Modification to Terminological Entry (ISO 19157-1:2023)

[ReesePlews]

in ISO 19157-1:2023 the authoritative source reference for feature type (3.16) is incorrect. [shown as ISO 19156:2011, 4.7]

the correct authoritative source reference is [ISO 19156:2023, 3.9]

[PeterParslow]

In my opinion, "more authoritative" should be the most fundamental standard that created the definition, not the most recently published - therefore ISO 19101-1:2014

I guess this recurring issue won't go away until we have ISO 19173

[ReesePlews]

thank you for the discussion here.

actually 19101-1 did not create the "feature type" definition, that was originally from 19156:2011. 19101-1 was published in 2014.

19101 does "own" the following "feature" related terminological definitions:

• feature
• feature attribute
• feature catalogue
• feature instance

and i do agree with @PeterParslow that these "ownership" issues will not go away until our SMART Terminology Register is running with all of our terminology in one, accessible place.

[PeterParslow]

Reese,
This seems to be an artefact of the way compound word terms are handled. The ISO 19101 definition of feature includes the note

"NOTE A feature may occur as a type or an instance. Feature type or feature instance shall be used when only one is
meant."

(And has done since ISO 19101:2002) so it states what the phrases "feature type" and "feature instance" mean.

I don't know why ISO 19156 felt the need to define "feature type" as a term in its own right when it is a combination of two previously defined terms (feature, in ISO 19101; type in UML & a range of other IT standards)

[ReesePlews]

thank you @PeterParslow that is very interesting and i see your point.

i tend not to manage the notes to entry as rigorously as i do the definitions. however, 19101 does define "feature instance" so perhaps 19101 should have defined "feature type" too, in the same way, instead of just leaving mention to it in a note.

i guess the only way to possibly shed light on why 19156 wanted to create this definition is to directly talk with the PL. not sure if they are on github.

if everything goes according to plan with 19173 all of the authoritative source references will be then come from that "project" (the SMART Terminology Register). however the "lineage/ownership" of the individual terminological entries will still probably be managed by a combination of the originating document project team with input from the TMG and other TC members. decoupling the terminology from the individual standards will enable use to make revisions to the individual terminological entries, outside of a document revision.