ELL: re: On software for Basque and other lgs ("support")
Jeff ALLEN
jeff at elda.fr
Wed Jul 14 18:36:50 UTC 1999
id PAA24892
To: owner-endangered-languages-l at carmen.murdoch.edu.au
Precedence: bulk
Reply-To: endangered-languages-l at carmen.murdoch.edu.au
*** EOOH ***
Return-Path: <owner-endangered-languages-l at carmen.murdoch.edu.au>
X-Authentication-Warning: carmen.murdoch.edu.au: majodomo set sender to
owner-endangered-languages-l at carmen.murdoch.edu.au using -f
X-Sender: jeff!elda.fr at 192.168.1.1
Date: Wed, 14 Jul 1999 20:36:50 +0200
To: endangered-languages-l at carmen.murdoch.edu.au
From: Jeff ALLEN <jeff at elda.fr>
Subject: ELL: re: On software for Basque and other lgs ("support")
Content-Type: text/plain; charset="iso-8859-1"
X-MIME-Autoconverted: from quoted-printable to 8bit by carmen.murdoch.edu.au
id PAA24892
Sender: owner-endangered-languages-l at carmen.murdoch.edu.au
Precedence: bulk
Reply-To: endangered-languages-l at carmen.murdoch.edu.au
Warning, this is a long reply with lots of information, so get a cup of coffee
and maximize your e-mailer to full-screen size.
>From: Trond Trosterud <Trond.Trosterud at hum.uit.no>
>Subject: ELL: On software for Basque and other lgs ("support")
>
>In my opinion Jokin Garatea is missing the point. Let me start by the word
>"support":
>
>The English word "support" is ambiguous: 1. within an IT context, "support"
>means "provide, possess, etc.", so when MS Windows supports Pascal it means
>that it is possible to run Pacal programs on that platform. As the readers
>of this list know, there are many ways a program or platform may "support"
>a natural lg L, starting with "containing all the characters used to write
>L", via "containing a keyboard layout for L", to "spellcheckers for L,
>hyphenation for L, grammar tools, thesauri, html conversion, etc. for L,
>the menus and help files for L, the paper manual written in L". 2. Another
>meaning of "support" is "help, give financial aid, voluntarely work, voice
>favourable opinions, etc.".
Thank you Trond for the point of clarification, and you are very correct. For
a few articles about comparative descriptions of software support for Minority
languages, please see:
SOMERS, Harold. Language Resources and Minority Languages. In Language Today.
Number 5, 1998. Nottingham, UK: Language Publications Ltd. pp. 20-24.
OSTLER, Nicholas. 1999. Does Size Matter? Language Technology and the
Smaller Language. ELRA Newsletter, Vol4 No2 (April-June 99).
And others that describe the development of technologies for minority
languages
(the list is certainly not exhaustive; I am constantly finding/writing new
articles on the topic).
OSTLER, Nicholas. 1999. Fighting Words: As the world gets smaller, minority
languages struggle to stake their claim. Language International, April 1999.
pp. 38-39, 45.
Hogan, Christopher (1998) Embedded Spelling Correction for OCR with an
Application to Minority Languages. Paper presented at Workshop on Embedded MT
Systems, in conjunction with the AMTA 98 conference. 28 October 1998,
Langhorne, Pennsylvania.
Hogan, Christopher. "OCR for Minority Languages", In Proceedings of the 1999
Symposium on Document Image Understanding Technology, Annapolis, Maryland,
April 1999, pp. 235--244.
ALLEN, Jeffrey. 1998. Lexical variation in Haitian Creole and orthographic
issues for Machine Translation (MT) and Optical Character Recognition (OCR)
applications. Paper presented at the workshop on Embedded MT systems of the
Association for Machine Translation in the Americas (AMTA) conference,
Philadelphia, 28 October 1998.
ALLEN, Jeffrey. 1999. La standardisation du cr.le ha.ien par
l'interm.iaire de la linguistique computationnelle. Presented at the
9.Colloque du Comite International des Etudes Cr.les. Held at the
Universit.de
Provence, Aix-en-Provence, France, 24 - 29 juin 1999.
Brown, Ralf (1998) Improving Embedded Machine Translation with User
Interaction. Paper presented at Workshop on Embedded MT Systems, in
conjunction with the AMTA 98 conference. 28 October 1998, Langhorne,
Pennsylvania.
Allen, Jeffrey and Christopher Hogan. 1998. Evaluating Haitian Creole
orthographies from a non-literacy-based perspective. Paper presented at the
Society for Pidgin and Creole Linguistics meeting, New York City, 9-10 January
1998.
LENZO, Kevin, HOGAN, Christopher, and Jeffrey ALLEN. 1998. Rapid-Deployment
Text-to-Speech in the DIPLOMAT System. Poster presented at the International
Conference on Spoken Language Processing. 30 November - 4 December 1998,
Sydney, Australia.
ESKENAZI, Maxine, HOGAN, Christopher, ALLEN, Jeffrey, and Robert FREDERKING.
1998. Issues in database design: recording and processing speech from new
populations (poster session). In Proceedings of the First International
Conference on Language Resources and Evaluation, 28-30 May 1998, Granada,
Spain. Vol. 2, pp. 1289-1293.
Decrozant, Lisa and Clare Voss. 1999. Building a "Tri-Text": Steps in the
Conversion of a Hard Copy Document to an On-line Resource. In ELRA
Newsletter. Vol 4 No. 1 (January-March 1999), Paris: European Language
Resources Association. pp. 10-11.
Decrozant, Lisa and Clare Voss (1998) "Cross-Linguistic Resources for MT
Evaluation and Language Training" in Proceedings of the Natural Language
Processing and Industrial Applications Conference (NLP+IA-98), Moncton, New
Brunswick, Canada, August, 1998.
Decrozant, Lisa and Clare Voss. (1999) Extending Use of Linguistic
Resources:
>>From FALCon to STARLing, in Proceedings of the Advanced Information
Processing
and Analysis Systems Group (AIPA-99), Washington, D.C., March 23-24, 1999.
Jones, D. and R. Havrilla. 1998. Twisted Pair Grammar: Support for Rapid
Development of MT for Low-density langs. In Proceedings of the Association for
Machine Translation in the Americas (AMTA) 98 conference "MT and the Info
soup". pp. 318 ff.
ALLEN, Jeffrey. 1998. Intelligence artificielle: apport de la traduction
automatique et le traitement automatique de la parole au cr.le ha.ien.
Seminar and system demonstration presented at the Universit.Cara.e,
Port-au-Prince (Haiti), 28 March 1998.
BAKER, Paul, MCENERY, Tony and Lou BURNARD. Minority Language Engineering.
ELRA Newsletter. Vol 3 issue 4; November 1998, Paris: European Language
Resources Association. p. 9
CAMARA, .mile, C.estin NSTADI, V.onique REY, and Jean VERONIS (December
1995)
Traitement Informatique des Langues Africaines: Probl.es et Perspectives.
Action de Recherche Partag. ALAF (AUPELF-UREF) Document ALAF ALA1. Version
1.0. http://www.lpl.univ-aix.fr/projects/alaf/ALA1.html
MASON, Marilyn. 1999. Automated approach to Haitian Creole orthography
conversion: Can this methodology be adapted to other creoles? Presented at
the
9.Colloque du Comite International des Etudes Cr.les. Held at the
Universit.de Provence, Aix-en-Provence, France, 24 - 29 juin 1999.
MASON, Marilyn. (forthcoming) Short Note: Automated Creole orthography
conversion. Journal for Pidgin and Creole Languages (JPCL).
ALLEN, Jeffrey. (in preparation) Short Note: Issues about Empirical Data for
Creole Language Research. In preparation for Journal for Pidgin and Creole
Languages (JPCL).
and the MILLE project:
http://www.ling.lancs.ac.uk/monkey/ihe/mille/1fra1.htm
>From: Jokin Garatea <garatea at gaia.es>
>To: endangered-languages-l at carmen.murdoch.edu.au
>Subject: Re: ELL: On software for Basque and other lgs ("support")
>
>I have nothing against MS or whatever other company. But this is not, in my
>opinion, the best way to promote the use of endangered languages through the
>use of telematic means. If all the promotional actions are to subcontract big
>companies to translate their almost "monopolistic" tools, forget your
national
>lg industries, if ther is any.
I am not necessarily indicating that we must support "monopolistic" tools, but
rather that it is important to put "such" tools in the hands of the people,
especially those who make the decisions. I am currently the adviser of a new
technologies project where I discouraged the project leaders to seek
development from large technology players that wanted too much money. I
suggested that several specialized academic institutes (competing ones at
that)
be invited to a workshop in May 99 along with the elders and other members of
the native tribes that have been wanting to consider these technologies. I
see
the academic research centers (those that have an excellent reputation and who
produce real tools) as a means of promoting such minority languages. The work
costs less, and it allows native-speaking students the opportunity to possibly
attend top academic programs in top institutes and to work on their native
tongues for developing such technologies. Lots of these institutes are
interesting in developing procedures for rapidly porting the technologies from
one language to another. No one (not even Microsoft, Sony, IBM, National
Science Foundation, Europe Commission, etc) wants to consider investing
significant amounts of money in projects that take 3, 4, 5 years. Granted,
there are still a few European proposals that allow 2-4 year project
proposals,
but projects now tend to be 1 and 2 year projects with intermediate
deliverable
prototypes and products. I know this because I create project proposals in
the
field all the time, and I also put out call for project proposals and evaluate
them. My last call for proposals required that the databases be created
within
a period of 12 months. The problem with 2-4 year projects is that the
return-on-investment period is too long. I'm currently working on another
major project where similar projects have taken 2 years and we are trying to
streamline the new project, with experience from other similar projects on
other languages, to cut the development and time-to-market down to 1 - 1 1/2
years. This is typical of all projects (yes, and for the academic ones that
are government funded). I am not talking about on-going medical research on
cancer, and such. We all know that this requires long-term funding. But
language technologies are at a point where government funding has
significantly
diminished over the past 5 years and the only way to continue the work is to
seek corporate investments. And the companies who are willing to provide such
financial backing want to see the products developed in a 1-2 year time
frame. This is reality and this is what we all must keep in mind for the
coming years.
>Minority lgs need a clear strategy to survive developing their own industries
>with the help, or not, of big companies and with the help, by sure, of their
>local administrations. This is a clear statement that we in the Basque
country
>are trying to undertake with the, very important, help of our local
>administrations.
I do not deny the need to develop local industries. I am in full support
(moral support I should clarify, since the word support is ambiguous as Trond
has wisely noted) of it. It is more a question of creating the
infrastructure. Please also note that the large corporations are not the
only ones developing functional technologies. Marilyn Mason of Mason
Integrated Technologies Ltd (contact: "Marilyn P. Mason" <mmason at mit.edu> &
MariLinc at aol.com) has developed orthography software that can be used on
multiple platforms. She is very willing to send information (including a
diskette) of what her program can do. She presented a demo recently at the
Etudes Creoles conference in Aix-en-Provence and is scheduled to be in
Seychelles or Mauritius for a conference this coming fall. As a small
start-up
company, she is actively seeking partnership with local groups to port her
technologies to different languages, possibly within the framework of US and
European-funded projects. She has worked in Zaire and in Haiti and with local
ethnic groups in Boston, USA and is familiar with the issues of cultural
localization. Her software is usable, functional, and accurate and has
benefitted from detailed comments from such people as Yves Dejean of the
Ministry of Education in Haiti. His comments were immediately integrated
(within a couple of days) to form a new version of her software. Turn-around
time with her is very fast because she has a small company. I would highly
suggest that ELL listers contact her for further information about what she
has
done.
>As far as I know, most of endangered lgs lack of local administrations
able to
>help them, so, are they condemmned to loose their technology train and also
>their lg?.
What I have been doing for Haitian Creole language technologies over the past
few years is to develop products (machine translation, terminology databases,
parallel aligned text corpora, translation memory, spellcheckers for OCR and
word-processing, speech recognition, speech synthessis) that are developed in
academic institutes in a rapid-development and rapid-deployment timeframe.
Once this is done for one or two languages (we did this for French, Spanish,
Croatian, Haitian Creole, and Korean), the objective is to improve the
processes and port them to new languages. I had a translation laboratory of
many native language translators per language, and we sought their language
and
cultural expertise for the translation and localization processes, hiring most
of them for 2-3 years each to work on the initial prototype demonstrations for
the ongoing maintenance work of the subsequent prototypes and final
(non-commercial) products.
>For, as far as I see, they will not able to subcontract a big
>company and neither to promote their own industry.
I agree with you here partially. I would not say that it is necessary to
subcontract the work to the big companies, but seeking funding from them is
indeed a possibility. Some big companies that I know of are sometimes
interested in (co-)funding projects that can prove that products and services
can be developed in rapid timeframes.
>The solution to this. I do not know. However we are trying to develop one:
>>Rapid convergence of hitherto separate technologies are already changing the
>nature of work and organisations. Versatile applications of existing and
>emerging technologies are being demonstrated in educational and training
>situation and it is no longer possible to ignore serious consideration of
their
>deployment in the minority languages.
Yes, I agree.
>We, in GAIA, based on our experience in this field and according to the lack
of
>beneficial results in the different program implementation for minority
>languages, have based our strategy in those three different lines:
>
>1. Marketing of Minority Languages.
>2. Concretion of the role of the Education Centres.
>3. Build up an important industry sector of marketable IT and Multimedia
>products addressed to minority languages.
>
>This approach does not mean to subcontract any tool. That means short term.
>Our language has survived to a lot of problems, among them a strong
>dictatorship, we are taking advantage of different programmes to finance our
>activities in that field. And it if is not easy for us, eventhough we can
count
>with our administration, what about other languages?.
>
>Our approach is not the best of course.
On this point, I think it is very important to emphasize that there is not "a"
single approach, (not even "a" single multi-faceted approach), that will be
the
answer to all situations. I believe that there are elements, strategies and
modules that can be adopted from previous experiences and adapted to new
situations. It is similar to the case of a family where the parents can
provide the same family education, same care, same attitude, same rules, etc
for 5 different children. But nothing guarantees that all 5 children will
grow
up and want to seek the same profession, have the same tastes and desires, and
even have the same life-style. It is very possible to have different
children
in a family that accept and reject the system proposed by their parents.
It the same with considering approach for minority (I use this term in a very
general way) language development processes. There is no single approach
that
will work in all contexts. Some countries have more financial backing than
others, some have more government backing than others, some have more industry
potential than others, etc...
What I see important is developing approaches that take the general core of
strategies and allow them to be adapted, customized, localized (as we say in
the translation and information technology sectors) to the new situation
for it
to be successful "for the customer/group of people" in question. The concepts
of globalization and localization have come about over the past 10 years
because large corporations 10+ years ago were trying to force a single
strategy, a single mindset, a single mentality and worldview upon other
cultures, and it did not work. They realized that not everyone wanted English
for everyday work products and procedures, that not everyone had the same way
of thinking, etc. The result was a significant, and ever-increasing, emphasis
on adapting the services and products to local peoples, and this is now a
major
preoccupation for hundreds, if not thousands, of companies.
>that It will take more than approaches to
>keep most European Minority Languages from becoming extinct.
Please note that the term "minority language" has different semantic
connotations in its use in Europe and the US, and is therefore culturally
ambiguous in English. "Minority Language" in Europe tends to refer to the
language of immigrants (e.g. Indian Hindi and Punjabi speakers in the UK,
Turkish speakers in Germany, etc). The EC has not been very enthusiastic
about
supporting this (at least, not until very recently).
However, in the US, the term "Minority Language" has been recently adopted by
computational linguists (taken from sociolinguistic circles) to refer to
languages of inferior social class even in their own countries (e.g. Haitian
Creole in Haiti with respect to French). A recent trend is to use the term
"Minority Language" to refer to low-density languages. A lot of Natural
Language Processing and Computational Linguistics conferences lately (1998 &
99) have been focussing on the following topics with special panels,
tutorials,
workshops, and thematic sessions.
1) minority languages
2) sparse-data languages
3) low-density languages
4) neglected languages
and 5) lesser-commonly taught languages
* LREC98 Workshop on Language Resources for European Minority Languages
* AMTA98 Neglected Language panel
* AMTA98 Workshop on embedded MT systems (with 3 papers on Minority languages)
* TMI99: Sergei Nirenburg (tutorial) Acquisition of Knowledge
about
a Low-Density Language for Use in MT
* CRL (New Mexico State University) summer school: A Survey of Language
Engineering Applications http://crl.nmsu.edu/summerschool
( This course will introduce language engineering applications such as
machine translation, information retrieval and extraction, ....
and the Expedition environment for configuring
machine translation systems for low-density languages.)
* Society for Pidgin and Creole Languages (SPCL) 98 (1 paper on Creole
language
technologies)
* Etudes Creoles 99 (several papers on Creole language technologies)
* CATANAL 99 Workshop. 20-21 May 1999. (workshop report: Wilkness, Peter
* and
Nora Deans. 1999. From Kayaks to Cyberspace: encounters of cultures and
technology - Computer-Assisted Translation of Alaska Native Languages
(CATANAL). A workshop (20-21 May 1999) report. International Arctic
Research
Center and University of Alaska-Fairbanks. Manuscript)
... and I am sure that the list goes on. I would appreciate any further input
from other ELL listers on conferences where technologies for minority,
endangered, etc languages have been discussed and presented.
Note: In most cases, even Korean and Arabic are considered to be a
lesser-commonly taught languages and potentially low-density because the
electronic data isn't always available to the public.
>If all it took was
>approaches, then the minority languages would not be in the sad condition
that
>most of them are in now because many of them have been exposed to approaches.
>What then? Lots of different approaches have been tried.
Again, there is not a single approach, nor is there a single multi-faceted
approach.
>These are not
>startling innovations; what we need is a critical mass of committed people,
and
>this critical mass can only be created through continuous capillary
>infiltration of information and encouragement.
Yes.
>Our approach is intended to be a part of such an effort.
I would like to add here a number of comments that I made on a very related
topic on another list at the end of June 99. I have had to remove the
comments
that provoked my reaction since I do not yet received permission from the
other
person to cite his comments. I can however cite my own.
----
Subject: #99: computers & economy : Allen comments
From: Jeff ALLEN <jeff at elda.fr>
As I stated during both of my presentatations at the Etudes
Creoles conference last week in Aix-en-Provence, one of the major setbacks
for the written Haitian Kreyol language at present is that there may be an
orthographic standard, but the written language (that is, the lexicon) has
no standard. Now, I am not blaming anyone for this, and it is easy to
point fingers. I am just making a statement based on objective facts of
thousands and thousands of sentences evaluated by texts written by 15
different teams of Haitians. The variability of written forms is such that
2/3% (yes, two-thirds) of all of the lexical items (words) in a 1.3 million
word Haitian Kreyol database (collected from 15 different sources) are
represented by 2 or more variant spellings. All of the statistics are
clearly outlined in my papers. Why? Well no one is investing in decent
literacy programs, and there is no real backing from the government. Haitians
in Haiti are learning to read and write in whatever way possible.
And as I said last week at the conference, lots of us can talk about
language, and about what needs to be done, but convincing the Minister of
Education, and the Minister of Culture, and others to invest in such
programs is not an easy task. They don't necessarily want to hear "words"
about what needs to be done to turn an oral language of the masses into a
written language. They want to see products and services that are going to
raise their country on the economic scale . Roads are products. Irrigation
systems are products. Crops are products. Mobile/cell phone networks are
services. Internet access is a service, etc......
[Jeff adds comment here: And I would like to emphasize to the ELL list that
the above-mentioned areas of products and services depend on new
technologies.
The roads, irrigation, and agricultural domains are significantly supported by
heavy-machinery developed by Caterpillar Inc (I worked for Caterpillar for 2
years as a trainer of technical writing and translation for their whole
line of
machine products) and its competitors. Such companies are on the cutting edge
of new technologies. In the 2 years I worked on systems, we saw transitions
through 3 different types of new directional steering systems (steering wheel
to joystick to hand-inside-of-driving-glove) being developed and implemented.
Phone and internet networks are constantly evolving and growing. In general,
all sectors are influenced by new technologies.]
To such ministers, learning to write A, B, C in Haitian Creole probably
doesn't seem like a viable economic avenue to take. They will simply say
"Just have the people learn to read and write in French" (yes, I do know
that it is not the native language of the majority of the people). The
question here is how can one turn "language" into a product that is going
to attract the attention of the ministers to really do something about it.
Well, it is necessary to develop products and services.
When I have given full bilingual English/Haitian Creole dialogue
demonstrations of the bidirectional speech translation systems that we
developed at Carnegie Mellon University in 1996-98, people were always
amazed. You see, no one can call Haitian Creole "broken French", because
it is "impossible" to translate from "broken French" into English using a
computer system. Those demonstrations automatically raised the level of
awareness of decision-makers in major international corporations about what
a language is. To them, from their worldview of global marketing, Haitian
Creole can be considered a full-fledged language because there are computer
systems that can do something with it. Sure, I know that no language is a
language because it is in a computer, but most people think from the
perspective that a language must really be a language if the computer can
do something with it.
However, we have a problem: When a (written) language is in such a state
of transition as I have discussed in past messages and in all my studies on
language "statistics", it is very difficult, and virtually impossible, to
develop a functional spell-checker for it. How can one convince Microsoft
to invest money in developing Kreyol versions of Word, Excel, Outlook,
Access etc. when the no one can decide on what is the acceptable spelling
of the menu items and words within the program? These are "real" questions
that people ask. Thus, lexical standardization is a significant issue. I
deal with this on a daily basis in working with terminology databases for
all of the European languages and with companies such as IBM, Intel,
Microsoft, Sony, Ford, Renault, Matra, Nortel, General Motors, Siemens,
Philips, etc.
The way to get government support, backing and funding is to develop
prototypes of language products, give demos of them that work and are
functional, and show that much more can be done in a limited, and
reasonable time frame. Such technologies will lead to the expansion of the
Haitian Creole language on the Internet, thus promoting Haiti, Haitian
culture, Haitian government, Haitian economy, Haitian import and export,
etc.....
Take into consideration the following statement:
The European Commission has claimed that languages which do not take a full
part in the electronic media are doomed to stagnate, if not atrophy:
.many of the minority languages are experiencing difficulties, often under
the influence of changing patterns of communication. Penetration of the
new technologies could substantially accelerate this process, threatening
to diminish the linguistic and cultural diversity of European society.
.The rapid rise in use of information and communication technologies will
naturally favour languages which can be successfully processed. Languages
supported by key software products offering powerful facilities for
manipulating text also provide almost unlimited access to information
services in those languages.The long-term viability of languages not
specifically supported is therefore put at risk.
EC proposal for a Council Decision, Multilingual Information Society,
1995
I mentioned last week at the Etudes Creoles conference that all
languages
from now on will be significantly altered by new technologies. It is
one
of "the" primary means by which languages will thrive and survive on
the
global market. Those that have internet sites will promote
themselves.
Those that do not may get left behind.
[Jeff adds comment here: And I would like to emphasize to the ELL list that I
said "thrive and survive on the global market" in my statement above. I did
not
say that the language might not survive in the own local areas. Yet, I do see
in many cultures a movement to global awareness and a desire to acquire the
languages and concepts of the "so-called" international languages and cultures
in order to give the greatest chance to their children for success. This is
reality. I have many friends from different countries in West Africa that
only
speak in the local language with their parents, but speak in French with their
siblings and with their friends. This will greatly influence the future
generations because we pass to our children the language that we speak the
most
and best, unless we specifically make efforts to do otherwise. Even in my
current context, I have not spoken English at home for the past 5+ years
because my wife and I have used French while living in the US and in France.
At work, I do not use English, except in e-mail messages and reports with
external organizations. So, when I have children, it will be comfortable to
speak to them in French -- my complete matrix language -- although English is
my "so-called" mother tongue. I will have to make the effort to speak to them
in English, although I will probably always speak to my wife (their mother)
and
her entire family in French. It is inevitable that psychological,
psycholinguistic, sociolinguistic and other extralinguistic factors (like
economy) play a significant role in language choice. These types of factors
are what mold and form the future at micro and macro levels.]
And can we say that Haitian Creole as a language of the masses will not die
off? Well, all of those people who strive to get their kids educated in
French in the country will do so for the economic well-being of their
children. And those kids will grow up thinking that French is what they
need to pass on to their children. And in a few generations, with that
kind of attitude, Haitian Creole could run the risk of being much less
spoken than it is today. Language survival is based on motivation and
attitude, and motivation and attitude are significantly influenced by
living conditions and the economy.
I therefore see no problem in promoting the Haitian Creole language at an
international level, promoting internet sites that market it at a global
level, making ministers in Haiti aware of the potential of this language at
all levels, and convincing managers in major software and technology
corporations/companies that Haitian Creole is just like any other language
and that it also can be invested in.
[here my correspondant stated that the Haitian culture is not yet at the
computer age and that the people will not use a computer system or machine
that
they will need to adapt their language to]
But you need to look beyond a simple bilingual translation software system.
I am not talking about a single system developed at a single university.
I am talking about major influences on the global economy for the coming
century. I am talking about Internet sites, Internet language-specific
search engines, voice dictation software, spell checkers, information
retrieval, multi-media and multi-modal applications, computer-aided design
and graphics, workflow applications, authoring systems and desktop
publishing, computer assisted language learning (CALL), MIDI systems and
music software, scanners and optical character recognition (OCR), etc.....
You cannot say that these different types of systems and applications do
not affect you, other Haitians, the president and the ministers in Haiti,
import and export companies, tourism in Haiti, Haitian telecoms,
newspapers, radio, TV, etc..... All of this WILL happen through a computer
during the coming century. The typewriter is obsolete. It is a rapidly
dying machine. It is therefore necessary that Haitians will have access to
the technologies and those in colleges and universities in other countries
work together and with other non-Haitians in a networking and partnership
way to promote the Haitian Creole language, Haiti, etc.
Jeff Allen wrote:
> "My use of Kreyol texts is not to judge the intellectual or factual
>level of the content, but rather to gather all available texts and
>through computational means to determine some correlation between
>literacy,
>education, reading, writing and the resulting texts that are available."
>
[here my correspondant stated that such mechanical and statistical studies are
simply something that stays in a laboratory and that they are not valuable for
people who themselves use the language]
I deal with the global marketing of languages on a daily basis. I estimate
the market segmentation of language technologies for every major language.
I gather information on every player in all of the different sectors
mentioned further above. Yes, these are all statistics, otherwise known as
facts. My research on Haitian Creole has been based on facts and
statistics because there are too many people out there who base their
evaluations on intuitions, on subjective feelings, etc. I see
presentations all the time where people say that X is more than Y, but
don't show any real figures for it. How can you prove it? Show me the
figures. Basic intuition just doesn't cut it when you are dealing with
top-level decision-makers. The Haitian minister of education will laugh at
such subjective, intuitive comments, but if you present hard cold facts
about language to him, along with a functional system, and significant
promise for a real product that can be objectively evaluated on benchmark
tests and ongoing tests, then the facts and statistics can lead to
financial investment and full-support.
Statistics are what managers, ministers, presidents, and others want to see
because numbers speak out loud very clearly. People make decisions every
single day based on numbers and statistics presented in spreadsheets,
databases, pie charts, etc, and these decisions are those that affect the
economy and the lives of nearly every single person on this entire planet.
What I have been trying to do for Haitian Creole over the past few years is
show that it has great potential for the world market. All that is needed
now are a number of talented native Haitians with a real desire to do
something for their language, for their country, and who can take my work
and do something with it.
[here my correspondant stated that dealing with people, with mentality of the
people, with what they do, and with language is subjective]
But those subjective issues are not what the decision-makers are going to
make promises and decisions on. I deal with such decision-makers every day.
They want objective facts.
[here my correspondant stated that I am emphasizing the non-importance of
education, of socio-economic and development, and that I do not touch on
these]
I think that I have made myself clear that my "statistical studies" on
lexical frequency in Haitian Creole are in fact crucial to the marketing
potential of the language, of the country, and of its economy for the future.
All of the points I have raised above indicate that the computer creates
products, services, jobs, and increases the growth of the economy. It may
not reach every single Haitian. But then, does every French kid in France
have a computer (no, only 9% of households have computers in France). But
what is important is opening the opportunity to as many as possible, and
ensuring that the Haitian Creole language will have a long life expectancy.
----
I do not profess to have all of the answers or the perfect solution, but I
believe that there are strategies to be developed that can be based on
previous
experience in other situations.
Best,
Jeff
=================================================
Jeff ALLEN - Technical Manager/Directeur Technique
European Language Resources Association (ELRA) &
European Language resources - Distribution Agency (ELDA)
(Agence Europe'enne de Distribution des Ressources Linguistiques)
55, rue Brillat-Savarin
75013 Paris FRANCE
Tel: (+33) 1.43.13.33.33 - Fax: (+33) 1.43.13.33.30
mailto:jeff at elda.fr
http://www.icp.grenet.fr/ELRA/home.html
----
Endangered-Languages-L Forum: endangered-languages-l at carmen.murdoch.edu.au
Web pages http://carmen.murdoch.edu.au/lists/endangered-languages-l/
Subscribe/unsubscribe and other commands: majordomo at carmen.murdoch.edu.au
----
More information about the Endangered-languages-l
mailing list