docs: micro edit for conceptual clarity

Author: ox-harris

site/engineering/realtime-engine.md

@@ -11,7 +11,7 @@
 
 The Realtime Engine is the core of LinkedQL’s [live queries](/capabilities/live-queries). It is **an _in-memory_ compute + cache layer** that lies between the storage layer and the application layer, transforming storage-level mutations — from WAL, binlog, or in-memory emitters — into logical events that materialize as live state, extending SQL’s reach from static queries to fully “live” queries.
 
-Unlike replica-based "live query" solutions that operate only against local databases or materialized stores, the engine is designed to operate seamlessly across storage backends — from local databases to traditional databases such as PostgreSQL and MySQL — allowing both to participate equally in the "realtime" space. LinkedQL thus brings reactivity into the realms of traditional SQL databases, erasing the local/remote distinction and making live queries a universal concept across storage backends.
+Unlike replica-based "live query" solutions, like PGLite, that operate only against local databases or materialized stores, the engine is designed to operate seamlessly across storage backends — from local databases to traditional databases such as PostgreSQL and MySQL — allowing both to participate equally in the "realtime" space. LinkedQL thus brings reactivity into the realms of traditional SQL databases, erasing the local/remote distinction and making live queries a universal concept across storage backends.
 
 This paper describes the engineering behind this design — from change detection and normalization to live, observable objects that self-update in real time.