# openEHR-implementers Digest, Vol 50, Issue 2

**Category:** [Implementers (archive)](https://discourse.openehr.org/c/implementers-archive/158)
**Created:** 2011-07-01 15:26 UTC
**Views:** 1
**Replies:** 1
**URL:** https://discourse.openehr.org/t/openehr-implementers-digest-vol-50-issue-2/14088

---

## Post #1 by @Justinas_Prelgauskas

Hello, Thomas,

We could use "unknown" value for all NodeIDs that are not stated explicitly,
however, this brings more issues:
1) How would we deal with C_Object.NodeID validation requirement "Each node_id must be
defined in the archetype ontology as a term code" (AOM1.5.pdf, page 41)?
2) How would we conform to "NodeID" naming semantics, especially when specializing such "unknown" nodes?
It would be incorrect to name specialized node "unknown.1.1", because it would not conform to pattern "atNNNN {.0}* .N" (ADL1.5.pdf, page 106).
3) How would we support alternatives by using "unknown" NodeIDs?
4) How would we resolve such "unknown" ARCHETYPE_INTERNAL_REFs that have occurences of more than "1"? Compilation would result in duplicated paths in operational template.

Regards,
Justinas Prelgauskas

2011/6/2 <[openehr-implementers-request@openehr.org](mailto:openehr-implementers-request@openehr.org)>

---

## Post #2 by @thomas.beale

sorry for late response...


> Hello, Thomas,
>
> We could use "unknown" value for all NodeIDs that are not stated explicitly,
> however, this brings more issues:
> 1) How would we deal with C_Object.NodeID validation requirement "Each node_id must be
> defined in the archetype ontology as a term code" (AOM1.5.pdf, page 41)?

this is node_ids that are of the form 'atNNNN' and 'acNNNN'

this test has been modified to read:

Semantic identifier of this node, used to distinguish sibling nodes of the same type. [Previously called ‘meaning’]. Each node_id must be defined in the archetype ontology as a term code, or else have the special Anonymous_node_id value.

This value is defined as:

const anonymous_node_id[1]: String = “unknown”

in the class C_OBJECT.

> 2) How would we conform to "NodeID" naming semantics, especially when specializing such "unknown" nodes?

'unknown' is just a placeholder for nodes with no logical node id; no rules apply to it

> It would be incorrect to name specialized node "unknown.1.1", because it would not conform to pattern "atNNNN {.0}* .N" (ADL1.5.pdf, page 106).

none of these rules apply to any node id other than the at- and ac-coded nodes.

> 3) How would we support alternatives by using "unknown" NodeIDs?
> 4) How would we resolve such "unknown" ARCHETYPE_INTERNAL_REFs that have occurences of more than "1"? Compilation would result in duplicated paths in operational template.

you can't create an archetype internal ref to a non-identified node (i.e. including any node with the 'unknown' identifier.

hope this clarifies.

- thomas

---


**Canonical:** https://discourse.openehr.org/t/openehr-implementers-digest-vol-50-issue-2/14088
**Original content:** https://discourse.openehr.org/t/openehr-implementers-digest-vol-50-issue-2/14088