For decades, researchers have worked towards increasing the recognition strength of efficient but non-general (LL- and LR-based) parsers and increasing the efficiency of completely general algorithms such as Earley's and Generalized LR (GLR). In other words, we can't feed yacc any old grammar or easily build a recursive descent parser for complex languages. But, tools that handle any grammar have traditionally been very inefficient. 

I have devoted my research life to increasing the strength of the simple and efficient top-down LL parsers, culminating in the powerful ALL(*) strategy of ANTLR 4. In this talk, I will summarize the key practical advances in parsing from the last 25 years, provide algorithm comparisons, and separate the promises from reality.