Re: Procedural Guidelines for Proposed New or Revised Romanization Tables

Posting to Autocat

On 22/04/2011 16:31, Aaron Kuperman wrote:

The issue we should be discussion is whether we should Romanize at all. Romanization is inherently a problem since a systematic Romanization needs to be based on a single dialect, and there conflicts with how the language is Romanized in practice. For example, an Arab or Jew who uses English as a second language will Romanize Arabic radically differently than one who was taught French as a second language (e.g. “sh” as opposed to “ch”).

While some Romanization is necessary so monolingul Americans can identify the books, perhaps that can be limited to headings under authority control and the short title. As it is a large amount of the record in Romanized, which is a tremendous waste of cataloging resources, and of no benefit to end users (who after all, prefer the original rather than the Romanized text).

I agree with this. My own experience is that library technicians need it, along with students of the language who need help. Also, someone who doesn’t know the language can often figure out a lot by looking at the romanization, e.g. when I would catalog a translation from e.g. Chinese, lots of times I could get an idea of the original title from the romanization. Of course, I still needed someone to confirm my suspicion!

In these discussions I would also want to study the incredible new possibilities available through tools such as Google Translate, e.g. “Literaturnaia gazeta” in Russian and through Google
Translate|en. I introduced this tool into the AUR catalog, where, if you look in the right hand column, you will see the Google Translate tool and you can change your language. It still seems like magic!

But still, with these kinds of tools, what is/will be the purpose of transliteration?