Language Resources and Localisation 
Reinhard SCHÄLER 
Localisation Research Cen
Department of Computer Science and Information Systems (CSIS)
University of Li
Limerick, Irel
Reinhard.Schaler@ul.ie 
 
 
Abstract 
Localisation is one of the fastest growing 
industrial sectors in the digital world. Since 
the mid-eighties, the role of localisation has 
developed and changed dramatically. 
Localisation has been redefined as the 
provision of services and technologies for the 
management of multilinguality across the 
global information flow. This paper discusses 
the need for easily accessible dedicated 
language resources for localisation, provides a 
practical example of what can be achieved 
with appropriate language resources in the 
context of localisation and proposes a strategy 
to acquire, maintain and make them easily 
accessible. 
1 Introduction 
Since its emergence in the mid 1980s, 
localisation has largely been defined as the 
linguistic and cultural adaptation of products for 
specific locales. Over the past 20 years, 
Localisation has become one of the engines driving 
the development of the multilingual information 
society and probably the first industrial sector 
where language resources have been used widely 
and consistently on large-scale commercial 
projects. 
Localisation professionals must prepare very 
large amounts of digital content simultaneously for 
different markets in acceptable quality and at 
affordable costs. This is only possible with the 
support of language resources, such as written and 
spoken corpora, translation memories and 
terminology databases, as well as the appropriate 
software tools for the acquisition, preparation, 
collection, management, customisation and use of 
these resources. 
In the following paragraphs, we will provide an 
overview of the current state of the localisation 
industry and its requirements, and focus on those 
aspects of the localisation process where the 
successful use of language resources is crucial for 
tre (LRC)  
 
merick 
and 
the timely delivery of multilingual digital content. 
We will describe some of the most widely used 
language resources in localisation. Finally, we will 
describe major current research efforts relating to 
language resources and localisation, and highlight 
the opportunity for the establishment of a 
Language Resources Centre for Localisation. 
2 The changing face of localisation 
In this section, we will provide some 
background information which will help to explain 
the reasons for the changing needs of the 
localisation industry, particularly in relation to 
language resources. 
2.1 Growing language populations create 
commercial opportunities 
Any content made available on the digital 
networks (e.g. the internet) becomes instantly 
available to millions of people across the globe. To 
make this content accessible, however, it needs to 
be localised. In today’s cyberspace, posting digital 
content in just one language is not sufficient 
anymore. 
Of a total online language population of 680 
million in September 2003, only 35.5% spoke 
English as a mother tongue, but 25% spoke 
Chinese, Japanese or Korean (CJK) and another 
Source: Global Reach (global-reach.biz/globstats)
 
Figure 1: Online language population
25% Spanish, German, French, Italian and 
Portuguese. 
In this scenario, localisation is a pre-requisite for 
the provision of equal access to the digital 
information society independently of an 
individual’s cultural and linguistic background or 
their geographical location and, at the same time, 
offers enormous business potential. In addition, it 
is evident that localised digital content 
(applications and systems) is a pre-requisite for the 
preservation of linguistic and cultural diversity in 
the digital world. 
According to US-analysts Allied Business 
Intelligence, the world-wide market for translation 
and software or web localisation is growing from 
US$11 billion (1999) to US$20 billion this year. 
However, although much digital content is 
created in Asia and Europe, 95% of localised 
digital content still originates in the USA. 
 Localisation
Vectors of scalability and growth
Geography / Languages
Content
Medium of delivery
Europe
Documents
Manuals
Asia
Global
CD-ROM
Online
Pure Internet-based
General
technical
Any
content
Culture
Symbols
Rights
Values
 
Figure 3: Vectors of scalability and growth 
2.2 Localisation services – redefined 
While politicians all over the world want to 
make Information Society Technologies (IST) 
available and accessible in the language and locale 
of the people they represent, software and digital 
content publishers need to respond to the demands 
of their customers by supporting a wide variety of 
local languages and cultures in their products. 
 
Since its emergence in the mid-1980s, the 
localisation industry has taken on the task of 
responding to these requirements of business and 
politics. Initially seen as just one of many service 
suppliers to the general IT sector, it is now taking 
on a more independent role translating the global 
digital content challenge into new business and 
social opportunities. It is the localisation industry 
that can enable the open, pluralist, user-friendly 
and inclusive multilingual and cross-cultural 
information society. 
The dramatic change in the role and the funtion 
of localisation has happened in parallel with the 
development and changes of the IT and content 
publishing sectors in general, now all converging 
in the digital world. 
 
The businesses of computing (hardware, 
software and services), communications 
(telephony, cable and satellite), and content 
(publishing, entertainment, advertising) are coming 
together to create the new digital media industry. 
New media publishing on the Internet combines 
computing, entertainment, broadcasting, music and 
video production. 
The issues faced by a wide variety of formerly 
independent, unconnected traditional content 
publishing industries joining in the digital world 
include the need to handle, control and translate 
larger amounts of text than ever before into an ever 
increasing variety of languages in parallel with the 
development of the original version, within a tight 
budget and according to strict quality guidelines as 
well as the need to adapt — not just translate — 
their products to the culture and locale of the target 
market. 
IT provides the framework for the convergence 
of these activities. The localisation industry 
provides the framework for the convergence of the 
multilingual aspects of these activities. 
Localisation becomes the catalyst for electronic 
multilingual production and publishing. 
On the background of these developments, the 
concept of localisation is being redefined as the 
provision of services and technologies for the 
management of multilinguality across the global 
information flow. 
Timely and cost effective delivery of high 
quality digital content to the global marketplace 
has become the major growth area for the 
localisation industry. It opened the relatively 
narrow software localisation industry to a wider 
range of players who are broadening traditional 
roles within the software localisation industry. 
Yet, the localisation industry does not have 
access to a robust infrastructure comprising 
Multilingual software = Multilingual information society
Preservation of
linguistic diversity,
cultural heritage
Development  of
business growth,
competitive advantage
Inclusion
of all countries and sections of society 
in the global village
Political focus Business focus
Figure 2: Multilingual software and Infosociety 
language data and tools, which are a prerequisite 
for the timely and cost effective creation and 
deployment of multilingual, cross-cultural and 
multimodal digital content. There is an urgent need 
for the development of the structural basis to make 
a sustained internationalisation and localisation 
effort possible, especially for less widely spoken 
languages. 
2.3 Localisation research 
The new role of the localisation industry also 
creates new opportunities and requirements for 
research and development. 
How can the business requirement for a 
reduction in production cost, combined with fast 
throughput time and high quality be satisfied in the 
context of localisation? 
One answer to this question is by access to 
adequate language resources, tools and standards. 
Re-use of already translated material, translation 
recycling, can not only speed-up translation, it can 
also at least help to ensure a more consistent use of 
terminology and thus increase the quality of the 
translation. 
Access to adequate terminology tools and 
resources can also help to achieve better 
translations at a lower cost. 
To make appropriate use of these technologies 
and resources, their role must be assessed in the 
context of localisation. 
3 Language resources and localisation 
Language data or resources have been defined as 
a set of speech or language data and descriptions in 
machine readable form, used e.g. for building, 
improving or evaluating natural language and 
speech algorithms or systems, or, as core resources 
for the software localisation and language services 
industries, for language studies, electronic 
publishing, international transactions, subject-area 
specialists and end users. 
Examples of language resources are written and 
spoken corpora, computational lexicons, 
terminology databases, speech collection and 
processing, etc. Basic software tools are also 
important for the acquisition, preparation, 
collection, management, customisation and use of 
these language resources and other resources. (see: 
http://www.elra.info/article.php3?id_article=35, 
consulted May 2004) 
Currently, there are no large collections of 
language data relevant to localisation easily 
accessible. While individual digital content 
developers, especially large multinationals, have 
very large collections of multilingual language data 
available and use them in a highly effective and 
efficient way, these data collections are not 
available to the localisation community in general. 
Large multinational content publishers, however, 
have shown how the efficient use of language 
resources can help to achieve astonishing 
production results leading to what some have 
described as the translation factory. 
• Start: mid-eighties
– Packaged softward -> multimedia -> content
• Ireland: the world centre (certainly the European centre)
• 95% of source orginates in the USA
• International market more important for pubishers than domestic markets
• MS: >60%, >US$5b ., >1,000 projects/year, Ireland: US$1.9b revenue (2001)
Revenue per Service-Segm ent
(in US $m )
0
500
1000
1500
2000
2500
3000
2000 2001 2002 2003 2004 2005
Softw are Localisation Web Localisation Translation Interpretation
Source: IDC, Worldwidc Globalization and Localization Services Market Forecast and Analysis, 2000-2005, 2001, www.idc.com
Figure 4: Revenue per service segment 
3.1.1 Case study 
Tony Jewtushenko, Tools Manager with 
Oracle’s Worldwide Translation Group, presented 
the following example of the use of language 
resources in localisation at the LRC’s 2003 Annual 
Localisation Conference. 
Project constraints 
 � 4m wordcount software strings; 
 � 30 languages simultaneous release; 
 � 13k localisable files; 
 � Localisation group in Dublin; 5,000 people 
world-wide distributed development team. 
Objectives 
 � 24/7, 100% automated process – no 
exceptions 
 � Translation in parallel with development 
 � Translation begins at code check-in 
 � Translation “on demand” – no more “big 
project” model 
Solution: the translation factory metrics 
 � Current throughput: 100,000 language 
check-ins per month 
 � 2 million files per month 
 � 98% of words leverage 
 � Average time to process a file: 45 seconds 
 � Fully scalable “add-a-box model” 
 � Simpship of all 30 languages 
 � International version testing before US 
release 
 � Reduced no. of release engineers (20->2) 
resulting in US$20m saving per year 
 � Positive ROI within 1 year 
It is important to keep in mind that Oracle is one 
of the world’s leading digital publishers and runs 
one of the most sophisticated internationalisation 
and localisation operations in the world. Oracle is 
also centrally involved in the development of two 
key standards under the umbrella of OASIS, the 
XML-based Localisation Exchange Format 
(XLIFF) and the Translation Web Services Group 
(TWS), which, combined, have the potential to 
fundamentally change not just the way localisation 
is done by Oracle, but by every digital publisher 
bringing its contents to the global market. 
3.1.2 Support for SMEs and researchers 
There are no good reasons why the work of other 
organisations, such as small and medium sized 
enterprises (SME’s) and research organisations, 
should not benefit from the intelligent use of 
linguistic resources, including language data, tools 
and standards. While large organisation catering 
for the main language markets have access to the 
finances necessary for the development and 
maintenance of this linguistic infrastructure, 
smaller organisations and those catering for 
financially less significant markets will need the 
support of a shared and widely supported 
infrastructure. 
This infrastructure would need to cover: 
 � Multimodal digital content in source and 
target languages; 
 � Monolingual and multilingual 
terminology; 
 � Translation memories. 
3.2 Linguistic tools 
Linguistic tools are seen by the Localisation 
Industry Standards Assocication (LISA) generally 
as still an emerging sector (Localization Industry 
Primer) although, according to LISA, enormous 
progress has been made over the past years in the 
area and a number of productivity enhancing tools 
are now in use, without which the localisation 
industry as we know it today would not be able to 
operate. 
3.2.1 Current situation 
The issues, which are addressed by linguistic 
tools and technologies answer some of the central 
questions asked by localisation professionals 
around terminology handling and update 
processing. 
Terminology handling 
 � Where can translators find standard 
terminology in multiple languages? 
 � How can multilingual terminology be 
processed so that it can be made readily 
available and easily accessible? 
 � Are there feasible models and mechanisms 
to maintain and constantly update 
multilingual terminologys so that 
modifications can be made accessible to 
translators instantaneously? 
 � How can changes in previously agreed 
terminology be automatically integrated in 
already translated text? 
 � How can translated texts automatically be 
checked for the correct and consistent use 
of terminology? 
Update processing  
 � How can versions of the same source 
material be compared against each other 
automatically? 
 � How can overlaps be identified, marked 
and analysed? 
 � How can source and target language 
content be compared and aligned? 
 � How can already translated text fragments 
(exact or fuzzy matches) automatically be 
used for the generation of a new target 
version? 
 � How can the limitations of Translation 
Memory Systems be overcome?  
The linguistic tools and technologies most 
widely needed and developed for use in the 
internationalisation and localisation effort include: 
 � Terminology management systems, which 
aid the collection and use of specialised 
vocabularies; 
 � Translation memories, which are designed 
to facilitate the reuse of previous 
translations; 
 � Machine translation, which provides actual 
linguistic analysis and conversion of texts 
from source language into the desired 
target language; 
 � User interface and user assistance visual 
translation environments, which aid 
translators to interactively work with 
compiled and uncompiled resource files in 
a variety of formats; 
 � Language data analysis tools, which 
rapidly compare and analyse old and new 
source material; 
 � Sophisticated matching tools leveraging 
material from previous projects; 
 � Natural language parsers; 
 � Extract-and-Insert tools; 
 � Parsers for natural language digital content 
in compiled sources. 
While large and sophisticated localisation 
operations have easy access to relevant linguistic 
third party technologies and in-house tools, smaller 
operations often do not. Reviewing the large 
variety of sophisticated tools and technologies 
available on the market, they often shy away from 
the purchase and implementation of tools because 
of the perceived high-risk factor attached to their 
deployment. 
3.2.2
allow s
direct and online acces
tools and technologies and detailed inform  
about these.  
this infrastructure has been  
Localisation Tools and Technolog
Showcased (
ELECT project. 
such as the Globalisat
Association  
Localisation 
the expertise available within the ELECT
consortium
detailed back
tools available was prepar
indivi
potential users of linguistic tools and technolo
specifically
sized enterpri
tool to use for their particular localisatio
independent, 
inform
3.3
establish
association, the Localisat
Association 
professionals recognised that the
im
would l
time ne
projects and increase the 
delivered to international audiences. 
standards for linguistic resources pl
overall localisation process (source: 
www.i18n.ca): 
 
3.3.1 Current situation 
A large number of standards relevant to 
linguistic resources in the context of localisation 
have been published by a number of organisations 
identified by LISA as being involved in the 
development of standards. Among these are: 
International Standards Organisation (ISO) – 
This is a network of national standards institutes 
from 140 countries working in partnership with 
international organizations, governments, industry, 
business and consumer representatives. The ISO 
sees itself as a bridge between public and private 
sectors. 
 Impact 
A support infrastructure must be put into place to 
maller players involved in localisation 
s to the widest variety of 
ation
A first step in the implementation of
 the establishment of the
y Laboratory and 
LOTS) as part of the European-funded 
Cooperating with leading industry associations, 
ion and Localisation 
 (GALA) and The Institute of
Professionals (TILP), and building on 
 
, a sophisticated online library with 
ground information on each of the 
ed and published. 
The detailed LOTS-sponsored reviews of 
dual tools and technologies will allow 
gies, 
 those working in small and medium 
ses, to base their decision on which 
n needs on 
well-researched and easily accessible 
ation. 
 Standards 
Standards played a central role in the 
ment of the localisation industry’s first 
ion Industry Standards 
(LISA). Very early on, localisation 
 successful 
plementation of widely recognised standards 
ower the cost of localisation, shorten the 
cessary for the successful completion of 
quality of the products 
The following graphic visualises the role 
ay in the 
i18n Inc.: 
 � Technical Committee 37 - Terminology 
and other language resources 
 � ISO 639 - Language Codes 
 � Terminology Data Categories - ISO 12620 
 � MARTIF - ISO 12200 - Machine-readable 
terminology interchange format 
 � Terminology Work - ISO 704 
 � Vocabulary - ISO 1087-1 - Part 1: Theory 
and Application 
 � Vocabulary - ISO 1087-2 - Part 2: 
Computer Applications 
 � Terminological Markup Framework - ISO 
DIS 16642 
 � ISO639-1 : New ISO standard for the 
identification of languages names 
Localisation Industry Standards Association 
(LISA) – This organisation has published a variety 
of standards relevant to the use of language 
resources in localistion: 
 � TMX the exchange standard for 
translation memory data between tools 
and/or translation vendors aiming at little 
or no loss of critical data during the 
process. 
 � TBX the open XML-based standard 
format for terminological data. 
 � OLIF the XML-compliant standard for 
terminology offering support for natural 
language processing (NLP) systems, such 
as machine translation, by providing 
coverage of a wide and detailed range of 
linguistic features. 
 
Figure 5: Standards in the localisation 
OASIS – This international, not-for-profit 
consortium designs and develops industry standard 
specifications for interoperability based on XML. 
Two of these have been developed specifically 
with localisation in mind: 
 � XLIFF – the XML-based Localisation 
Interchange File Format 
 � TWS – the Translation Vendor Web 
Services Standard 
The Free Standards Group Open 
Internationalization Initiative (Openi18n.org) – 
This non-profit initiative aims to accelerate the use 
and acceptance of open source technologies 
through the application, development and 
promotion of interoperability standards. 
Termnet – The International Network for 
Terminology promotes co-operation in the field of 
terminology internationally, so as to stimulate the 
development of the terminology and knowledge 
market, as well as terminology proper. Termnet 
publishes terminologically relevant data in both 
printed and computerised forms and thus makes it 
accessible to a large circle of users.  
Unicode – The Unicode Consortium is a non-
profit organization founded to develop, extend and 
promote the use of the Unicode Standard, which 
specifies the representation of text in modern 
software products and standards. The membership 
of the consortium represents a broad spectrum of 
corporations and organisations in the computer and 
information processing industry. Membership in 
the Unicode Consortium is open to organisations 
and individuals anywhere in the world who support 
the Unicode Standard and wish to assist in its 
extension and implementation. Unicode’s most 
visible activities include the holding of the 
Internationalisation and Unicode conference twice 
a year. 
WC3 – This consortium develops interoperable 
technologies (specifications, guidelines, software, 
and tools) to lead the Web to its full potential as a 
forum for information, commerce, communication, 
and collective understanding. 
3.3.2 Impact 
Currently, there is not central repository of 
standards relevant to the development and 
maintenance of linguistic resources for localisation 
comprising language data, tools and standards 
which is easily accessible to the localisation 
community. Furthermore, and equally important, 
no independent organisation or consortium is 
currently set up to demonstrate the effective and 
efficient use of linguistic resources in a localisation 
environment following industry-standard 
approaches and using state-of-the-art technologies. 
4 The Localisation Tools, Technologies and 
Resources Laboratory 
4.1 The Rationale 
The establishment of the Localisation Tools and 
Technologies Laboratory and Showcase (LOTS) as 
part of the European-funded ELECT project was 
the first attempt to make a repository of language 
resources covering linguistic data, tools and 
standards available and easily accessible. 
Although LOTS is located at the Localisation 
Research Centre in Limerick, it is also available 
online via www.electonline.org. 
In one location, LOTS provides different user 
groups with access to the widest possible range of 
tools and corresponding resources. 
Figure 6: The LOTS online desktop 
 � Students, Trainers can quickly get an 
overview of tools, technologies and 
resources relevant to localisation. 
 � Researchers can experiment with state-
of-the-art technologies and resources 
comparing the results of their efforts 
with commercial offerings. 
 � Professionals can test whether particular 
applications are appropriate to cover 
their specific needs. 
 � The LRC uses the facilities available in 
LOTS to verify standards and 
interoperability issues. 
4.2 Tools and Technologies 
LOTS was established with the support of the 
localisation tools and technology developers. All 
the resources available to LOTS have been given 
to the LRC by their owners free-of-charge. 
Twenty-four companies, resprenting the majority 
of localisation tools and technologies worldwide, 
have so far contributed to LOTS. 
 
Figure 7: The LOTS laboratory 
The LRC has estimated that the laboratory 
contains tools and technologies worth more than  
€350,000. 
These are accessible on a large number of PCs 
running in LOTS under a variety of operating 
systems in more than a dozen different languages. 
They are also available online on the LOTS 
server which brings LOTS directly to the desktop 
of users anywhere in the world. 
4.3 Resources 
In addition to the tools and technologies, the 
resources available on the LOTS server and in the 
laboratory include the most widely used file 
formats and standards. 
While tools and technologies are well covered in 
LOTS, the coverage of corresponding resources as 
used by these tools could be improved. 
The LRC is working with content and 
technology developers on agreements which will 
allow a wider deployment of authentic source 
material in a wider range of languages and file 
formats. 
5 Language Resource Centre for Localisation 
Although with LOTS the LRC has established 
the foundations for a language resources centre for 
localisation, a consortium representing the main 
actors in localisation, based in a number of 
European countries, including the accession 
countries, will need to be established to develop 
and maintain such a repository long-term. This 
consortium will also need to develop long-lasting 
relations with the leading industry associations and 
standards bodies, e.g. the Globalisation and 
Localisation Association, LISA, OASIS, Unicode, 
W3C and The Institute of Localisation 
Professionals (TILP). 
The centre, proposed to be located at the LRC, 
will provide access to 
 � Linguistic resources 
 � Methodologies and guidelines 
 � Localisation scenarios 
Each of these components will be explained 
more in detail in the following sections. 
5.1 Repository of linguistic resource 
The centre will establish, develop and maintain a 
repository of linguistic resources for localisation 
physically based at the LRC but accessible online 
over the Internet covering: 
 � A large variety of multimodal authentic 
digital content in source and target 
languages; 
 � Mono- and multilingual terminology; 
 � Translation memories; 
 � Linguistic tools and technologies used for 
the automatic processing of digital content; 
 � Guidelines and standards for the 
development and processing of digital 
content to be localised. 
Together, the efforts in these areas will deliver 
the structural basis for a sustained 
internationalisation and localisation effort, 
especially for less widely spoken languages where 
market forces often provide insufficient incentives. 
5.2 Methodologies and guidelines 
The centre will provde access to methodologies 
and guidelines for the verification of standards 
compliance and interoperability verification for 
linguistic resources in a multilingual, multicultural 
and multimodal localisation environment. This 
work will be based on industry-standard 
approaches and be guided by established principles 
and procedures. The reports will be sourced in 
cooperation with relevant associations and 
standards bodies and public deliverables of 
European Union funded projects. 
5.3 Localisation scenarios 
The centre will, in consultation with the wider 
localisation community, build a laboratory-based, 
automated localisation environment mirroring real-
world, authentic localisation scenarios. This 
environment will be used to showcase, verify and 
demonstrate best practice in localisation making 
use of the linguistic resources available through the 
centre. 
Results from each of these three areas of activity 
will be made available to the wider digital content 
and localisation communities using different 
dissemination strategies. It is envisaged that the 
development of a market place for these linguistic 
resources will guarantee the sustainability of the 
effort. 
6 Conclusion 
We have shown how the needs and requirements 
of localisation have developed over the years. The 
enormous pressure on localisation providers to 
produces language versions of original material 
simultaneously with the production and publication 
of the original can only be addressed making 
efficient and effective use of customised language 
resources covering linguistic data, tools and 
technologies, and appropriate standards. 
Access to language resources for SMEs and the 
research community can be realised within a 
widely supported Language Resource Centre for 
Localisation, built on the foundations of the 
Localisation Tools and Technology Laboratory 
(LOTS) at the LRC. 
The establishment of the Language Resources 
Centre for Localisation is a five-year project. 
During this time, the LRC plans to create a 
sustainable, accessible and financially viable 
linguistic infrastructure for the internationalisation 
and localisation communities. 
Its overall aims are to: 
 � Pool together linguistic infrastructure 
resources for the localisation industry, 
including digital content, monolingual and 
multilingual terminology, translation 
memories, tools and technologies, as well 
as relevant standards. 
 � Establish a linguistic resources support 
network within the localisation industry 
covering digital content publishers, service 
providers, technology developers, 
standards bodies and standards verification 
initiatives. 
 � Provide convenient access to relevant 
linguistic resources for content developers, 
service providers, as well as suppliers of 
localisation services and solutions. 
 � Develop methodologies for the verification 
of standards compliance. 
 � Implement localisation scenarios in a 
laboratory environment aimed at 
demonstrating state-of-the-art, best 
practice localisation technology and 
process solutions, and at verifying relevant 
localisation standards. 
 � Work towards the establishment of a 
market place and a viable linguistic 
resources provider network for the 
localisation industry. 
Activities during years 1-2 will focus on the 
establishment of a repository of linguistic 
resources. During this period, the LRC will build a 
core group of partners supporting its development. 
This core group will be backed by the LRC’s 
Contact Group. 
Activities during years 3-5 will see the 
implementation of a financially viable standards 
testing and interoperability verification system at 
the LRC. While few commercial organisations 
have managed to make a profit to sustain this kind 
of activity in other industrial sectors, there is no 
doubt that the operation of such a centre is 
financially viable long-term in a not-for-profit 
environment such as that provided by the LRC at 
the University of Limerick. 
7 Acknowledgements 
We would like to acknowledge the support of the 
European Union for the European Localisation 
Exchange Centre (ELECT) under its eContent 
Programme, contract 52005. 

References  
Electonline – the online resource for the 
localisation community. www.electonline.org, 
last visited 10 June 2004. 
Language resources: the ELRA definition. 
http://www.elra.info/article.php3?id_article=35, 
last visited 20 May 2004. 
Localization Industry Primer. LISA. 
www.hltcentral.org/usr_docs/call_docs/ 
eContent/call1/LISA%20Primer.pdf, last visited 
13 May 2004. 
Schäler, R., Michael Carl, Andy Way. 2002. 
Toward a Hybrid Integrated Translation 
Environment. In: “Proceedings of the Fifth 
Biennial Conference of the Association for 
Machine Translation in the Americas Conference 
(AMTA)”, Tiburon, California, 08-12 October 
2002.  
Schäler, R. 2002. The European Localisation 
Exchange Centre (Keynote). In: Proceedings of 
the Twenty-first International Unicode 
Conference, Unicode Localization and the Web: 
The Global Connection, Dublin, Ireland, 16-17 
May 2002. 
Schäler, R. 2002. The XLIFF Standard. Panel 
Session with Ian Dunlop (Novell), Tony 
Jewtushenko (Oracle), Wojtek Kosinski and 
Peter Reynolds (Bowne), in: Proceedings of the 
Twenty-first International Unicode Conference, 
Unicode Localization and the Web: The Global 
Connection, Dublin, Ireland, 16-17 May 2002.  
Standards in the localisation process. Graphical 
representation. www.i18n.ca, last visited 10 May 
2004. 
