,D).  As is, even if you have some 
magic that makes all uses behave correctly, I will find it 
confusing.

I'm also skeptical about the usefulness of functional chaining 
like P.foo().bar() given the performance characteristics. How 
often should a function on a dictionary return another 
dictionary?


---

Different question: Can this work?



?- X = P.x, P = point{x:3, y:4}.


>> I guess there are two main uses for a dict:

>> a) as a struct where there are fixed elements of varying types.
>> b) as a map where there are variable elements of fixed types.

>Not sure I get that.

I mean this: 
a)window{top: 0,bottom: 100,left:2,right:12, background: red, controls: List}


b)dictionary{cat, 'small animal with whiskers', dog, 'domestic animal that barks'}