ICANN Language Issues: Looking Down the Road

[Sam Lanfranco]

Michael Karanicolas has written: "I still believe that based on the ICANN
meeting reports that language services are largely unused by the community
and represent a huge overhead that we maintain purely to make ourselves
look inclusive rather than providing a set of services that are actually
used by the community". 
ICAHN is not alone with this dilemma, and I also feel that the expenditures
mainly result in nice optics, while miss-allocating increasingly scarce
ICANN resources. In other work that I do there are a variety of approaches
to deal with multiple languages and equitable access by all. Here is an
effort out of the Bibliotheca Alexandrina to deal with the language issue
at the text level. While this applies to the language of health science,
there are possible lessons here with regard to ICANN's overall text
strategy in a multilingual world. Admittedly we are much less down this
road when it comes to simultaneous translation, but savings on any
linguistic front free resources for the more difficult multilingual issues.
 
Sam Lanfranco
	Multilingual Science   

	There are 6500 languages in the world, but only one language of science,
English. 
	The Library of Alexandria and Scientists in Developing countries aim to
change this.  
	"Health is a state of complete physical, mental, and social well-being and
not merely the absence of disease or infirmity". (WHO)  
	"الصحة هي حالة من الصحة البدنية والعقلية
والاجتماعية الكاملة وليس مجرد غياب المرض
أو العجز"(Arabic)  
	Here is the WHO definition of Health in English and Arabic.  In the next 5
years, all journals can become multilingual.  
	The Library of Alexandria is ushering in a new era of multilingual
science.  Reducing the burden of translation, indeed promotes openness and
scholarship.  
	“We build too many walls and not enough bridges” (I. Newton)   

	One of the major walls of science is one of different languages.  
	We are writing to you with a simple idea. In 2006 Vint Cerf (the father of
the Internet) told us about google translate, and we were excited to build
a language translator into our Supercourse, and the Library of Alexandria. 
However, much of it produced was Gibberish.  We tried again five years ago,
and google translate captured the “gist” of the materials, however, it
was still a bit messy.  In the past few months we have seen a major
improvement in translation capabilities with a new approach called Google
Neural Machine Translation which produced marked improvement in accuracy,
and we feel is now acceptable to build multilingual science world wide.  
	Multilingual Translation of Science   

	We propose to establish a beta version of a translation button to be made
available to all journals for free.  We would capture metrics of usage, and
concerns of translation. We would like to have you join. We are planning a
6-month beta version; after which we will decide whether to continue. We
will help you add the button to your journal and site.  
	Please contact Ronald LaPorte, Ph.D. (Emeritus professor University of
Pittsburgh)  

	ronaldlaporte at gmail.com (mailto:ronaldlaporte at gmail.com)  

	Ismail Serageldin, Ph.D.  

	Professor Emeritus, Library of Alexandria, Alexandria, Egypt
---- Original Message ----
From: Michael Karanicolas 
To: NCSG-DISCUSS at LISTSERV.SYR.EDU
Sent: Fri, Dec 28, 2018, 12:01 PM
Subject: Re: Thoughts on the Proposed ICANN Org Budget for FY20

Interesting conversation. Thanks very much Ayden for going through the
budget doc to kick this off - which is important work.
"I still believe that based on the ICANN meeting reports that language
services are largely unused by the community and represent a huge overhead
that we maintain purely to make ourselves look inclusive rather than
providing a set of services that are actually used by the community. I
haven’t looked at recent figures, but back during ICANN Dublin I believe
that the majority of the interpretation channels were used by less than 10
people at huge expense to the org, we need to really look at wether
language services are actually delivering any ROI or are we merely spending
to look good in this area."
As an English speaker, I'm always a bit hesitant about jumping into this
issue, as I know I have the privilege/advantage of having my native
language as the default. It's quite possible that ICANN's main concern here
is the optics, but that's not something to be entirely discounted either.
As an organization that strives to be international and inclusive, I think
they like having a UN-esque model of translation. Still - I wonder if there
isn't a better way of doing things, like potentially arranging an opt-in
system, where users could pre-request language services for a meeting, and
they'd only be provided if a certain number of people registered for a
given language. That could also help them be better targeted as well - and
have interpreters in the sessions where they're most needed, as opposed to
just in the main rooms by default. May be something that's worth raising.

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ncsg.is/pipermail/ncsg-discuss/attachments/20181228/9d84f4c0/attachment.htm>