What did you do?
While reviewing Japanese translation consistency for the term "operator" (see #23594), I found that glossary.md defines "Operator" strictly as a PD scheduling concept:
Operator
An operator is a collection of actions that applies to a Region for scheduling purposes. Operators perform scheduling tasks such as "migrate the leader of Region 2 to Store 5" and "migrate replicas of Region 2 to Store 1, 4, 5".
An operator can be computed and generated by a scheduler, or created by an external API.
However, the bare word "operator" is also used elsewhere in the docs for a completely different concept — a node/step in a query execution plan (e.g. TableScan, HashAgg, Exchange, IndexLookUp), for example in partitioned-table.md ("each operator supports direct access to multiple partitions") and throughout the EXPLAIN-related pages. This second sense has no glossary entry of its own and is not distinguished from the PD-scheduling sense anywhere in the docs.
What did you expect to see?
Either:
- A separate glossary entry (or a disambiguation note on the existing "Operator" entry) for the execution-plan sense of "operator", so readers landing on the glossary from an EXPLAIN-related page aren't given an unrelated PD-scheduling definition, or
- Confirmation that this overlap is intentional/acceptable, so translation teams (and future doc edits) know to keep treating both senses as the same term rather than trying to disambiguate them.
What did you see instead?
A single "Operator" glossary entry that only documents one of (at least) two distinct technical meanings the word carries elsewhere in the docs, with no cross-reference or disambiguation between them.
Additional context
This surfaced while normalizing the Japanese translation's katakana spelling for "operator" (オペレータ → オペレーター) across both PD-scheduling and execution-plan contexts. Since the English source itself doesn't distinguish the two senses, the Japanese translation currently mirrors that (uses the same term for both) — but it would help future translation/editing work to have an explicit answer on whether that's the intended approach.
What did you do?
While reviewing Japanese translation consistency for the term "operator" (see #23594), I found that
glossary.mddefines "Operator" strictly as a PD scheduling concept:However, the bare word "operator" is also used elsewhere in the docs for a completely different concept — a node/step in a query execution plan (e.g.
TableScan,HashAgg,Exchange,IndexLookUp), for example inpartitioned-table.md("each operator supports direct access to multiple partitions") and throughout theEXPLAIN-related pages. This second sense has no glossary entry of its own and is not distinguished from the PD-scheduling sense anywhere in the docs.What did you expect to see?
Either:
What did you see instead?
A single "Operator" glossary entry that only documents one of (at least) two distinct technical meanings the word carries elsewhere in the docs, with no cross-reference or disambiguation between them.
Additional context
This surfaced while normalizing the Japanese translation's katakana spelling for "operator" (オペレータ → オペレーター) across both PD-scheduling and execution-plan contexts. Since the English source itself doesn't distinguish the two senses, the Japanese translation currently mirrors that (uses the same term for both) — but it would help future translation/editing work to have an explicit answer on whether that's the intended approach.