ISSUE: retroactivity of sub-string APS in WebCGM

[Lofton Henderson <lofton@rockynet.com>]

[...2nd of 2 new issues about sub-string APS in WebCGM...]

ISSUE:  Should the sub-string APS functionality that will be documented in 
WebCGM 2.1 also apply to 2.0 (and even 1.0)?

DISCUSSION:  It is a fact about ISO corrigenda that they are, after 
approval, considered to be part of the subject standard from Day 1 of that 
standard.  I.e., it is as if sub-string APS had been a clear explicit part 
of CGM:1999 since 1999.

Therefore, the question arises:  are sub-string APS legal in the 1.0 and 
2.0 profiles?  Presently, the text of 1.0 and the text of 2.0 do not 
mention sub-string APS, and do not place any restrictions on them.  But 
since they are a part of CGM:1999, and not restricted, it could be argued 
that they are legal in 1.0 and 2.0.

It is possible that we want them to be legal, but that would have 
conformance implications on all implementations.  And if we *did* want them 
to be legal, we would certainly want to impose whatever profile 
restrictions we decide for 2.1 [1].  Therefore, if we did want them to be 
legal, we would need an erratum to 1.0 and an erratum to 2.0.

[1] http://lists.oasis-open.org/archives/cgmo-webcgm/200801/msg00038.html

In fact, if we do *not* want them to be legal in 1.0 and 2.0, we should 
probably still have an erratum to explicitly prohibit them, now that the 
CGM:1999 standard has been clarified.  This would be non-substantive -- no 
implementors thought sub-string APS were possible in 1.0/2.0 -- and a 
low-priority project.

OPTIONS:  (Chose one or more of the following:)

1.) legal in 2.0, clarify with 2.0 erratum
2.) legal in 1.0, clarify with 1.0 erratum
3.) not legal in 2.0, clarify with 2.0 erratum
4.) not legal in 1.0, clarify with 1.0 erratum
5.) other

RECOMMENDATION:  #3 and #4.

Regards,
-Lofton.