[Lexicon] The Preferred Vocabulary

Tony Proctor tony at proctor.net
Mon Nov 3 11:54:45 CST 2014


The Lexicon has made huge progress in gathering together established usage for words across multiple fields. Whether we like/agree with them or not, they're currently out there and being used in that way.

It's early days still but I would like to set the ball rolling for the next stage of dealing with terminology: that of deciding on a preferred set of terms for our own writings, and especially for public-facing works such as standards. The non-software fields, such as research, transcription, archiving, etc., have a more established (although not unique) set of terms and concepts. Software, on the other hand, partly because there is no precedent for some of the things we need to do, has duplicated, conflicted, or a dearth of relevant terms depending on the topic.

It will be hard to achieve a preferred set given the disparate usage already out there, and the risk of ambiguity with everyday terms, or clashes with terms already used in a software context. I previously raised this latter issue by mentioning XML's ubiquitous use of 'element', 'attribute', 'document', and 'tag' which we must strive to avoid.

One case where my own terminology has gradually evolved to something I'm happy with is in the area of evidence-person versus conclusion-person, and its generalisation to other subjects. I'd like to present that here now, mainly to kick off some discussion. Part of this scheme involves the use of hyphenation to create compound words; something I've already suggested may be necessary in order to avoid confusion and clashes.

        - - - -

In order to generalise the concept of 'person' to include other things (e.g. places and groups in STEMMA's case), I've used the term 'subject'; this being the a "subject mentioned in some document".

In order to refer to the mention in a specific document, and the various details/properties/relationships associated with that mention, I've used the term 'reference'.

In order to refer to the computer representation of a 'subject', I've used the word 'entity'.

This means that a person-reference is then distinct from a person-entity, and a place-reference distinct from a place-entity (something that many genealogists don't notice), and both are distinct from the real-world person or place, respectively. In the generic sense then these become subject-reference and subject-entity.

    - - - -

I'm not asking everyone to accept these. As I said, I want to kick off some discussion. However, I would like people to think along those lines and consider how hyphenation might be used to create a very precise set of terms that no one is going to get upset by or confused by.  :-)

    Tony
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://fhiso.org/pipermail/lexicon_fhiso.org/attachments/20141103/3835e2f3/attachment.html>


More information about the Lexicon mailing list