# "Huffman coding is unsuitable for text files"?

My lecturer for information theory says that "Huffman coding produces efficient codes but is unsuitable for text files where the letters are represented by a fixed length ASCII code".

I do not understand this statement. I understand that as you change the text file, you'll have to update the probabilities and thus the Huffman table/code itself but I don't think this is what is being referred to here.

Can someone explain?

Thank you so much for all your help

• Why don't you ask your lecturer? Jan 11 '17 at 13:26
• The lecture course is finished and have attempted contact with the lecturer; I wouldn't be asking on here if I hadn't tried to get an answer already. Jan 11 '17 at 23:45