Re: Temporal validitity of subject indicators?

On 05/11/2011 03:09 AM, Robert Cerny wrote:
> A very good advice in the light of this issue was given years ago on 
> this mailing list by Bernard Vatant: if you are not a 100% sure that 
> two things are the same, make them different. That is why i was a 
> little surprised to see the role types merged with the topic types in 
> the last draft of TMCL metaschema.

My 0.02 Euros: It can bite you either way.  Everything matters.

In my own mind, "topic type" is identical to "subject type".  I'm under 
the impression that some in our community feel differently about that, 
for reasons I don't really understand.  I'm not worried about it because 
there are many universes.

But among those who agree that

* "topic type" === "subject type", and that

* a role type is a subject, and that

* topic types are subjects,

then :

the set of all role types must be a subset of the set of all topic 
types.  The two sets are not the same sets, but any role type must also 
be a topic type.

Steve N.