On 13/01/2014 15.18, Wojciech Siemaszkiewicz wrote:
I am not sure I understand this comment from Aaron about transliteration.
“The effort involved in Romanization is often greater than the effort involved in the rest of cataloging, and usually in [is] worthless to users.”
The romanization of non roman alphabets was introduced to cataloging so we can read titles, etc. if we are not familiar with the original language. How is that worthless to users beats me. It was created for the users to begin with so they can find material in foreign languages.
Wojciech is right. Transliteration is vital for many users who are learning a foreign language, but it is also very important for staff who can often pick up quite a little bit on the job, once something has been transliterated. For instance, the Russian баскетбол or ресторан looks incomprehensible to someone who doesn’t know, but when transliterated, they spell “basketbol” (basketball) and “restoran” (restaurant).
It would be nice however, if the transliteration could be done automatically somehow. Perhaps it will be possible. Cyrillic is a one-to-one transliteration, after you determine the correct language. This means that a letter may be the same in Russian or Ukrainian or Serbian but each language transliterates that same letter differently.
Still, doing an automatic transliteration of Cyrillic alphabets, should be easier than Arabic or Hebrew. Google Translate has its problems but is really an amazing tool.