Variable names are multilingual, and used as locators

[DanSmith]

Should variable name be a simple string, or is a new locator variable name needed to describe the actual column name/locator in various data file formats and databases.

[borsna]

Makes sense to make the variable name a simple string, it should have a 1:1 match to the data file.
The label is probably a point where localization makes sense.

[wlthomas]

That also reflects the structure of codebook. Wendy L. Thomas ISRDI [retired] email: ***@***.***

…

On Wed, Apr 15, 2026, 04:08 Olof Olsson ***@***.***> wrote: *borsna* left a comment (ddialliance/ddimodel#66) <#66 (comment)> Makes sense to make the variable name a simple string, it should have a 1:1 match to the data file. The label is probably a point where localization makes sense. — Reply to this email directly, view it on GitHub <#66 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJT47LE6FTJGPQL7JICUPD4V5GRDAVCNFSM6AAAAACXZ6A63CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DENJQG4YTEOBWHA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[wlthomas]

I've been thinking about this further. While it makes sense in terms of a variable name, "name" has been ubiquitously throughout DDI-L for all versionable and many identifiable objects. This means that it may be the name of a section, a data collection instrument, etc. etc. In these cases the name, like the title of a book, may be expressed in different languages. This is probably an artifact of the inclusion of name as a common element within complex structures and expanding its meaning and use beyond the specific usage of name in terms of a variable. Redefinitions of terms that have been in use since 2003 may be a 4.1 or later issue when a simplification of the model may be in order. I think it may exceed the purpose of the 4.0 model. Note that Dan has already introduced the basic structure of a Describable Type which includes a standardized name, label, and description. If the definition and type of "name" is changed in one place it should be done across the board. I don't think that would necessarily be appropriate.

…

On Wed, Apr 15, 2026 at 4:08 AM Olof Olsson ***@***.***> wrote: *borsna* left a comment (ddialliance/ddimodel#66) <#66 (comment)> Makes sense to make the variable name a simple string, it should have a 1:1 match to the data file. The label is probably a point where localization makes sense. — Reply to this email directly, view it on GitHub <#66 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJT47LE6FTJGPQL7JICUPD4V5GRDAVCNFSM6AAAAACXZ6A63CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DENJQG4YTEOBWHA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Wendy L. Thomas ISRDI [retired] email: ***@***.***