It is pretty fascinating; a book of Ossetian fairy tales landed on my desk not long ago. One minority language of the former USSR really stumped us for a bit because instead of being in Cyrillic script it used a modified Romanization system that only lasted for a few years in the 1930s. I forget which language it was, though I believe it was one of the smaller ones that is now transliterated beginning with a K and was/is spoken in Siberia. I believe it might have been a reader for new literates. 
From: Constructed Languages List [[log in to unmask]] on behalf of MorphemeAddict [[log in to unmask]]
Sent: Thursday, September 11, 2014 4:30 PM
To: [log in to unmask]
Subject: Re: THEORY: Recognizing languages without really learning them

Krista, it seems to me that you have my dream job. I hope you enjoy it.

If I encounter a book for sale and can't identify the language, that very
fact makes it desirable. It doesn't happen very often. The last it did, the
book, which I thought was Persian, turned out to be Dari, according to the
Russian colophon. I was close.
I used to have lists of short words to help me distinguish i.a. the Indian
languages in Devanagari (Hindi, Sanskrit, Marathi, et al.). I've probably
lost that list.


On Thu, Sep 11, 2014 at 2:18 PM, Krista D. Casada <[log in to unmask]> wrote:

> Actually, I do this about five days a week. I catalog foreign language
> books for my university library, and my job is to match them with the
> appropriate records in the OCLC catalog, an online database used by
> libraries around the world. Today I am mainly working with Japanese and
> Turkish books that have been donated. I don't speak either language,
> although I know a little bit of both. For older Japanese and Chinese books,
> I keep a file of frequently-used characters to help me approximate titles
> (due to budget constraints I don't have a scanner), and occasionally get to
> narrow down my choices by entering page numbers or publication years.
> You all are about to convince me to ask for a raise.
> Krista
> ________________________________________
> From: Constructed Languages List [[log in to unmask]] on behalf
> of Gary Shannon [[log in to unmask]]
> Sent: Thursday, September 11, 2014 12:53 PM
> To: [log in to unmask]
> Subject: Re: THEORY: Recognizing languages without really learning them
> This would be easier, I think, if you only had to teach how to recognize
> them in print. The foreign language section of one of my local used book
> stores often has books filed in the wrong language, but it's easy to see
> how a few simple rules would go a long way toward separating, say, Spanish
> and Portuguese. If it looks like Spanish, but there's a "ã" anywhere in he
> text, it's Portuguese.
> It would probably be pretty easy to come up with a list of words or letter
> sequences that are very common in one particular language but do not occur
> in any other language. And of course, you could include "shapes" in that
> list for Cyrillic, Chinese, Japanese, Korean, Burmese, Arabic, etc.
> Take a look at
> --gary
> On Thu, Sep 11, 2014 at 3:06 AM, Leonardo Castro <[log in to unmask]>
> wrote:
> > Hi, everyone!
> >
> > If you were offered a one-billion-dollars job to teach 007 how to
> recognize
> > the top 100 most spoken languages with no doubt, but still not having to
> > speak all of them, would you accept it? What would be your teaching
> method?
> > Suppose you have six months to accomplish the goal and that 007 is a good
> > student with full time available to do whatever you want to learn it
> during
> > those 6 months.
> >
> > And what if the 20 most spoken sign languages were included in the list,
> > would you still accept that job?
> >
> > If you don't have those skills yourself, suppose that they give you an
> > additional month before beggining teaching for you to develop them. :-)
> >
> > Até mais!
> >
> > Leonardo
> >