03-05-2018, 04:02 AM
Thank you all for your replies. I will continue reading.
I recently read D'imperio ("Enigma" and the 2002 paper) and am working thru Nick's blog backwards to 2010 - where there was a mention of Chinese - phonetic alphabets of any kind seem to be a century after VMS so that's a fair point.
Also, I did not realise how bad the available transcriptions are - you could have a whole conference just on the correct folio order, let alone correctly recognising EVA a and o - no wonder statistical programs don't work very well on the whole corpus!
By contrast, I was impressed with the smaller and very carefully matched selection of input data used by D'Imperio.
I recently read D'imperio ("Enigma" and the 2002 paper) and am working thru Nick's blog backwards to 2010 - where there was a mention of Chinese - phonetic alphabets of any kind seem to be a century after VMS so that's a fair point.
Also, I did not realise how bad the available transcriptions are - you could have a whole conference just on the correct folio order, let alone correctly recognising EVA a and o - no wonder statistical programs don't work very well on the whole corpus!
By contrast, I was impressed with the smaller and very carefully matched selection of input data used by D'Imperio.