ArticlePDF Available

Translation Challenges in the Localization of Web Applications

Authors:

Abstract and Figures

This preliminary study aims at exploring the nature of challenges that translators face when they take on a localization project of a web application. Taking into account that localization is an activity constrained by time, process and economic resources, translators need to make use of their full skill set to overcome the various challenges imposed by the source text and the localization process itself. For the purpose of this study, an ad hoc monolingual English corpus composed of the user interface strings of web applications has been used. Since multiple types of challenges are found in a localization project of this nature, this paper focuses on those related to internationalization practices and to constraints imposed by the translation memory segmentation process. Although localization is a mature field and a great deal of guidelines and best practices is available for content creators and tool developers, as found in this qualitative study, localizers can still suffer the consequences of deficient internationalization practices and non-ergonomic translation tools.
Content may be subject to copyright.
235
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266
Translation Challenges in the Localization of
Web Applications
Elena de la Cova
mecovmor@upo.es
Universidad Pablo de Olavide, Seville
Recibido: 20/06/2016 | Revisado: 20/07/2016 | Aceptado: 08/11/2016
Abstract
This preliminary study aims at exploring the nature of challenges that translators face when
they take on a localization project of a web application. Taking into account that localization
is an activity constrained by time, process and economic resources, translators need to make
use of their full skill set to overcome the various challenges imposed by the source text and
the localization process itself.
For the purpose of this study, an ad hoc monolingual English corpus composed of the user
interface strings of web applications has been used. Since multiple types of challenges are
found in a localization project of this nature, this paper focuses on those related to interna-
tionalization practices and to constraints imposed by the translation memory segmentation
process.
Although localization is a mature eld and a great deal of guidelines and best practices is
available for content creators and tool developers, as found in this qualitative study, localiz-
ers can still suffer the consequences of decient internationalization practices and non-ergo-
nomic translation tools.
Keywords: localization, web applications, challenges, internationalization, translation memory systems
Resumen
Desafíos de traducción en la localización de aplicaciones web
Este estudio preliminar tiene como objetivo explorar la naturaleza de los desafíos a los que
los traductores se enfrentan cuando se embarcan en un proyecto de localización de una apli-
cación web. Dado que la localización es una actividad condicionada por tiempo, procesos
y recursos económicos, los traductores tienen que poner en marcha todas sus competencias
para superar los muchos desafíos impuestos por el texto fuente y por el proceso de localiza-
ción en sí.
En este estudio, se ha utilizado un corpus monolingüe ad hoc en inglés compuesto por men-
sajes de la interfaz de aplicaciones web. Puesto que en este tipo de proyecto de localización
existen diferentes tipos de desafíos, este artículo se centra en aquellos relacionados con las
prácticas de internacionalización y con la segmentación de las memorias de traducción.
A pesar de que la localización es un ámbito de considerable madurez y de que los creadores
de contenido y los desarrolladores de herramientas tienen a su disposición un gran abanico
de instrucciones y directrices, en el presente estudio cualitativo se concluye que los localiza-
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
236
dores siguen sufriendo las consecuencias de prácticas de internacionalización decientes y
de herramientas de traducción poco ergonómicas.
Palabras clave: localización, aplicaciones web, desafíos, internacionalización, sistemas de memoria de traduc-
ción
1. Introduction
The localization industry stresses the importance of cultural adaptation of prod-
ucts and services for their success in target markets (DePalma 2006; Schäler 2007).
Unfortunately, the impact of localization on a given business often becomes visible
in society only when its results are poor. There are a number of authors that have
explored translation mistakes in localization. Diéguez and Lazo (2011) analyze suc-
cesses and errors found in localized websites and categorize the translation techniques
followed according to Hurtado Albir’s classication (2001: 266-271). Jiménez-Cre-
spo has written extensively about the corporate website as a text genre from different
perspectives (2008; 2009a; 2009b; 2011a; 2011b; 2012; 2013), and has proposed a
holistic error typology based on a monolingual comparable corpus of Spanish orig-
inal and localized corporate websites. His research provides an additional empirical
methodology to current quality assessment systems in the localization industry, which
lack a comprehensive outlook and could benet from aspects identied in his error
typology such as pragmatic errors (Jiménez-Crespo 2011a). Additionally, the quality
of website localization has also caught the attention of researchers like Medina and
Ramírez (2015), who have evaluated the quality of the “Products” section from SME
corporate websites.
Pym (2010) highlights the lack of studies about translation mistakes in localization,
despite the severe consequences these mistakes have on businesses. A case to illustrate
this took place in 2015, when the well-known Spanish retailer Zara (from the Inditex
retail group) published in their German online shop a new model of sandals called
dreifarbige Sklaven Sandalen” (direct translation from the Spanish “sandalias de
esclava”, in English “three-colored slave sandals”). According to the RAE Spanish
dictionary (Royal Spanish Academy), the word “slave” (“esclava”) in Spanish can
also refer to a kind of simple bracelet design. Therefore, Zara apparently used the
word “slave” in an attempt to compare their sandal design to that of the bracelet. How-
ever, translating “sandalias de esclava” literally into “dreifarbige Sklaven Sandalen
immediately triggered the offense from the German audience because “Sklaven” does
not have the same meaning as in Spanish (type of bracelet). This error went viral on
social media (as can be seen on Twitter under the hashtag #sklavensandalen), and
Zara had to issue a press release explaining that they do not promote slavery and it
was a translation mistake. It was, nevertheless, too late and criticism spread to other
business areas, such as the working conditions of their labor in developing countries.
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 237
Although they apologized and removed the word “slave” from the design name, the
company reported a decrease in their sales in Germany the days following the “trans-
lation mistake” (lavanguardia.com, 2015). Germany is one of Zara’s biggest markets
and therefore this issue had severe consequences.
In order to provide content with optimal quality and adequate localization, transla-
tors need to have the right competences and tools. Conversely, the reality of the trans-
lation and - especially - the localization industry does not always allow translators to
make the most out of their translation skills. Deadlines are usually tight, translation
instructions might not be very clear and the translation environment limits the way
translators work. According to Alonso (2016), professional translation environments
might currently be ruled by conict, opacity and mistrust between translators and
direct clients or managers, due to the computer-mediated nature of communication
and of translation production networks (Abdallah and Koskinen 2007). Localization
is often the last stage before content is published (Herrmann and Sachse 2005: 69)
and, depending on the process, it may happen that localizers do not have the freedom
or the possibility to make the changes necessary to adequately adapt the content to
their markets. Additionally, communication barriers between authors and localizers
frequently prevent feedback to ow along the globalization process, which would be
greatly benecial for content creation (Fenstermacher 2006: 82).
This paper aims at exploring the challenges that localizers face when they are
exposed to the localization of a web application project. Given that there can be a
myriad of challenges in this process, this qualitative study focuses on how the “inter-
nationalized text” (Pym 2004), or the lack of proper internationalization, can affect
the process. In addition, since the intersection between translation and technology is
inherent in localization (Jiménez-Crespo 2010) and it affects how localizers work, the
translation environment has been taken into consideration in the present study. In par-
ticular, the translation memory segmentation process has been observed as a potential
challenge that translators have to overcome.
In order to identify translation challenges arising from the localization of web
applications, an English monolingual corpus of web applications has been analyzed.
In this paper the concept of challenge is essentially understood as the notion of trans-
lation problem, which has been widely analyzed by different authors in Translation
Studies, such as Wilss (1982), Lörscher (1991), Kiraly (1995), Presas (1996), Nord
(2005), Toury (2010) or PACTE group (2011), among others. Although exploring this
complex concept in detail is not the purpose of this paper, a denition of translation
problem will be provided to frame the perspective of the study:
A translation problem is an objective (or inter-subjective) transfer task which every translator
(irrespective of their level of competence and of the technical working conditions) has to solve
during a particular translation process.
(Nord 2005: 166)
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
238
As will be seen in this empirical study, many of the translation challenges can be
solved more or less successfully depending on various factors but they are still con-
sidered translation problems. Therefore, according to the perspective of this study, a
translation problem is viewed as such regardless of whether it can be solved or not
and regardless of the experience or “level of competence” of the translator, who might
have been exposed to a particular challenge in the past and learnt to overcome it. That,
however, does not make the challenge itself less of a challenge; it just makes it easier
to sort out for a given translator.
This paper intends to depict some of the challenges that localizers face when they
take on the localization of a web application which, if not dealt with adequately, could
cause localization and translation errors. Therefore, this qualitative study should be
seen as an attempt to explore the previous step to the localized outcome (the trans-
lation stage). Unfortunately, due to non-disclosure agreements signed by translators
- among other factors - researchers might nd it difcult to observe and reect on a
paper what translation challenges localizers deal with in professional projects and how
they solve them (Rojo 2013: 71). This could be seen as a methodological constraint
imposed by the localization industry processes. In this sense, the present qualitative
study aims at exploring the localization of web applications at the translation stage.
In order to do that, an empirical approach based on a corpus - as recommended by
Olohan (2004) and Gamero (2001) cited in Jiménez-Crespo (2011a: 319) - together
with the researcher’s intuition and professional experience in this eld have been used
to determine what constitutes a localization challenge.
Since this study focuses on the localization of web applications, section 2 covers
basic concepts related to localization. Section 3 presents the topic and the theoretical
background behind each of the challenges identied. Section 4 explains the method-
ology followed and illustrates those translation challenges with examples from the
analyzed corpus. Finally, section 5 highlights the conclusions of the study.
2. Localization
The localization industry was born in the eighties as a result of the technology
boom and the arrival of desktop computers to the general public (Castells 2005; Es-
selink 2003). As technology became more and more present in the daily life of or-
dinary people, the demand for computer programs in local languages increased. A
new translation prole was required; mainly a professional translator with technical
skills who could embark on new and technologically advanced projects. As Esselink
and Cadieux (2002: 2) put it: “when people started translating software, some of the
changes required were not, strictly speaking, translation: changes to character encod-
ings, date and time formats, sorting rules, etc.” Due to the need of specialized teams to
translate or localize software, many high-tech companies outsourced their localization
projects to translation vendors in Europe, mostly in Ireland, which offered certain tax
benets to multinational high-tech companies based in Ireland (Esselink 2003). Mi-
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 239
crosoft and Oracle, among many others, had their European headquarters in Dublin
and, as their need for localization services grew, the number of localization vendors
established in Ireland grew as well (e.g. INK, later Lionbridge).
The term localization has been traditionally dened by LISA (Localization Indus-
try Standards Association)1 as:
Localization involves taking a product and making it linguistically and culturally appropriate
to the target locale (country/region and language) where it will be used and sold.
(LISA 1998: 3)
Although this denition does not specify what kind of products are typically lo-
calized, in a broad sense localization refers to adapting some kind of technology or
digital product to a given target culture from several perspectives. As GALA2 (Glo-
balization and Localization Association) explains:
True localization considers language, culture, customs, technical and other characteristics of
the target locale. While it frequently involves changes to the software writing system, it may
also change the keyboard usage, fonts, date, time and monetary formats. Graphics, colors and
sound effects also need to be culturally appropriate.
(GALA 2013)
Thus, when a new piece of software is designed to be launched in ten different
international markets, localizers step in to adapt it culturally and linguistically to
make sure that its users feel that it has been originally written and designed in their
own language. And yet, that goal is not so easy to achieve if the product has not been
properly internationalized, as it will be shown later on.
According to Jiménez-Crespo (2011: 4), there are ve types of localization: soft-
ware, websites, video games, small devices and multimedia content. However, there
are two primary and distinctive localization processes: software localization and web-
site localization or, as Mata (2005: 190) suggests, computer program localization
(Localización de productos informáticos or LPI) and web content localization (Local-
ización de contenido web or LCW). The former refers mainly to software products but
also to mobile apps, video games and software documentation and the latter applies to
products hosted in or created by technologies traditionally linked to the World Wide
Web.
Software localization has been dened by Esselink (2000: 57) as “the translation
of all graphical user interface (GUI) components of a software application, such as
dialog boxes, menus, and error or status messages displayed on screen”. Software lo-
calization involves working with or working around elements from the programming
environment or processes, like placeholders, control codes, and concatenated strings.
Furthermore, software localization implies taking into account aspects such as space
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
240
restriction, formats conversion, lack of context, bug xing and image localization
(Alonso and De la Cova 2013).
Space restriction is a particularly challenging aspect in software localization be-
cause text space in a software application is limited (even more in mobile applica-
tions). Therefore, the possibility of text expansion or restriction should be considered.
Languages do not take the same space or character length for a given text. For exam-
ple, Spanish uses 17% more characters than English for the same text (Oliver, Moré,
Climent 2008: 178).
In addition, image les and multimedia content are increasingly part of software
localization projects but their localization is seen by companies as quite an endeavor
more from the economic and organizational perspective than from a technical point
of view (Mata 2009).
In terms of website localization, although several authors have dened it (Corte
2002; Cronin 2003; Sandrini 2005; Pym 2011a), for the purpose of this study, the
following denition will be used:
A complex communicative, cognitive, textual and technological process by which interactive
digital texts are modied to be used in different linguistics and sociocultural contexts, guided
by the expectations of the target audience and the specications and degree requested by
initiators.
(Jiménez-Crespo 2013: 20)
This denition stresses the importance of the target audience and the specications
requested by the client or the product manufacturer, which is relevant to this study
because those specications can greatly impact the translation process and, to a cer-
tain extent, limit it.
Although software and website localization still entail different processes and re-
quirements, the difference between them are starting to fade out. Web applications
share some features of client-server traditional programs but are generated dynam-
ically using the web browser as a client. As Yu and Offutt (2002: 5) explain, web
applications generally work like this: “a client rst retrieves information from servers
in the form of HTML les, then sends requests to servers, and nally expects replies
as HTML les.”
It is therefore in this workow of content where localizers intervene and where – as
will be explained below - they face a number of challenges posed by the internation-
alization aspects of the text, as well as by translation tools themselves.
The situation of these challenges can be illustrated by the “dialect of resistance and
accommodation” (Pickering 1995 in Olohan 2011: 344). According to this “dialect”,
resistance is offered by an entity, for example, technology; and accommodation is
seen as the human response to that resistance. Olohan (2011: 342-355) applied this
sociological theory to translation studies and, through the analysis of translators’ posts
on a TM tool online forum, she identied the dialect of resistance and accommodation
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 241
(“dance of agency”) in the interaction between the TM tool and the translators. In the
wider context of localization processes, the internationalized text and technological
barriers could provoke resistance among translators; whereas accommodation would
take place when translators, forced by their skopos - together with tight deadlines and
client expectations as depicted by Alonso (2016) - work around these challenges and
end up aiming to produce an acceptable output.
3. Challenges in the localization of web applications
Although web applications can cause a variety of challenges for translators, due to
space limitations, this study does not cover all of them but instead focuses on interna-
tionalization related problems. In particular, it discusses to what extent internationali-
zation decient practices (both technical and language related) leave certain traces in
source texts that constrain translation activity. In addition, since localizers generally
work with translation memory systems, challenges related to how the source text is
segmented and presented to translators have also been taken into account.
Before moving to the empirical analysis of our corpus, we will review the existing
literature pertaining to our object of study, i.e. localization challenges related to inter-
nationalization factors and segmentation processes. More particularly we focus on: i)
challenges caused by software internationalization decient practices, ii) challenges
resulting from inadequate content internationalization, and iii) challenges due to the
rigidity of the segmentation of translation memories.
3.1. Software internationalization
The internationalization process takes place during product development in order
to enable localization:
Internationalization (I18n) is the process of generalizing a product so that it can handle multiple
languages and cultural conventions without the need for re-design. Internationalization takes
place at the level of program design and document development.
(LISA 1998: 3)
Software internationalization is critical for the correct localization of a product,
and it comprises two main activities: 1. Separating text from the development code (so
that it can be easily translated in an independent resource), and 2. Enabling the code to
support linguistic and cultural adaptation (Esselink 2000; McKethan and White 2005;
Schäler 2007; Massion 2011).
In terms of the rst activity, if the text and the source code are not appropriately
separated there can be such complex and costly localization problems that localiza-
tion might even be canceled. Therefore, programmers should separate user interface
text and source code according to the set guidelines of the development environment
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
242
(Herrmann and Sachse 2005: 47). Although the separation of text for translation is
the right practice to follow, it can trigger other sorts of challenges for localizers (see
sections User interface text separation and Ambiguity).
With regards to the second activity, when developing a piece of software, the
engineer should program the code in such a way that it will be able to display bi-di-
rectional text (left-to-right or right-to-left), e.g. to handle its localization in Hebrew.
Additionally, it should allow international characters, such as Japanese kanjis, if it is
going to be localized in that language, for which they need to make sure that charac-
ter encoding is right. Therefore, it is paramount to decide whether specic software
is going to be localized further down the line or not to avoid unnecessary problems
and costs after the software has already been developed (Herrmann and Sachse 2005;
Asnes 2007; Horvath 2007; Yang 2007).
In addition, there are certain international practices to be considered when devel-
oping software so that its localization can be carried out appropriately and without
causing problems for translators. Most leading technology manufacturers provide
detailed guidelines on how to internationalize their products correctly, such as Mi-
crosoft, Oracle or Google, as well as web technology contributors as the W3C, which
offers broad web internationalization guidance in its “W3C Internationalization (I18n)
Activity”3 website.
Horvath (2007: 3) suggests a list of ten internationalization tips to overcome most
common internationalization challenges. They range from using Unicode functions
and methods to preparing a translation kit that contains all necessary information for
translators. Among those ten tips, two are especially relevant for this study: user in-
terface text separation and avoiding ambiguity.
Although internationalization is a widely documented process and there are numer-
ous best practices guidelines available for publishers to use, localization problems can
persist if those are not implemented. As Dunne (2006: 6) points out:
The relative success of translation and localization efforts depends to a great extent on the
successful implementation of internationalization strategies. However, these in turn depend
on an enterprise level commitment to globalization strategies.
Moreover, the recent Multidimensional Quality Metrics (MQM) framework, which
denes quality metrics for translation quality assessment, includes internationalization
as one of the main dimensions of the ten identied:
Internationalization covers areas related to the preparation of the source content for subse-
quent translation or localization. Internationalization issues may be detected through problems
found in the target (particularly from those included in Locale convention (locale-convention),
but an Internationalization audit is generally conducted separately from a general assessment
of translation quality. (Lommel, Burchardt and Uszkoreit 2016)
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 243
In sum, these details reveal that although internationalization is widely implement-
ed for content globalization and best practices documentation are available, it is still a
cause of concern in the eld in terms of quality assessment, and also in terms of costs
inefciencies (LISA 2007).
3.1.1. User interface text separation
As mentioned earlier, separation of text from code is critical for the correct and
efcient localization of a product. In fact, the “Mozilla Localizability Guidelines”
mention as their rst recommendation “Don’t mix data and code” (Mozilla 1998) re-
ferring essentially to avoiding hard-coded strings and separating text in a resource le.
A resource script le is “a text le containing descriptions of resources from which the
resource compiler creates a binary resource le” (Esselink 2000: 473). Hard-coding
is a coding process that involves “the embedding of translatable strings in the body
of programming code rather than in separate resource les” (Esselink 2000: 469). To
put it another way, when an element such as a name or a message is hard-coded, the
translator does not have access to it in context (or at all); meaning that the linguistic
context where it appears (sentence or TM segment) does not contain the element itself
(which is embedded in the code). Several internationalization best practice guides
(Mozilla 1998; Vine 2002; Horvath 2007; Yang 2007) advice against hard-coding
because it can potentially cause translation quality issues, translation inconsistencies
or the publication of outdated or untranslated content.
Similarly, engineers resort to different ways of using elements that work like place-
holders but are not accessible to translators. Placeholders or variables are dened as
“characters that are usually preceded by a percentage (%) sign and replaced by anoth-
er word, value or string at application run-time” (Esselink 2000: 68). This is illustrated
in the following example drawn from our corpus:
Are you sure you want to reset <strong><span class=”reset-password-userna-
me”></span></strong>’s password?
On this example, the HTML element <span> would be replaced at runtime by the
username of the person whose password will be reset (i.e. Are you sure you want to re-
set Peter’s password?). Wrapping text with <span> tags provides a hook to an element
most likely via JavaScript4 (Roberts 2014). If the <span> element was a placeholder,
the translator would have access to it and would be able to move it where necessary
depending on their language structure. For example, in Spanish the order of the pos-
sessive noun phrase elements would be reversed:
¿Seguro que deseas restablecer la contraseña de <strong><span
class=”reset-password-username”></span></strong>”?
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
244
However, in this particular example, the <span> element would most likely not be
accessible in a TM environment and therefore the localizer would not be able to edit
it. The sentence would be parsed in two different source segments (SS):
TM SS 1: Are you sure you want to reset
TM SS 2: ‘s password?
Although in this case the translator could tweak the phrasing to adjust to the Span-
ish language structure, the solution is far from ideal because the TM target segments
(TS) would not be accurately stored (as will be described further in section 4.2.3.):
TS 1: ¿Seguro que deseas restablecer la contraseña de
TS 2: ?
3.1.2. Ambiguity
The second internationalization decient practice that Horvath (2007: 3) advises
against is ambiguity, which results mostly from not having context in GUI texts (ei-
ther linguistic or visual context).
Although separating text from source code is a fundamental internationalization
practice, it means that the GUI text will be stored in a resource le or database (basi-
cally in a correct resource format) without context. That resource le then is convert-
ed to a le format that translators can use in their translation memory systems (e.g.
XLIFF le).
The source les are usually programming code les-e.g. instructions in the programming lan-
guage C++- and resource les (with the extension .rc), which include user interface elements
such as dialog boxes, which constitute the main elements to be translated. (Sandrini 2008:
170)
Therefore, if the software strings do not contain contextual information, they might
be ambiguous and difcult to understand (Massion 2011; Horvath 2007). Unfortu-
nately, this is too often the case (Safar and Machala 2010). Figure 1 is an example of
a typical resource le where the text of the software is stored:
se n d e b a r
issn-
e 2340-2415 | Nº 27 | 2016 | pp. 235-266 245
Figure 1. Resource fi le of WinHTTrack
In the previous gure, the savvy localizer would be able to identify certain ele-
ments (buttons, menus, etc.) that would add some kind of context to their translation.
Nevertheless, quite frequently, the le sent to translation would look very much like
this, i.e. a list of strings without any contextualization or information of the nature of
the user interface component:
Figure 2. GUI string list of WinHTTrack
Lack of context is one of the main challenges that translators face when localizing
a piece of software (Esselink 2000; Ludwigsen, Williams and Polis 2009; Safar and
Machala 2010; Sikes 2010; Jiménez-Crespo 2013; Dunne 2015; Edwards 2015). Am-
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
246
biguity can cause confusion, translation errors and, ultimately, it can affect the quality
of the localized product negatively (Herrmann and Sachse 2005). Words in isolation
can cause confusion, have different meanings depending on the context (Bernal Me-
rino 2007: 31-32) and they can also create problems to Machine Translation systems
when options are not appropriately tagged (Roturier and Lehmann 2009).
To avoid working with de-contextualized strings that show in the resource le
without any textual order, developers should include a description eld where they
can enter comments in order to provide context and additional information for trans-
lators (Ludwigsen, Williams and Polis 2009).
As Edwards points out (2015: 20):
(…) context is so critically important to our work because so much depends on conveying the
right information across linguistic and cultural barriers. Without all the proper contextual cues
and metadata, our jobs become extremely challenging — if not even impossible in some cases.
This process of localizing text without context or descriptions could be termed
“blind localization” because translators work very often without knowing what they
are actually translating.
3.2. Content internationalization
As described in the previous sections, the term internationalization mainly applies
to the technical process of designing a software or digital product in such a way that it
enables localization; alternatively, the term can also apply to writing content with an
international audience in mind (Alonso and De la Cova 2013). This concept is broadly
referred to as “writing for translation” (also “writing for an international audience”
or “writing world ready content”), but to emphasize the importance of considering
international aspects when writing content, in this paper the phrase “content inter-
nationalization” is the one primarily used. As Brown (2003: 4) states: “Writing for
translation requires the technical writer to express complex concepts in terms clear
enough to move easily from one language to another”.
Basically, writing for translation involves thinking ahead and writing text with an
international audience in mind when a product is going to launch globally. This is cen-
tral in avoiding problems at the localization/translation stage, like unnecessary costs
and quality issues (Hoft 1995; Amant 2005; Sichel 2009; Yang 2007). In addition,
Keller (2011) points to resulting time inefciencies when working with TM systems
as a further problem.
In particular, unnecessary costs can be prevented to a large extent if writers take
into account that their output will be sent for translation but, unfortunately, that does
not happen as often as it should (Brown 2003; Sichel 2009), as will be shown in sec-
tion 4.2.2. Localization is a more expensive task than writing; for example, manufac-
turers may want to take into consideration that “an average of $4 is spent on localiza-
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 247
tion for every $1 spent on technical documentation” (Brown 2003: 4). Therefore, for
a business that values their ROI (return on investment), this data should make them
rethink their writing process when localization enablement is needed.
Because every culture has different ways of conveying basic information, it seems
sensible that technical writers are aware of them in order to avoid problems and to
help their companies prosper in a world where online communication is essential
(Amant 2001). However, as it is not realistic to expect that writers would learn in
detail about global cultural differences, there is a considerable amount of guidelines
and manuals to help them with that task. This is particularly relevant for technology
companies that have a substantial international presence and whose revenues largely
come from international markets. For instance, Microsoft, Google or Symantec have
large teams of writers (also called technical writers, editors or communicators). As
Amant (2000: 326) points out:
English’s status as an international language does not mean that English-speaking techni-
cal communicators can now rest on their linguistic laurels. Rather, technical communicators
should be aware of important technical publications printed in languages other than English
in order to get a more comprehensive picture of the developments in a certain eld.
Text written for translation should avoid culture-specic and location-specic ref-
erences (Kamerer 2003), as they can be difcult to convey in other languages. Visual
representations can also be problematic if they are transferred to international mar-
kets, although they are a very effective means of communication if used in the right
way (Amant 2005).
Apart from the fact that the text should be written with translation in mind, the
contextual information helps translator learn about the subject matter and bridge the
gap between cultural differences (Flint et al. 1999: 239).
As established above, cultural references are one of the key elements to consider
when writing content for an international audience. Hoft (1995) refers to these ele-
ments as international variables: “International variables are the localizable elements.
International variables identify supercial and deep cultural differences” (Hoft 1995:
19). Usually, these variables are related to politics, economy, religion, education, lin-
guistics and technology.
In order to help writers create “localization friendly” content, companies such as
Microsoft and Mozilla create their own internationalization guidelines. Similarly,
there is a signicant amount of scientic production related to writing for translation
primarily in the Technical Communications eld. Nonetheless, there is only a limited
amount of scientic production arising from Translation Studies about the importance
of designing technical content for a global audience. De la Cova (2014) suggests a
categorization of guidelines for international writers, specically designed for web
content localization. These categories include cultural references (e.g. humor), time
and localization references (e.g. date formats), clear writing (e.g. avoiding ambiguity),
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
248
world-ready wording (e.g. rethinking slang), acronyms (e.g. DIY), colors and images
(e.g. cultural-rich images) and punctuation (e.g. number formats).
3.3. Rigidity of translation memories
The third translation challenge highlighted in this paper derives from the rigidity
of translation memories (TMs).
Translation memory is a technology that enables the user to store translated phrases or
sentences in a special database for local re-use or shared use over a network. Translation
memory systems work by matching terms and sentences in the database with those in the
source text. If a match is found, the system proposes the ready-made translation in the target
language.
(Esselink 2000: 362)
Born in the seventies and widely implemented in the nineties (Somers, 2003),
TMs have evolved considerably in the last decades, becoming an essential tool for
professional translators and other language services providers. TMs have changed and
shaped the way translators work (Bowker 2002) and have brought benets in terms of
cost savings and productivity gains (Lagoudaki 2006; O’Hagan 2009). Consequently,
there is a large volume of studies related to the impact of TMs on the translation pro-
cess (Bowker 2005; Mossop 2006; Biau Gil and Pym 2006; Jiménez-Crespo 2009;
Pym 2011b; Olohan 2011; Christensen and Schjoldager 2011; LeBlanc 2013; Alon-
so 2015; Alonso and Calvo 2015; Bundgaard, Christensen and Schjoldager 2016).
Research topics range from functional aspects about TMs, to the way they alter the
translation process or to their inuence on cognitive processes.
LeBlanc (2013) has conducted a study with Canadian translation vendors to evalu-
ate their perception of translation technology, in particular regarding TMs. The results
show the main advantages and disadvantages of using this technology from their per-
spective. On the one hand, among the advantages, LeBlanc highlights that TMs help
increase productivity; they improve consistency; they eliminate repetitive work; they
are used as databases; and they have a pedagogical function. On the other hand, among
the disadvantages, translators have argued that TMs change their relationship with the
text (due to segmentation); they are a barrier to creativity; they make translators pas-
sive; they have an effect on their natural reexes; they make beginner translators too
dependent on them; they are sometimes polluted and contribute to error propagation;
they have a demanding impact on their productivity requirements; they are used as
rigid guidelines for using exact or full matches; and nally, they can potentially have
a negative effect on translators satisfaction if misused (LeBlanc 2013: 6-10).
One of the main challenges detected in the present study comes from the process
of segmentation. Segmentation is one of the ve main processes of TM systems: text
segmentation, alignment, indexing, search and retrieval (Lagoudaki, 2006: 3). It is de-
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 249
ned as the “division of text into translatable units, such as sentences or paragraphs”
(Esselink 2000: 473) or “other more or less easily distinguishable text portions, such
as titles” (Somers 2003: 34). When translators enter their translations in target seg-
ments, the source and the target segments form the so-called “translation unit”.
Different authors have focused on segmentation and how it affects the translation
process (Dragsted 2006; Colominas 2008). For instance, Colominas (2008: 352) af-
rms that adding sub-sentential segmentation (chunk segmentation) to TM technolo-
gies might enhance their recall (target-text proposal) and precision (usability).
Challenges related to segmentation were also mentioned by translators in LeB-
lanc’s study, generally with regard to the inexibility of the sentence-by-sentence
approach, which they describe as too mechanical and unnatural and possibly leading
to problems in the cohesion of the text (LeBlanc 2013: 9). According to Bass (2006:
76), segmentation is one of the technological barriers identied in his study as obsta-
cles to quality in the localization industry. Another relevant study (Jiménez-Crespo
2010: 203) comparing original and TM translated texts from the same genre shows
that translated text replicates source superstructure, which would imply that TMs do
restrain and constrain translators.
4. Empirical study
4.1. Methodology
This is a qualitative and descriptive study based on a single monolingual English
ad hoc corpus composed of the GUI strings of three web applications from the cloud
storage and social media eld. The present study is a preliminary work related to a
wider PhD research project (De la Cova, in progress) in which localization problems
are being identied in a single monolingual English corpus and categorized following
a proposed set of criteria.
According to EAGLES (1996), a corpus is “a collection of pieces of language that
are selected and ordered according to explicit linguistic criteria in order to be used as
a sample of the language”. The corpus analyzed in this study is a single monolingual
English ad-hoc corpus of 101,263 source words (18,911 strings), composed of the
GUI strings of three web applications. For illustrative purposes, we have provided
translations into Spanish when applicable.
These web applications have been chosen according to representativeness criteria
in the application category they fall under, as they are some of the most used cloud
storage and social media services of the last years.
In terms of corpus size, it satises the research purpose of this preliminary study.
Although there does not seem to be an agreed optimal corpus size among researchers
(McEnery and Wilson 2001, Sinclair 2002, Corpas and Seghiri 2006, Reppen 2010),
some authors, such as Kennedy (1998 in Olohan 2004: 46), agree that a big corpus
is not necessarily better than a smaller one and that the aim of the research should
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
250
shape the corpus design criteria. Similarly, Bowker and Pearson (2002: 45) state that
“there are no hard and fast rules that can be followed to determine the ideal size of a
corpus”, and the researcher should determine what is the appropriate size depending
on the purpose of the project and the availability of data. These authors suggest 25,000
words as an initial goal for a corpus size, which could be expanded.
Given the nature of the texts and the purpose of this analysis, a manual analysis
seemed the most appropriate approach. As argued by Sinclair (2000 in Olohan 2004:
62-63), in qualitative studies corpus tools might be less useful than in automated
analyses, where annotated corpus seem more suitable. Moreover, since the aim of the
study was to observe the source text as translators do when they undertake a locali-
zation project in order to identify potential challenges, we concluded that using a TM
tool would be a feasible approach. As a result, our corpus was processed in a TM tool,
and segments were analyzed in the TM environment.
For the present study, the corpus has been processed in Smartling5 Translation
Management System because it is a tool specically designed for websites and dy-
namic web applications. Smartling TM follows regular industry segmentation prac-
tices and includes standard TM features. One of the most innovative aspects of the
Smartling TM system is that it is cloud-based, which appears to be the industry ten-
dency for translation software (DePalma 2015; Toon 2015).
In terms of the object of research, as outlined in the previous section, for the pur-
pose of this study, only translation problems related to or stemming from internation-
alization factors have been selected in our corpus of web apps. Additionally, since the
source text is generally presented to translators in a TM system, certain challenges
caused by it have also been taken into account. With the aim of illustrating these types
of localization challenges, the following sections show the results from our corpus
analysis.
4.2. Description of challenges
4.2.1. Challenges caused by software internationalization decient practices
In terms of software internationalization practices, two principal challenges have
been identied in the corpus. Firstly, challenges subsequent to not separating appro-
priately text and code. And secondly, challenges related to ambiguity.
4.2.1.1. User interface text separation
The extraction of text from code into a separate le (resource le) is an essential
internationalization practice to avoid localization problems. When an application is
not appropriately internationalized, it can happen that not all text is separated in a re-
source le. As a result, translators might not have access in the source text to certain
localizable elements that are necessary for linguistic or stylistic correctness. This can
be seen in the following example.
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 251
Plan changes will be applied to your next billing date on <span
class=”billing-date”></span>
The string above is segmented by the TM system as shown below.
TM SS: Plan changes will be applied to your next billing date on.
As explained in section 3.1, the red <span> element behaves like a placeholder but
it is not presented in the source text; therefore, translators might have difculties in
nding out what it is.
In this case, translators might assume that the proposition “on” is followed by a
date with the format of month and day (and optionally, year), such as “July 1st, 2016”,
but it could be followed instead by the segment “the last day of the month” or “Friday,
July 1st”, to mention a few options. Without further context or access to isolated ele-
ments, translators might want to play it safe and avoid making any kind of grammar
inconsistency and therefore, translate the strings as below:
TS: Los cambios de plan se aplicarán en la próxima fecha de facturación:
4.2.1.2. Ambiguous strings
Context is fundamental for proper translation as explained above. Unfortunately,
the internationalization process, though necessary, often causes the de-contextualiza-
tion of user interface strings.
The two following segments are illustrative examples of de-contextualized strings
that can pose quite a challenge to translators:
TM SS: Logged in
TM SS: Removed the sign-in URL
Without additional context, one would have serious difculties translating these
strings into Spanish (and probably into other languages) because there is not an ex-
plicit subject in any of them. In the rst example, it is unknown whether one person
has logged into a corporate email account or whether two people have logged into it.
Similarly, in the second example, it is not clear whether “a person has removed the
sign-in URL” as opposed to “a glitch has removed the sign-in URL”. Therefore, with-
out further information, a wide variety of possible scenarios could apply depending on
the web application features. As before, the safest and most neutral translation would
probably be to provide an impersonal translation in Spanish:
arT Í CU L Os O ri Gin aLe s
de la Cova, E. Translation Challenges in the Localization of Web Applications
252
TS: Sesión iniciada or Se inició sesión
TS: URL de acceso eliminada or Se eliminó URL de acceso
In this case, it was later found out that there was an implicit singular subject on the
strings and therefore they could have been translated as “Ha iniciado session” and
Ha eliminado la URL de acceso” respectively. This challenge would have been easier
to overcome had there been a comment or a description with additional information
on the user interface strings, as explained in section 3.1.
Another ambiguous example found in the corpus is the following:
TM SS: Like “name”
As user interface strings are usually extracted without context in a separated le,
possibly the rst reaction of localizers of web applications would be to translate this
string as the popular phrase of social media “Like” (in Spanish “Me gusta”), assuming
that the web application they are translating has some kind of “Like” button. In social
media sites or applications, the button or option “Like” is generally used to click or
tap on a post (picture, comment, etc.) that a user likes on a given page. Facebook is
the company that made this phrase popular and it has now spread to other social media
websites like Pinterest and Instagram, although they use different icons (a thumb-up,
a heart, etc.) to illustrate the “like” action:
Figure 3. Facebook Like button in English and Spanish
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 253
Alternatively, the string might simply refer to a prepositional phrase (preposition
+ noun), as in “Like Maria”. In this case, unless translators are provided with further
details, they might follow their intuition and risk translating the string as “Me gusta”,
potentially entering a translation mistake in the target text and consequently polluting
the TM (which was one of the drawbacks of using TMs according to translators in
LeBlanc’s study). The string actually referred to “Like” as preposition (e.g. “Como
María”) and not to the social media phrase:
TS: Como “name”
In addition, the phrase “Like” introduces a wide variety of linguistic questions
and concerns when translated into Spanish (and other languages). For instance, when
people “like” a post on a given page, the text to describe that action shows the num-
ber of people that have liked that post, e.g. “5 Likes”. This phrase would probably
be translated into Spanish using “Me gusta” because that is the most widespread
translation, but as the source phrase is pluralized (“Likes”), translators might have
doubts on how to write it: 5 Me gusta, 5 me gusta, 5 “Me gusta”, etc. On top of that,
the string “likes” in isolation might also make translators wonder whether it is a verb
(third person singular) as in the example above or a pluralized noun. And that would
open a new set of challenges.
4.2.2. Challenges resulting from inadequate content internationalization
As mentioned in section 3.2, “writing for translation” (or content international-
ization) refers to writing content with an international audience in mind, knowing
that the written text is going to be sent for translation and that certain neutralization
guidelines need to be followed. Apart from cultural and location-specic references,
language structures that can constrain translation are also discouraged. In order to help
developers write localizable content, certain technology companies, such as Mozilla
or Microsoft, publish detailed internationalization and localization guidelines which
stress the importance of avoiding assumptions with regards to language structure and
singularities (Mozilla 1998; Microsoft 2016). Among the different guidelines, they
remind developers that English word order and lack of gender might not replicate well
(or at all) in other languages.
To examine this challenge, two examples from the corpus will be analyzed.
TM SS: You found a %s!
The placeholder “%s” could potentially be replaced by any singular noun which
makes sense in terms of the application features (le, folder, error, match, etc.). Since
there are two genders in Spanish (masculine and feminine), “a” could theoretically be
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
254
translated by “un” or “una” when there is no context or string description, as shown
in this string:
TS 1: ¡Has encontrado un/una %s!
Or it could also be translated in other ways to avoid using “un/a”.
TS 2: %s: elemento encontrado
TS 3: ¡Se encontró %s!
TS 4: ¡Has encontrado: %s!
At application run-time, the placeholder would be replaced by an element (most
likely depending on the user action), so the rst segment would show in the user in-
terface like “¡Has encontrado un/una coincidencia!” which certainly does not sound
natural or correct in Spanish. The other segments would be displayed like “coinciden-
cia: elemento encontrado”, “Se encontró coincidencia” or “¡Has encontrado: coinci-
dencia!”, options clearly showing the user that this is an automatically generated text.
Neither of them is ideal because the target user would probably feel the text as not
written in their language. This problem could easily be solved if developers entered
comments in the description eld or if they used a different phrase or text to inform
users that some element has been found in their system.
In the second case taken from the corpus, the translator will come across a similar
challenge because writers have not taken into account the different constructions of
languages:
TM SS 1: for the last
For this string, fortunately, there is some preview context. It is part of an applica-
tion feature by which the user can generate reports on different measurable actions
(saved items, posts published, items purchased, etc.) for a given period of time (e.g.
week, 6 months, year, 2 years). The preview of the report feature would look some-
thing like this:
Show “saved items” for the last “week”
The underlined elements are four different translatable TM segments, and the quot-
ed phrases are items on drop-down menus (that the user can choose from depending
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 255
on what kind of report they want to get). As a result, they are segmented in the TM
system like this:
TM SS 1: Show
TM SS 2: saved items (or posts published, purchases)
TM SS 3: for the last
TM SS 4: week (or 6 months, year, 2 years)
The noun on TM SS 4 constrains the translation of “for the last” both by gender
and by number in Spanish because the drop-down menu includes both masculine and
feminine nouns (and plural masculine nouns) as options.
Considering that when an application is already developed and sent for transla-
tion, it is not likely that developers will want to make changes unless an important x
needs to be made, translators will have to work with the text sent for translation, and
to creatively think for feasible solutions.
In the present example, a possible but not ideal solution would be to translate the
string “for the last” as “para anterior/es”. “Anterior” is a neutral adjective in Spanish
working for masculine and feminine, and a good option to avoid too many alternative
options (para el/la último/a) that would interfere with the user interaction with the
application. It is advisable to use a precise and concise language when translating
software to make sure that the use of the software is not hindered by language (Mi-
crosoft 2016).
4.2.3. Challenges related to the rigidity of segmentation in TMs
As seen on section 3.3, the process of segmentation can heavily constrain the lo-
calization process. As Austermühl (2006: 77) states:
Among the complaints from the translator community are the rigidity of source
text structures, the dominance of the sentence or sub-sentence phrases as primary
translation units, incompatibilities within one TM or between TM and term bases
contents, faulty yet untouchable segments, the lack of creativity for the translator
as autonomous text producer, the lack of co-text and context for the segments to be
translated, and the lack of motivation or freedom to go beyond the simplistic source
text structures and the preexisting translations imposed upon the translator by the TM
system.
There are several challenges related to segmentation, but for the purpose of this
study only two will be highlighted. Firstly, translators work with isolated units (seg-
ments) that potentially affect textual cohesion, especially because many TMs do not
have a feature for segments merging which would be useful in many cases (as shown
below) and because TMs generally do not allow for text structure changes. Secondly,
and due to the singularities of every language, sometimes it seems unavoidable to
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
256
store segments incorrectly. Since sentences are split into segments, if source and target
languages have different sentence structures, translators might have to play around
with TM segments to produce a grammatically and stylistic correct sentence in the
target language.
This two-fold segmentation problem can be exemplied with the following string
found in the analyzed corpus:
Internal and external collaboration
This sentence is split into two segments in the TM system:
TM SS 1: Internal and external
TM SS 2: collaboration
Since merging segments is not allowed in the TM system used, it is almost impos-
sible to store target segments correctly in the TM because Spanish usually requires
that the noun precedes its adjective (its modier) in a noun phrase. A potential trans-
lation would be:
TS 1: Colaboración interna
TS 2: y externa
As a result, the TM would save one translation unit as:
EN: Internal and external – ES: Colaboración interna
And the second translation unit as:
EN: collaboration – ES: y externa
Consequently, the TM would be polluted. There are other workarounds for this
segment, but none of them would avoid this problem completely. Although this solu-
tion seems acceptable from a translation point of view, the pollution of the TM could
affect productivity in future projects and hinder translation quality if they are not used
cautiously (Bass 2006: 76).
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 257
5. Conclusions
Localization is everywhere. As a ubiquitous phenomenon, it is in the websites we
read, in the devices we use and in the apps that make our life easier. However, trans-
lated content might not always sound natural in a given content, or might not even be
correct, as exemplied earlier with the Zara translation mistake. In fact, some soft-
ware manufacturers offer their users the possibility to report language problems they
might come across (e.g. Google, Dropbox6, Pinterest).
As this paper has tried to show, the localization of web applications can pose sever-
al challenges related to internationalization inadequate practices and to technological
barriers (e.g. TM segmentation).
In terms of internationalization factors, two main sets of challenges have been
identied: one related to the internationalization of the software itself (technical pro-
cess), and another related to “content internationalization”. The rst set of challenges
generally arises from the separation of text and code that, although essential for the lo-
calization process, can cause ambiguity problems (leading to “blind localization”) and
the mixing of code and text in the resource le if the application has not been correct-
ly internationalized. The second set of challenges pointed out is inadequate content
internationalization (or writing for translation), which can constrain the localization
practice when writers do not consider cultural or language differences, e.g. structure.
As regards to technological barriers, TM system alteration of the text through
segmentation can seriously limit translators and cause linguistic and stylistic errors
as well as impact text cohesion. As mentioned earlier in this paper, localization is
characterized by the intersection between translation and technology (Jiménez-Crespo
2010) so, given the outlook of this study, segmentation was taken into consideration
as one of the technological challenges that localizers face.
Programmers, technical writers and localization engineers can help overcome these
problems. If isolated UI strings (due to the separation of text and code) included com-
ments or descriptions, the task of translators would be easier. Moreover, if writers
wrote content with an international audience in mind and following internationaliza-
tion guidelines, translators would not feel constrained by the source text. There are
multiple guidelines and best practices available for content creators. However, the
missing piece for implementing those in the content creation process might be found
in businesses making certain decisions based on ROI considerations since localization
is a process constrained by economic decisions. Similarly, if more resources were in-
vested in making sure that TM systems help translators produce a correct and natural
output instead of making them resort to workarounds, then text cohesion and time
efciencies would surely improve.
Both the fact that translators are only one of the actors involved in the transla-
tion production networks (Abdallah and Koskinen 2007) and that internationalization
strategies often follow beyond the scope of the translation process serve as arguments
for adopting a wider approach to the topics discussed in this paper. One could say that
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
258
translators at times lack the necessary freedom to create correct and natural target
texts. As Harcz (2016: 23) argues:
The extent to which liberty can be taken when rendering a script from one language to another
is determined by several factors. Most notably: the targeted audience; the purpose (or intended
use) of the translation; the subject matter of the document; the client who ordered the job; the
specic instructions received when the order was placed; and my personal preferences and
habits as a translator.
Furthermore, according to Alonso and Calvo (2015), in the future, tools will har-
moniously complement translators’ own capacities, as in the scenario depicted by
these authors in their trans-human translation approach. Translators will interact with
technology “as though this were really just an extension of their capabilities and
creating a process with a social, creative and learning dimension” (Ibid.: 151), thus
overcoming current tendencies towards the dehumanization of translation processes
(Biau-Gil and Pym 2006) (cited in Alonso and Calvo 2015: 148).
Going back to Pickering’s dialect of “resistance and accommodation” (in Olohan
2011: 344), one could argue that, although professional translators might be able to
“accommodate” and work around challenges (“resistance”) producing an acceptable
output, these solutions might not be the most desirable options to successfully localize
a product. This is particularly the case if one considers that localization entails “taking
a product and making it linguistically and culturally appropriate to the target locale
(country/region and language) where it will be used and sold” (LISA 1998: 3). There-
fore, one might wonder if the challenges identied in this preliminary study prevent
true localization from happening.
As a nal remark, if localization quality evaluations considered the nature of source
texts and the environments that translators work with when assessing quality, metrics
might shed some light into what is causing certain translation errors. In the recent
“Multidimensional Quality Metrics” project (QTLaunchPad 2014), it is advised that
translation quality scorecards consider internationalization aspects:
If Internationalization problems occur, they will negatively impact the quality score for the
translated content, even though they are beyond the control of the translator. Any metrics
used for translator assessment MUST exclude issue types that are beyond the control of the
translator.
However, to consider internalization in localization quality evaluations, manufac-
turers and vendors should thoroughly audit source content and, unfortunately, this
does not seem to be a standardized practice so far.
This paper has tried to ag the need to spend more resources in the internation-
alization of web applications, or to increase the “enterprise level of commitment to
globalization strategies” (Dunne 2006: 6), considering that even some of the biggest
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 259
technology publishers have shortcomings on the internationalization process. The
fact that few companies measure the ROI of localization and that they lack a compre-
hensive localization strategy (DePalma 2006) could explain why these shortcomings
still persist. As Giammarresi argues: “Just as localization project pain is evidence of
internationalization issues, internationalization issues are themselves symptoms of
deeper, more systemic ills” (2011: 19).
Although localization professionals have been stressing for years the importance
of thinking about localization during product development and of providing adequate
tools for localization processes, according to the present study, translation problems
still arise from decient globalization processes. Hopefully, the day will come when
most technology companies truly view localization as a strategic business decision
that affects their revenue and their branding. When that happens, the challenges iden-
tied in this paper could be overcome.
6. Bibliography
Abdallah, Kristiina and Koskinen, Kaisa (2007). Managing trust: Translating and
the network economy. Meta: Journal des Traducteurs / Meta: Translators’ Journal
52 (4), 673-687.
Alonso, Elisa (2015). Analysing Translation Professionals in the Information So-
ciety and their Use and Perceptions of Wikipedia [online]. JoSTrans: The Journal
of Specialised Translation 23 (January 2015), 89-116. <http://www.jostrans.org/
issue23/art_alonso.php> [Accessed 25 March 2016].
Alonso, Elisa (2016). Conict, opacity and mistrust in the digital management of
professional translation projects. The International Journal of Translation and In-
terpreting Research 8 (1), 19-29.
Alonso, Elisa and Calvo, Elisa (2015). Developing a Blueprint for a Technolo-
gy-mediated Approach to Translation Studies. Meta: Journal des traducteurs Meta:/
Translators’ Journal 60 (1), 135-157.
Alonso, Elisa and De la Cova, Elena (2013). Introducción a la localización. Materi-
ales docentes (An Introduction to Localisation. Training Materials). Seville (Spain):
[self-edited].
Amant, Kirk S. (2000). Expanding translation use to improve the quality of tech-
nical communication. IEEE transactions on professional communication 43(3),
323-326.
(2001). Cultures, computers, and communication: Evaluating models of inter-
national online production. IEEE Transactions on Professional Communication 44
(4), 291-295.
— (2003). Designing effective writing-for-translation intranet sites. IEEE Transac-
tions on Professional Communication, 46 (1), 55-62.
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
260
— (2005). A prototype theory approach to Web site localization: an analytical meth-
od for technical communicators. 2005 IEEE International Professional Communi-
cation Conference Proceedings, 101-107.
Austermühl, Frank (2001). Electronic tools for translators. Manchester: St. Jerome.
— (2006). Training translators to localize. In Translation Technology and its Teach-
ing. Anthony Pym, Alexander Perekstenko and Bram Starink (eds.), 69-81. Tarrag-
ona: Intercultural Studies Group.
Asnes, Adam (2007). Change your encoding, change your company. Getting started
guide: Internationalization. Multilingual 18 (3), 5-6.
Bass, Scott (2006). Quality in the real world. In Perspectives on Localization. Ki-
eran J. Dunne (ed.), 69-84. Amsterdam-Philadelphia: John Benjamins.
Bernal Merino, Miguel Ángel (2007). What’s in a ‘Game’?. Localisation Focus The
International Journal of Localisation 6 (1), 29-38.
Biau-Gil, José Ramón and Pym, Anthoy (2006). Technology and translation (a ped-
agogical overview). In Translation Technology and its Teaching. Anthony Pym,
Alexander Perekstenko, and Bram Starink (eds.), 5-19. Tarragona: Intercultural
Studies Group.
Bowker, Linda and Pearson, Jennifer (2002). Working with specialized language: a
practical guide to using corpora. London-New York: Routledge.
Bowker, Linda (2005). Productivity vs quality. A pilot study on the impact of trans-
lation memory systems. Localisation Focus 4, 13-20.
Brown, M. Katherine (2003). Trends in Writing for Translation. Multilingual 14
(7), 4-8.
Bundgaard, Kristine, Christensen, Tina Paulsen and Schjoldager, Anne (2016).
Translator-Computer Interaction in Action—an Observational Process Study of
Computer-Aided Translation. Journal of Specialised Translation 25, 106-130
Castells, Manuel (2005). Internet y la sociedad red. In Por otra comunicación: los
media, globalización cultural y poder. Denis de Moraes (ed). 203-228. Barcelona:
Intermón Oxfam Editorial.
Christensen, Tina Paulsen and Schjoldager, Anne (2011). The Impact of Transla-
tion-Memory (TM) Technology. In Human-Machine Interaction in Translation:
Proceedings of the 8th International NLPCS Workshop. Bernadette Sharp, Mi-
chael Zock, Michael Carl and Arnt Lykke Jakobsen (eds). 119-130. Copenhagen:
Samfundslitteratur.
Colominas, Carme (2008). Towards chunk-based translation memories. Babel 54
(4), 343-354.
Corpas Pastor, Gloria and Seghiri Domínguez, Miriam (2006). El concepto de rep-
resentatividad en la Lingüística del Corpus: aproximaciones teóricas y metodológi-
cas. Documento técnico BFF2003-04616 MCYT/TI-DT-2006-1. Universidad de
Málaga.
Corte, Noelia (2002). Localización e internacionalización de sitios web. Tradumàti-
ca 1, 1-8.
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 261
Cronin, M. (2003). Translation and globalization. London-New York: Routledge.
De la Cova, Elena (2014). Localización de sitios web: ¿a quién van dirigidos? IV
Congreso de la Sociedad Española de Lenguas Modernas. Sevilla: Sociedad Es-
pañola de Lenguas Modernas
— (in progress). Localization problems of Online Help (temporary title). Doctoral
Thesis, Universidad de Sevilla
DePalma, Donald (2006). Quantifying the return on localization investment. In Per-
spectives on Localization. Kieran J. Dunne (ed.), 15-36. Amsterdam-Philadelphia:
John Benjamins.
— (2015). The cloud: a translation business essential. Multilingual 26 (1), 26-27
Diéguez Morales, Mª Isabel y Lazo Rodríguez, Rosa Mª (2011). El español en In-
ternet: aciertos y errores en sitios web localizados del inglés. Onomázein: Revista
de lingüística, lología y traducción de la Ponticia Universidad Católica de Chile
24, 299-326.
Dragsted, Barbara (2006). Computer-aided translation as a distributed cognitive
task. Pragmatics & Cognition 14 (2), 443-464.
Dunne, Keiran (2015). Localization. In Routledge Encyclopedia of Translation
Technology. Sin-wai Chan (ed.), 550-562. London-New York: Routledge.
(2006). A Copernican revolution. In Perspectives on Localization. Kieran J.
Dunne (ed.), 1-11. Amsterdam-Philadelphia: John Benjamins.
• Edwards, Kate (2015). The death of context. Multilingual 26 (8), 20-21.
Esselink, Bert (2000). A practical guide to localization. Amsterdam-Philadelphia:
John Benjamins.
— and Cadieux, Pierre (2002). GILT: Globalization, Internationalization, Localiza-
tion, Translation. The LISA Newsletter: Globalization Insider XI (1.5).
— (2003). The evolution of localization. The Guide to Localization. Multilingual
14 (5), 4-7.
Fenstermacher, Hans (2006). Authors, localizers and language barriers. Multilin-
gual 17 (1), 82
Flint, Patricia, Van Slyke, Melanie Lord, Stärke-Meyerring, Doreen and Thompson,
Aimee (1999). Going online: Helping technical communicators help translators.
Technical communication 46 (2), 238-248.
Giammarresi, Salvatore (2001). Strategic views on localization project manage-
ment. In Translation and localization project management: the art of the possible.
Kieran J. Dunne, and Elena S. Dunne (eds.), 17-51. Amsterdam-Philadelphia: John
Benjamins.
• Harcz, Daniel B. (2016). Taking liberty as a translator. Multilingual 27 (1), 23.
Herrmann, Achim and Sachse, Florian (2005). Internacionalización de aplicaciones
de software. In Traducción y localización. Mercado, gestión y tecnologías. Detlef
Reineke (ed.), 45-70. La Palmas de Gran Canaria: Anroart.
Hoft, Nancy L. (1995). International technical communication: How to export in-
formation about high technology. New York: John Wiley & Sons, Inc.
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
262
Horvath, Alan (2007). Think internationalization in everyday design. Getting start-
ed guide: Internationalization. Multilingual 18 (3), 3-4.
Hurtado Albir, Amparo (2001). Traducción y traductología. Madrid: Cátedra.
Jiménez Crespo, Miguel Ángel (2008). El proceso de localización web: estudio
contrastivo de un corpus comparable del género sitio web corporativo. Tesis doc-
toral, Universidad de Granada
(2009a). The effect of Translation Memory tools in translated Web texts: Ev-
idence from a comparative product-based study. Linguistica Antverpiensia 8,
213–234.
— (2009b). El uso de corpus textuales en localización. Tradumàtica: traducció i
tecnologies de la informació i la comunicación 7, 1-15
— (2010). The intersection of localization and translation A corpus study of Span-
ish original and localized web forms. Translation and Interpreting Studies 5(2),
186-207.
— (2011a). To adapt or not to adapt in web localization: a contrastive genre-based
study of original and localized legal sections in corporate websites [online]. Jos-
trans: The Journal of Specialised Translation 15, 2-27. <http://www.jostrans.org/
issue15/art_jimenez.php> [Accessed 15 March 2016].
— (2011b). A corpus-based error typology: towards a more objective approach to
measuring quality in localization. Perspectives, Studies in Translatology, 19 (4):
315-338.
— (2012). Loss or Lost in Localization: A Corpus-Based Study of Original and Lo-
calized Non-Prot Websites [online]. Jostrans: The Journal of Specialised Trans-
lation 17, 136-165. < http://www.jostrans.org/issue17/art_jimenez.php> [Accessed
16 March 2016].
• — (2013). Translation and web localization. London-New York: Routledge.
• Kamerer, Laurie (2003). Avoiding a US-centric Writing Style. Multilingual 14 (7),
22-23
Keller, Nicole (2011). Creating translation-oriented source documents, Multilingual
22(7), 43.
Kiraly, Don C. (1995). Pathways to translation: Pedagogy and process (No. 3).
Ohio: Kent State University Press.
Lagoudaki, Elina (2006). Translation memories survey 2006: Users’ perceptions
around tm use. Proceedings of the ASLIB International Conference Translating &
the Computer (28) 1, 1-29.
LeBlanc, Matthieu (2013). Translators on translation memory (TM). Results of
an ethnographic study in three translation services and agencies. Translation and
Interpreting 5 (2), 1–13.
LISA (1998). LEIT phase I nal report. LISA Education Initiative Taskforce (LEIT),
Report November, 22.
Lommel, Arle and Ray, Rebecca (2007). LISA globalization industry primer. Ro-
mainmótier: LISA.
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 263
Lörscher, W. (1991). Translation performance, translation process, and translation
strategies: A psycholinguistic investigation. Target 4 (1), 126-129.
Ludwigsen, Scott, Williams, Jeff, y Polis, Jennifer (2009). The Guide to Translation
& Localization. Portland: lingo-systems.
Massion, Francois (2010). Automating translation quality management. Multilin-
gual, 21 (1), 41.
Mata, Manuel (2005). Localización y traducción de contenido web. In Traducción
y localización. Mercado, gestión y tecnologías. Detlef Reineke (ed.), 187-252. Las
Palmas: Anroart.
— (2009). Algunas pautas para el tratamiento de imágenes y contenido gráco en
proyectos de localización (II). EntreCulturas 1, 533-569.
Medina, Ana and Ramírez, Cristina (2015). La localización de la sección” produc-
tos” en sitios web de empresas exportadoras agroalimentarias [online]. InTRAlinea:
Online Translation Journal. <http://www.intralinea.org/specials/article/la_localiza-
cion_de_la_seccion_productos_en_sitios_web_de_empresas> [Accessed 2 April
2016]
McEnery, Tony and Wilson, Andrew (2001). Corpus linguistics: An introduction.
Edinburgh: Edinburgh University Press.
McKethan Kenneth and White, Graciella (2005). Demystifying software globali-
zation[online]. Translation Journal 9 (2), 1-8. <http://translationjournal.net/jour-
nal/32global.htm> [Accessed 2 April 2016]
Meersseman, Yann. The customer makes the difference. In LISA Best Practice
Guide. Quality Assurance- the client perspective. LISA (ed.), 35-37. Switzerland:
The Localization Industry Standard Association.
Mossop, Brian (2006). Has computerization changed translation?. Meta: Journal
des traducteursMeta:/Translators’ Journal 51 (4), 787-805.
Nord, Christiane (2005). Text analysis in translation: Theory, methodology, and
didactic application of a model for translation-oriented text analysis. Amster-
dam-New York: Rodopi.
Oliver, Antoni, Moré, Joaquim and Climent, Salvador (2008). Traducción y tec-
nologías. Barcelona : Editorial UOC, 2008.
O’Hagan, Minako (2009). Evolution of user-generated translation: Fansubs, trans-
lation hacking and crowdsourcing. Journal of Internationalisation and Localisation
1 (1), 94-121.
Olohan, Maeve (2004). Introducing corpora in translation studies. London-New
York: Routledge.
— (2011). Translators and translation technology: The dance of agency. Translation
Studies 4 (3), 345–357.
PACTE Group (2011). Results of the validation of the PACTE Translation Com-
petence model: Translation project and dynamic translation index. En Cognitive
Explorations of Translation. Sharon O’Brien (ed.), 30-53. London-Nueva York:
Continuum Studies in Translation.
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
264
Presas, M. Lluïsa (1996). Problemes de traducció i competència traductora. Bases
per a una pedgagotia de la traducción. Tesis doctoral, Universitat Autònoma de
Barcelona.
Pym, Anthony (2004). The Moving Text. Localization, Translation, and Distribu-
tion. Amsterdam and Philadelphia: John Benjamins.
• — (2010). Exploring Translation Theories. London-New York: Routledge.
— (2011a). Website localization. In The Oxford Handbook of Translation Studies.
Kirsten Malmkjaer and Kevin Windle (eds), 410-23. Oxford: Oxford University
Press.
— (2011b). What technology does to translating. Translation and Interpreting 3
(1), 1–9.
Reppen, Randi (2010). What are the key considerations?. In The Routledge Hand-
book of Corpus Linguistics. Anne O’Keeffe and Michael McCarthy (eds.), 31-37.
London-New York: Routledge.
Rojo, Ana (2013). Diseños y métodos de investigación en traducción. Madrid: Ed-
itorial Síntesis.
Roturier, Johann, and Lehmann, Sabine (2009). How to treat GUI options in IT
technical texts for authoring and machine translation. The Journal of International-
ization and Localization 1(1), 40-59.
Safar, Libor and Machala, Jiri (2010). Best practices in localization testing. Multi-
lingual 21(1), 28-32.
Sandrini, Peter (2005). Website localization and translation. In EU-High-Level Sci-
entic Conference Series: Challenges of Multidimensional Translation . Heidrun
Gerzymisch-Arbogast and Sandra Nauert (eds.), 131-138. Saarbrücken: MuTra
Conference Proceedings.
— (2008). Localization and Translation. In LSP Translation Scenarios. Heidrun
Gerzymisch-Arbogast, Gerhard Budin and Gertrud Hofer (eds.), 167. Wiesbaden:
ATRC Group.
Schäler, Reinhard (2007). Reverse localisation. Localisation Focus: the Interna-
tional Journal of Localisation 6 (1), 39-48.
— (2008). Linguistic resources and localization. In Topics in Language Resources
for Translation and Localization. Elia Yuste (ed.), 195-225. Amsterdam-Philadel-
phia: John Benjamins.
Sinclair, John (2002) Preface. In Small Corpus Studies and ELT. Theory and Prac-
tice. Mohsen Ghadessy, Alex Henry and Robert L. Roseberry (eds.), xii-xv. Amster-
dam-Philadelphia: John Benjamins.
• Sikes, Richard (2010). Five types of localization testing. Multilingual 14 (7), 4-8.
Sichel, Barb (2009). Planning and writing for translation. Getting Started Guide:
Writing for translation. MultiLingual, 20 (7), 3-4
Somers, Harold (2003). The Translator’s Workstation Computer and Translation:
A Translator’s Guide. Amsterdam-Philadelphia: John Benjamins.
• Toon, Alison (2015). Going global via the cloud. Multilingual26 (1), 26-27
se n d e b a r
issn-e 2340-2415 | Nº 27 | 2016 | pp. 235-266 265
Toury, Gideon (2010). What’s the problem with ‘translation problem’? Meaning in
Translation. Barbara Lewandowska-Tomaszczyk and Marcel Thelen (eds.), 231-
252. Frankfurt am Main: Peter Lang.
Vine, Andrea S. (2002). Internationalization in Software Architecture and Design.
Multilingual 13 (3), 15-22
Wilss, Wolfram (1982). The science of translation: Problems and methods. Tübin-
gen : Gunter Narr Verlag
Wu, Ye and Offutt, Jeff (2002). Modeling and testing web-based applications. GMU
ISE Technical ISE-TR-02-08. Technical reports, George Mason University.
Yang, Yanxia (2007). Extending the user experience to localized products. In Us-
ability and Internationalization. Global and Local User Interfaces. Nuray Aykin
(ed.), 285-292. Springer Berlin Heidelberg.
• Online Resources
EAGLES. What is a corpus? EAGLES Text Corpora Working Group Read-
ing Guide <http://www.ilc.cnr.it/EAGLES96/corpintr/node13.html#SEC-
TION00040000000000000000> [Accessed 2 February 2016]
GALA. Terminology. <http://www.galaglobal.org/> [Accessed 2013]
GALA. About GALA. <http://www.galaglobal.org/> [Accessed 5 March 2016]
La Vanguardia. Zara causa polémica en Alemania por la venta de unas ‘san-
dalias de esclava’. <http://www.lavanguardia.com/de-moda/moda/mu-
jer/20150409/54429772886/zara-polemica-alemania-sandalias-esclavas.html>
[Accessed 25 February 2016]
Lommel, Arle, Burchardt, Aljoscha, Uszkoreit, Hans y Melby Alan (2015). Multi-
dimensional Quality Metrics (MQM) Issue Types. <http://www.qt21.eu/mqm-de-
nition/issues-list-2015-12-30.html#internationalization> [Accessed 10 April 2016]
(2016). Multidimensional Quality Metrics (MQM) Denition. <http://www.
qt21.eu/mqm-denition/denition-2015-12-30.html#issue_types> [Accessed 16
October 2016]
Microsoft Developer Network. Coding for Internationalization. <https://msdn.mi-
crosoft.com/es-es/library/ms903791.aspx> [Accessed 10 April 2016]
Mozilla. Mozilla Localizability Guidelines. <http://www-archive.mozilla.org/docs/
refList/i18n/l12yGuidelines.html> [Accessed 10 April 2016]
QTLaunchPad (2014). High-Level Structure. <http://www.qt21.eu/launchpad/
node/1332> [Accessed 15 April 2016]
Roberts, Adam (2014). Span (HTML element). <http://www.sitepoint.com/
web-foundations/span-html-element/> [Accessed 15 April 2016]
Twitter. #sklavensandalen page <https://twitter.com/hashtag/sklavensandalen>
[Consulta: 15 marzo 2016]
W3C. <https://www.w3.org/International/> [Accessed 15 April 2016]
— Your rst look at JavaScript. <https://www.w3.org/wiki/Your_rst_look_at_Ja-
vaScript> [Accessed 10 April 2016]
art í cu l os o ri gin ale s
de la Cova, E. Translation Challenges in the Localization of Web Applications
266
Notes
1. LISA was a non-prot association founded in 1990 aimed at “promoting the localization and internation-
alization industry and providing a mechanism and services to enable companies to exchange and share infor-
mation on the development of processes, tools, technologies and business models connected with localization,
internationalization and related topics” (Lommel and Ray, 2007, p. 53)
2. The Globalization and Localization Association (GALA) is a leading trade association for the language
industry that supports “member companies and the language sector by creating communities, championing
standards, sharing knowledge, and advancing technology.” (GALA, 2016)
3. “The W3C Internationalization (I18n) Activity works with W3C working groups and liaises with other or-
ganizations to make it possible to use Web technologies with different languages, scripts, and cultures” (W3C
2014).
4. “JavaScript is a text-based language that does not need any conversion before being executed. (…) It is ex-
ecuted instantly by a type of program that interprets the code called a parser.” (W3C, 2012)
5. “Smartling provides an innovative, cloud-based translation management platform that lets companies quick-
ly translate and deploy their websites and dynamic web applications, mobile applications and business docu-
ments across the entire corporate digital infrastructure” (Multilingual, 2015)
6. Dropbox Help Request: https://www.dropbox.com/support/s/215/3281254/c/214
... When developing software for an international audience, the cultural references need to be considered, also called international variables, which are related to politics, economics, religion, education, linguistics, and technology, from internationalization to translation. Internationalization enables the software to support multiple languages without major engineering modifications; however, it also depends on a correct localization since, in order to adapt textual and non-text content, the localizer needs to know the location for which the product is intended, to achieve the equivalences of the new culture without major changes that compromise the quality of the product (18) . ...
... A number of studies that discuss software localization indicate the importance of professionals with technical and linguistic skills in the translation process; however, there are no guidelines for this stage, which can even be outsourced (18) . In addition, some issues are considered before and during localization, as they influence the translation process that will be adopted; for example, the nature of the project, the technologies involved, time, and the human and financial resources available (8) . ...
... Most internationalization and localization studies available are related to the experience and development of content localization tools for web or websites apps, which involve millions of lines of source code (8,9,16,18,20) . The evaluation of the textual interface of the translated software by the end users or specialists in the subject matter is rarely addressed (13) , which corroborates to reassert the originality of this study in the Brazilian scenario. ...
Article
Full-text available
Objectives: To localize the textual interface of the SureWash Pocket® mobile health app and to present the methodological experience used in the localization process. Method: Software localization study to Brazilian Portuguese. The localization process was carried out between December 2018 and June 2019 in the city of Cuiabá/MT, through six steps: permission to localize the software; two independent translations; synthesis of translations; evaluation of the validity of the textual interface by a Committee of Expert Judges; back translation; and publication. Results: The textual interfaces were organized into 107 items and analyzed by ten expert judges. In the first round, 64% of the items reached a pre-established minimum agreement level of 90%. During the second round, 39 remaining items were adapted according to consensual suggestions and the pre-final version was consolidated. Conclusion: The localization process suggests that the localized SureWash Pocket® is considered valid and can be used for an educational intervention on hand hygiene.
... The theme used for the website should also support several languages. (Cova, 2016). ...
... Si se emplea un sistema de gestión de contenidos (conocido con las siglas CMS, del inglés Content Management System ) como Wordpress o Joomla para la creación del sitio web, debe comprobarse que este es compatible con las lenguas meta a las que posteriormente se localizará el contenido. A este respecto, se recomienda informarse de antemano sobre los diferentes complementos o plugins que permiten incorporar traducciones al sitio web así como comprobar que el tema empleado para configurar la apariencia del sitio soporta varias lenguas (Cova 2016). ...
Conference Paper
Impulsadas por diversos factores tales como el desarrollo de la globalización y los constantes cambios económicos, un número cada vez mayor de empresas decide poner en marcha un proceso de internacionalización de su negocio. Para estas empresas la localización, actividad que supone tomar un producto –por ejemplo, un sitio web– y hacerlo lingüística, cultural y técnicamente apropiado para un país o región concretos, se presenta como una herramienta de enorme utilidad. En España, las pequeñas y medianas empresas (pymes) constituyen más del 99% del tejido empresarial. Sin embargo, la traducción y localización de sus sitios web no siempre presenta un nivel de calidad satisfactorio. Tras llevar a cabo una amplia revisión de la literatura relevante en los campos de traducción y localización, se efectuó un análisis de los sitios web corporativos de 70 pymes españolas con el objetivo de evaluar el grado y características de la localización web en este sector económico.
... Professional translation environments might currently be ruled by conflict, opacity and mistrust between translators and direct clients or managers, due to the computer-mediated nature of communication and of translation production networks [8][9][10]. Web localization is often the last stage before content is published [11] and, depending on the process, it may happen that localizers do not have the freedom or the possibility to make the changes necessary to adequately adapt the content to their markets [12]. ...
Conference Paper
Web localization constitutes a new field of study and professional intervention. From the translation perspective, localization equates to the website being adjusted to the typological, discursive and genre conventions of the target culture, adapting that website to a different language and culture. This entails much more than simply translating the content of the pages. The content of a webpage is made up of text, images and other multimedia elements, all of which have to be translated and subjected to cultural adaptation. A case study has been carried out to analyze the current presence of localization within Spanish SMEs from the chemical sector. Two types of indicator have been established for evaluating the sample: indicators for evaluating company websites (with a Likert scale from 0-4) and indicators for evaluating web localization (0-2 scale).
... Integrating translation and localization might require the formalization of a solid theoretical background. This can be hardly achieved, due to several factors like the confidential agreements signed by the translators, which can be a methodological constrain (de la Cova, 2016). Rojo (2013) asserts that researchers have difficulties in observing and reflecting upon the type of challenges localizers have to face when carrying out different projects, and also how they cope up with these challenges. ...
Article
Full-text available
This study proposes an adaptive methodology to overcome localization translation challenges. The objective of the study is to generate a theoretical framework for identifying localization translation problems and ultimately propose a user-centred and agile-based methodology to minimize translation errors. The main research question that this paper attempts to answer is the question of “What would be the best theoretical framework for identifying current translation problems and addressing the convergence of translation and localization according to the new developments in informatics and communication technologies?” To answer this question, it was important to dismantle the notions of translation, translation theory, and localization. Based on the revised new definitions adapted to the new socio-technological context of the present digital era, the challenges can be identified and addressed through the formulation of a new methodology. The new methodology involves several steps, including the selection of recognized techniques like the “rich points” model to identify the localization translation challenges, a set of quality criteria to evaluate the projects, and adopting a user-centred approach and agile methodology for the project management of localization translation projects in order to assure the satisfaction of the stakeholders and a rapid adaptation to changes in the requirements. The proposed methodology must be validated in the future by applying it to concrete cases of localization translation projects and assessing its utility and performance. Thus, it would be useful in the future for improving localization translation projects.
... The theme used for the website should also support several languages. (Cova, 2016). ...
Article
With the emergence of video game localisation studies in the last decade, scholarly interest in translation studies in this young discipline has increased. Although globalisation has encouraged video game companies to offer their products in as many languages and markets as feasible, this academic discipline is still an under-researched area, especially in the Arab context. This article presents an overview to engender a better understanding of the nature of video games and their localisation in the Arab market. This market is distinct due to its culture, politics and language. In addition to the translational challenges related to the Arabic language, this study deals with the specificities of video game localisation from political, cultural and linguistic perspectives. This research offers a historical overview of the localisation of video games into Arabic since the inception of the practice, including an outline of the main localisation milestones in this field.
Chapter
Full-text available
Transcreation is a relatively new practice in the linguistic sector which has become more important throughout recent years. Both companies and researchers have contributed to the definition of the phenomenon. However, the process has not involved those who carry out this work first-hand: transcreators, translators, linguists , etc. In this article, our aim is to learn the perceptions of the professionals who work in transcreation projects in order to determine whether the current definitions of the phenomenon are complete. For this, we have interviewed a qualitative and varied sample of members of the sector. The results suggest that the professionals share the current view of transcreation, although they highlight a second perception of the phenomenon as a multidisciplinary strategy that would overlap with other modalities of a creative nature. It is concluded that it is necessary to undertake additional research in order to confirm the validity of this new proposal.
Article
Full-text available
The sharp rise in the use of technology tools in the translation process has rendered human translators more invisible than ever. The importance of the role played by human translators in translation, however, cannot be denied or understated. This paper aims to examine the primary factors influencing the work of human translation combined with translation technology tools. Therefore, the paper provides an overview of the translation and language industries and insights into translation industry standards, quality concerns and the most frequently used tools, as they are aspects that influence and condition the translator's work today. The final and main section in this work emphasizes an increasingly common trend in translation: a human-assisted machine translation model based on the post-edition of the output from machine translation systems. By analyzing market studies, surveys and papers on the aforementioned aspects, this article confirms that the role of human translators in technology-driven translation processes will be as central in the future as it is today.
Article
Full-text available
Though we lack empirically-based knowledge of the impact of computer-aided translation (CAT) tools on translation processes, it is generally agreed that all professional translators are now involved in some kind of translator-computer interaction (TCI), using O'Brien's (2012) term. Taking a TCI perspective, this paper investigates the relationship between machines and humans in the field of translation, analysing a CAT process in which machine-translation (MT) technology was integrated into a translation-memory (TM) suite. After a review of empirical research into the impact of CAT tools on translation processes, we report on an observational study of TCI processes in one particular instance of MT-assisted TM translation in a major Danish translation service provider (TSP). Results indicate that the CAT tool played a central role in the translation process. In fact, the study demonstrates that the translator's processes are both restrained and aided by the tool. As to the restraining influence, the study shows, for example, that the translator resists the influence of the tool by interrupting the usual segment-by-segment method encouraged by translation technology. As to the aiding influence, the study indicates that the tool helps the translator conform to project and customer requirements.
Article
Full-text available
Most conspicuous initially with Japanese anime fansubs, fan-based translation has been developing since the 1980s. In the advent of widespread availability of Web 2.0 platforms, fan translation has become a global phenomenon despite its dubious legal status. Extending the emerging interest in fansubs and scanlation in translation studies to the little discussed translation hacking by video game fans, this article brings readers‘ attention to participatory culture manifest in user-generated content in the field of translation and localisation. The article describes the evolution from unsolicited fan translation to solicited community translation now called crowdsourcing and considers them in the framework of user-generated translation (UGT). The article provides interdisciplinary perspectives, drawing on insights from media and game studies to address UGT which could profoundly impact the profession of translation and localisation as new technological environments unleash the technical competence, genre-knowledge and unparalleled devotion of the otherwise untrained Internet crowd as translators.
Article
The use of corpora in translation studies, both as a tool for translators and as a way of analyzing the process of translation, is growing. This book provides a much-needed assessment of how the analysis of corpus data can make a contribution to the study of translation. Introducing Corpora in Translation Studies: • traces the development of corpus methods within translation studies • defines the types of corpora used for translation research, discussing their design and application and presenting tools for extracting and analyzing data • examines research potential and methodological limitatis • considers some uses of corpora by translators and in translator training • features research questions, case studies and discussion points to provide a practical guide to using corpora in translation studies. Offering a comprehensive account of the use of corpora by today's translators and researchers, Introducing Corpora in Translation Studies is the definitive guide to a fast-developing area of study.
Article
Translation and Globalization is essential reading for anyone with an interest in translation, or a concern for the future of our world’s languages and cultures. This is a critical exploration of the ways in which radical changes to the world economy have affected contemporary translation.
Article
Because of the hype surrounding computers, it is hard to determine whether they are doing anything more than speed up the writing and research process. Changes such as the advent of "collage" translations - where phrases are pasted into translations from old translations or original TL documents - have been enabled by technological change but they are driven by changes in the translation business. Plan de l'articleDiscerning the newDearth of observations of the workplaceChanges in translation as a businessThe demand for speed and the advent of "collage" translationMultimedia translation and changes in the division of labourPossible impact of voice recognition softwareImpact of English as a global languageThe view from the future