German blogger Noodlegei, who mostly writes on ways to use all the free applications and services from Google, some days ago tweeted a comparison of word occurrences in the Google books corpus. While I had read about this new corpus before, it was this tweet which made me think about using it for similar comparisons within the scope of this blog.
Whereas the word อำเภอ for the Thai districts is usually romanized as "Amphoe" following the RTGS romanization system, still rather popular on websites is the alternative transcription "Amphur" - Googlefight shows them as almost equally popular. But websites, especially with the thousands of copies of Wikipedia, are not that much good as a corpus, the books scanned for Google Books are of a much higher quality. And as a plus the Google Ngram viewer also includes the development over time. For the comparison of Amphoe vs Amphur it shows that the RTGS spelling is slowly becoming the more common one, but due to the rather low number of books containing this term years like 1982 with unusually many publications using Amphoe create a very significant peak, even with the smoothing function.
A second comparison of Muang vs Mueang for เมือง - the second one became the recommended transcription just ten years ago - shows a clear preference for the term Muang, with Mueang rising very slowly. However this comparison is slightly unfair, as there are also placenames where Muang is the official transcription, e.g. Ban Muang district, and also the districts of Laos are usually transcribed as Muang (ເມືອງ).
In German it has the Umlaut characters which fit perfectly the sounds of the two vowel, so for me the most natural way would be to spell them Amphö and Müang. These spellings are really unusual however - Amphö has only some 200 Google website hits, Müng slightly more with about 10.000, and even less when restricting to Google Books. I will of course stay with RTGS despite its shortcomings, and will also try my best to keep the Wikipedia articles to stay with the only standard it has.