"In closing, it should be emphasised (sic) that a sound methodology is highly important if the study of the linguistic landscape is to become a serious sociolinguistic research tool. Walking the streets and taking photos of anything that might strike one | as particularly curious, illustrative, or, worse still, 'representative' is unlikely to yield any scientifically relevant results. While this has been common sense in sociolinguistic research on spoken languages for decades, much remains to be done when it comes to empirical research into written language."
Peter Backhaus Linguistic Landscapes: A Comparative Study of Urban Multilingualism in Tokyo. Multilingual Matters 136. Clevedon, Buffalo, Toronto. 2007: 63-63.