[Nml-wg] References

[Freek Dijkstra]

Hi Aaron,

Thanks for your mail. I think I see your point, but I want it to be
clear (I struggled with some text in the document, and am not happy
yet). So I'm going to be devil's advocate for the next two mails.

> The problem is ambiguity. If I can define a new, named object inside of
> a relation element, how does a parser know that the named object inside
> the relationship is defined there, as opposed to elsewhere?

Why does a parser need to know know that the named object inside the
relationship is defined there, as opposed to elsewhere?

E.g. why can't it just create the object, even though the object may be
incomplete, or the object may be wrong.

If the rest of the definition is elsewhere in the topology description,
Can't it augment the created object with that information when it
encounters it later in the file? If not, why not?

Just mention one or two reasons, no need for details. I'll put in my
next question based on your answer.


(You explicitly mentioned typo detection as one reason, and I previously
mentioned that it is useful to distinguish between full object
descriptions and mere object references. Are these the two reasons? Are
they both valid enough to define a relation or parameter?)

Freek