AreaAudit

Which languages each generator says it supports, and where

Region subtag: the half of a code that names a market

A region subtag is the second part of a code like the Brazilian or German forms. The first part names the language; the subtag names the place whose norms the content follows. Leaving it off is legal, common, and hides the thing an audience can hear. As of 2026-09-22.

The two questions a language code can answerThe language part of a code answers what somebody would call the language. The region part answers whose version of it. A code that answers only the first is not a shorter version of one that answers both; it is an answer to one fewer question.Present in every codeLanguage partWhich language the content is in. Always present.Script partWhich writing system, where two are in use for one language.Region partWhose norms the wording follows. Optional, and usually absent.Present in a minority of the codes read here
Fig. 1 Spelling, vocabulary, date order and formality all move with the region part, which is why it is transcribed rather than normalised away.
What a code says with and without a region subtag. Recorded 2026-09-22.
Code written asWhat is pinned downWhat is left open
Language onlyWhich language the content is inWhich variant, spelling norm or accent
Language and regionThe language and the market it is written forWhether other markets are served too
Language and scriptThe language and the writing systemWhich country the wording follows

Inclusion rule. The three shapes a code can take in the material this register reads. Private-use and extension parts exist but appear nowhere in it. Order. Least specific first.

1One code, two jobs

The language part answers what somebody would call the language. The region part answers whose version of it. Those are different questions, and a code that answers only the first is not a shorter version of one that answers both.

The distinction has a cost attached. Spelling, vocabulary, date order, currency and formality all shift between variants of one language, and a page written for one market reads as slightly foreign in another.

2Why so many codes leave it out

A bare code is cheaper to maintain and more permissive: it invites every reader of the language rather than one country. For a marketing site that is often the right call, and it avoids committing to a variant nobody has budgeted for.

It is the wrong call wherever the variant is the decision. A buyer choosing a tool for one market gets no answer from a bare code, and cannot tell whether the vendor made a choice or has simply not made one yet.

3How this register treats it

The code is transcribed exactly as the vendor wrote it, subtag included, and never normalised. A cell reading one thing and a cell reading another are different facts, and collapsing them would delete the only variant evidence these pages carry.

Counting is done on the language part, so a vendor declaring one variant of a language counts once for it. The form is printed beside the count so the narrower claim stays visible next to the broader one.

A definition, not a measurement: no vendor figure appears here that is not also in the register. Codes as they appear per language are listed one language at a time.

Nearby terms: Language subtag, Script subtag. The whole vocabulary is at terms; nothing on this page is a claim about a named product.