opoudjis
commented
3 years ago @skalee Note, this impacts you.
Closed opoudjis closed 3 years ago
opoudjis
commented
3 years ago @skalee Note, this impacts you.
Disambiguating information on designations is not currently captured in the concept model, and is clearly required in iev-document.
The reason for this is a confusion between ISO 10241-1:2011 and IEC/ISO DIR SUP IEC:2016.
ISO 10241-1:2011 defines domain, as subject matter. It never explicitly defines usage information, but it repeatedly gives illustrations of it. From 4.1.2d, 4.4.3.2.2, 5.3, 6.2.3, and A.1.3.2.4, "usage information" can be any of:
grammarInfo, pronunciation, language, script, geographicArea
It is also very clear that these attributes are attributes of designations, not concepts.
Exceptionally, 4.1.2d allows that usage-specific information can also include domain, which is currently only a concept attribute. It is explicitly spoken of as a designation attribute anyway, and one could argue that it is inherited from the concept.
4.4.3.2.2 allows "other secondary data categories defined in the ISO Data Category Registry (DCR) on the basis of ISO 12620".
In 3.4.1.1.3, usage information is appealed to to disambiguate symbols. That does not say what usage information is, but it again makes it an attribute of a designation.
ISO 10241-1 states that IEC uses usage information in relation to a term's specific use, and that that usage information is not restricted to domain: 3.3.1 Note 4, 4.4.3.2.2 Note 3.
That's not inconsistent with ISO-10241-1 at all: usage information in that spec is a union of domain (of designation), grammarInfo, pronunciation, language, script, geographicArea.
The version of ISO/IEC DIR SUP IEC cited in ISO-10241-1:2011 is not online. Its successor https://www.iec.ch/members_experts/refdocs/iec/isoiecdir-iecsup%7Bed10.0%7Den.pdf is.
And the illustration is:
The thing in the the angle brackets in IEV, ISO/IEC DIR SUP IEC is explicitly saying, is a domain. ISO-10241-1:2011 is saying that earlier on, IEV allowed a more general category than domain; it didn't say what that was, but the ISO-10241-1 definition of what it could be is clear.
What is in the angle brackets is the domain of the designation, which IN THEORY is inherited from the concept. In practice, it cannot be: the domain of the designation is optional disambiguating text, and is language-specific. Multiple designations under the same concept can be in different languages, so they do not share the same common domain as language-specific text. Particularly when that text is supplied for some designations and not for others.
So IEC 161-02-19 has
The French "<of a harmonic>" is not a concept domain, inherited fromt the root. It is a disambiguation specific to French, and not inherited into English.
What is to be done:
usage info is not a usage note, and is not a property of concepts. ISO 10241-1:2011 6.2.7 is clear that the usage note is a note, and the rest of the spec is clear that usage info is already defined as one or more of grammarInfo, pronunciation, language, script, geographicArea. We will remove usageinfo from concepts.
We will add usageinfo as an attribute of designations, NOT concepts, to capture the miscellanea in ISO 12620.
We will add fieldofapplication to designations, to express the disambiguation illustrated under Rang, <d'un harmonique>. It is understood to be disambiguating text specifically, and to be published inline after a designation, rather than at the start of a definition, as for the concept domain. fieldofapplication is what the IEV calls this; it is a descriptor related to the domain, rather than the domain itself
If we ever need the pre-2016 IEC usage of usageinfo as disambiguation in brackets, we can use the usageinfo of designations. But in current documents, and by default, it is disambiguating text, and it is a fieldofapplication.
In any parsing of IEV, I strongly recommend that the text in brackets be treated as fieldofapplication, not usageinfo; but I will render both the same way.