Fw: Please find a pdf attached.

[Helena S Chapman <hchapman@us.ibm.com>]

FYI. I received this by the way of Unicode
Consortium recently. Feel free to agree or disagree with my resposne. 

I do have a question on something he
mentioned. I have NOT seen any production usage of pictograph (or emoji)
symbols coming through our translation pipeline at all. In the context
of text to speech for the purpose of translation/localization, what is
the common practice? It is not an XLIFF question per se...

----- Forwarded by Helena
S Chapman/San Jose/IBM on 04/30/2014 12:07 PM -----

From:      
 Helena S Chapman/San
Jose/IBM
To:      
 William_J_G Overington
<wjgo_10009@btinternet.com>
Cc:      
 "uli@unicode.org"
<uli@unicode.org>
Date:      
 04/30/2014 12:06 PM
Subject:    
   Re: Please find
a pdf attached.

---

William. If you look at XLIFF in details,
you might find that though it comes with lots of bells and whistles, the
core module itself is not as overwhelming as you expect. People always
have the option to pick and choose additional extensions including namespaced
custom extensions.

I'd like to also learn what real life
use case exists today for readouts.dat outside of XLIFF? You mentioned
transform from readouts.dat to XLIFF can happen. I agree and that's the
beauty of XLIFF where a universally, well informed localization enthusiasts
do use it for interchange purposes. Some tooling might even use it as an
internal processing format today though that was not the original intent.
So, unless there is a very good business reason we need something like
readouts.dat to exist, my recommendation at this point is to preserve it
for organization proprietary processing purpose only.

Thank you.

Best regards, 

Helena Shih Chapman
Globalization Technologies and Architecture
+1-720-396-6323 or T/L 938-6323
Waltham, Massachusetts




From:      
 William_J_G Overington
<wjgo_10009@btinternet.com>
To:      
 Helena S Chapman/San
Jose/IBM@IBMUS
Cc:      
 "uli@unicode.org"
<uli@unicode.org>
Date:      
 04/30/2014 11:57 AM
Subject:    
   Re: Please find
a pdf attached.

---

Hi Helena

Thank you for your email.

> I'd like to get a better understanding of what problems you are attempting
to solve that is not already solved by open standards such as OASIS XLIFF?

Well, until I read your email I knew nothing at all about XLIFF.

I found the following.

http://en.wikipedia.org/wiki/OASIS_%28organization%29

http://en.wikipedia.org/wiki/XLIFF

http://docs.oasis-open.org/xliff/xliff-core/xliff-core.html

I have not studied it all yet.

From looking at the example in the http://en.wikipedia.org/wiki/XLIFF
web page, it is possible that the answer to your question is that the readouts.dat
format will not solve any problem that could not also be solved using XLIFF.

However, the readouts.dat format is very lightweight and could perhaps
be useful in specific situations, such as, for example, if someone who
is able to translate into Latvian from English wants to copy a file of
emoji to English read-out labels and localize it and produce a file of
emoji to Latvian read-out labels using just WordPad on an ordinary PC.

Would that be useful if, say, a manufacturer of a text-to-speech system
included the ability to read in and apply a readouts.dat file, thereby
giving an end user the opportunity to customize the system using a readouts.dat
file that he or she had prepared using just WordPad and no special software
tools?

It appears that in relation to mainstream industrial use where XLIFF is
in use and all of the knowledge, skills and facilities are available to
use it that there may well be no use for the readouts.dat format.

Yet perhaps the readouts.dat format might find use as an easy to learn
and easy to use format for use by an end user in customizing text-to-speech
software packages for use with a particular language, particularly when
the language is not one specifically supported by the manufacturer of the
software package.

Maybe there could be a software tool that reads in a readouts.dat file
and produces an XLIFF file that contains all of the localization information
that is in the readouts.dat file.

When I saw the original HTML release of the text that is now in a pdf document
in the Unicode Technical Committee Document Register as document L2/14-093,
in particular the text that is now the first complete paragraph on page
4 of the L2/14-093 document, I thought that the sentence.dat format that
I had produced for my research in communication through the language barrier
could be adapted to produce an easy to use format to localize emoji to
text: the result is the readouts.dat format.

I am attaching two publications from my research as that background might
be of interest.

Also, a publication about a simplified spin-off variant that could perhaps
be useful in transliteration.

I fully accept that the readouts.dat system is not as comprehensive as
XLIFF, yet I am wondering notwithstanding that, whether the readouts.dat
format might be useful in some circumstances: for example if someone is
developing a text-to-speech system and needs a way to allow an end user
to supply to the system a user-supplied customized localization file for
the text equivalent of each of a number of pictograph characters and symbols.

I would be interested to know your comments if possible please.

Yours sincerely

William Overington

30 April 2014
[attachment "Communication_through_the_language_barrier_in_some_particular_circumstances_by_means_of_encoded_localizable_sentences.pdf"
deleted by Helena S Chapman/San Jose/IBM] [attachment "The_format_of_the_sentence.dat_file_used_for_automated_localization_of_encoded_localizable_sentences.pdf"
deleted by Helena S Chapman/San Jose/IBM] [attachment "The_format_of_the_translit.dat_file_suggested_for_possible_use_for_transliteration.pdf"
deleted by Helena S Chapman/San Jose/IBM]