1468   Re-use DO from other LN

Created: 05 Jan 2016

Status: In Force (green)

Part: Part 7-1 (2011; Edition 2)

Links: #1312 Presence condition of lnNs and dataNs , #1196 Extensions to standardized LN classes made by third parties

Page:

Clause:

Paragraph: 14.3.2

Category: Issue may impact interoperability of implementations of Edition 2

Issue

As discussed at the UCAIug telco January 5 the namespace rules will change in Edition 2.1. This may lead to interoperability issue with the current name space rules in some cases. Especially the re-use of an DO defined in part 7-4 (e.g. Str) but re-used in another LN from part 7-4 (e.g. GGIO). In Ed2.0 the re-used DO may have a dataNs referring to the standard or the dataNs may be missing (inherited from lnNs). According to tissue 1196 (Ed2.1) the DO shall have a private dataNs value.

Proposal

To prevent interoperability issue we propose in Ed2.0 to also allow a private dataNs value for re-used DO.

Example: re-use the DO=Str in LN=GGIO. The GGIO.Str may have:
- no dataNs or
- a dataNs with a value referring to the standard or
- a dataNs with a private value (new)

Discussion Created Status
08 Mar 16 In Force (green)
No change to 7-1 is needed (except according to 1196 in Ed2.1). This Tissue is more testing related than standard related. 19 Jan 16 Ballot Period
The Tissue is accepted with the addition, that usage of a private name space in dataNs also needs a TICS reference to Tissue 1196.
This is common practice already now.
19 Jan 16 Ballot Period
At the moment, I tend to consider that namespace interoperability is not yet a big issue and recommend to accept the compromise submitted by the proposal. 15 Jan 16 Ballot Period
This Tissue is superfluous. An Ed2.0 device will behave as defined in Ed.2.0. If it shall use a dataNs with a private value, it can refer to Tissue 1196 in its TICS. This is by the way the preferred solution to avoid later IntOp problems. 13 Jan 16 Discussion (red)
I agree with the proposal, as it offers a good compromise to allow a transition from Ed2 server test procedure V1.0 to an more strict definition documented in #1196 and in 7-1 edition 2.1 regarding namespace handling and life cycle of products.
Ed2 client shall today not refuse to the operator the usage of Logical Node extensions as soon as the used common data class for the extension is supported. Extensions of a logical node are recognizable as they include any data object that was not present in the Logical Node definition exposed in lnNs (resp. ldNs if lnNs missing).
07 Jan 16 Discussion (red)

 

Privacy | Contact | Disclaimer

Tissue DB v. 21.10.16.1