[wg11] "final" schemas and empty extensible selects

[Ed Barkmeyer]

It occurs to me that Part 11 Clause 8.4.2 apparently allows me to define an 
exchange schema in which an empty extensible select is never extended.  But 
then it seems also to be possible, per Clause 11.4.2, to "prune" an implicitly 
interfaced SELECT data type to empty, by not interfacing any of its domains.

The result might well be attributes that have no possible values.

The problem seems to be that EXPRESS does not have a notion of "final" schema 
= a schema intended to be the description of an actual population, as distinct 
from a schema whose intent is only to be "interfaced" into a "final" schema.
A schema intended to be "final" can also be interfaced into another "final" 
schema, but an "interface-only" schema cannot be the description of a population.

In a "final" schema (and only in a "final" schema), an empty select type 
should be invalid.

-Ed

-- 
Edward J. Barkmeyer                        Email: edbark at nist.gov
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263                Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263                FAX: +1 301-975-4482

"The opinions expressed above do not reflect consensus of NIST,
  and have not been reviewed by any Government authority."