The ICBN rule was that family names are based on the type genus name* with suffix -aceae, but a few descriptive family names were grandfathered in as optional alternatives - Compositae, Leguminosae, Labiatae, Umbelliferae, Gramineae, Palmae, Cruciferae, Guttiferae. I doubt that Stace has changed anything. I assume that the ICN hasn't changed anything either.
I see it as a personal choice by authors, with a long term trend to the -aceae forms.
- There are a few familiies, include Theaceae, Fabaceae, Caryophyllaceae and Winteraceae, based on genera which are not currently recognised (Thea -> Camellia, Faba -> Vicia, Caryophyllus -> Dianthus, Wintera -> Drimys).