The general trend in the development of parser theory is in the direction of exploring implementing methods of increasing power. In particular, ways of improving the efficiency of LR parsers and the generation of LR tables have been receiving a lot of attention. The value of increasingly powerful tools is questioned from the point-of-view of the need to keep definitions of languages understandable to the programmer. Consideration is given to Wirth's contention that recursive descent is the method of choice and alternatives are suggested.