With the aim of contributing to the internationalization of Spanish small and medium-sized companies, in this section we propose a series of practical recommendations for the localisation of corporate websites. This set of recommendations is based on the classification of criteria for the evaluation of the quality of the localisation proposed by Gutiérrez-Artacho and Olvera-Lobo (2017). This classification groups the criteria that influence the quality of the localisation in three large groups: linguistic, cultural and technical criteria.
The importance of linguistic aspects of localisation has its origin in the close relationship between localisation and translation. Several authors have described this relationship in different ways: while Esselink (2001) considers translation a fundamental phase of the localisation process, other authors consider localisation as a type of translation. Hurtado Albir (2001) uses the term traducción de productos informáticos instead of localización; even so, she considers that it is a clearly differentiated translation modality that includes several types of translation (legal, technical or audiovisual, for example).
Regarding the cultural character of the localisation, we can find its foundation in the widely described relationship between language and culture. One of the first to point out that relationship was Nida (1945): «The person who is engaged in translating from one language into another ought to be constantly aware of the contrast in the entire range of culture represented by the two languages». Hurtado Albir goes in depth in this relationship affirming that «la traducción no sólo se produce entre dos lenguas diferentes, sino también entre dos culturas diferentes; la traducción es, pues, una comunicación intercultural. El trasvase de los elementos culturales presentes en un texto es uno de los mayores problemas a que se enfrenta el traductor» (2001). It is logical to affirm that if translation and culture are two intimately linked concepts, also localisation and culture are. In the same way, localisation, as a process whose target is usually computer products such as applications or web pages, cannot be studied or developed without taking into account the various technical aspects that characterise it.
For the elaboration of these recommendations, the objective set is that they can be used for the localisation between any pair of languages. Despite this, it is true that the vast majority of examples that are offered involve the Spanish and English languages since, according to other studies that we are currently developing, they are one of the most common language pairs in the context of the localisation of Spanish pymes websites.
Most of the examples shown are drawn from Jiménez Crespo (2008).
The translator-localiser should pay attention to possible translation errors that affect the content. Some of these errors are the following:
- Misinterpretation: translation error that consists of attributing to the target text a sense contrary to that of the original text.
- Incorrect meaning: it consists of attributing to the target text a different meaning to that of the original text. For example, an incorrect meaning would be made when translating site map as mapa de la página web since the map represents the structure of the website in its entirety; not a single page. Therefore, the correct translation would be mapa del sitio web.
- Nonsense: formulations without sense.
- Omissions of information.
You can consult, among other authors, Hurtado Albir (1999) to obtain a more exhaustive list of possible translation errors.
1.2.1. Calques, loanwords and false friends
In the website field, due to the influence of the North American hegemony, it is frequent the use of calques or loanwords from English that are unnecessary because they have acceptable equivalents in Spanish. An example of calque from English would be the translation of go (the button that activates the search in the search engine) as *ir, instead of using a usual alternative in original texts in Spanish such as buscar. It would also be inadvisable to use raw loandwords (hosting) when there are equivalents in Spanish (alojamiento web), although it is true that there may be commercial reasons and web positioning to use some anglicisms especially extended among target speakers. False friends, as in any translation, should be avoided: success as translation of suceso, localidad as translation of location, advertisement as translation of advertencia.
1.2.2 Lack of terminology consistence
The use of terms within the website should be consistent, so it would be an error to alternate between the original and translated form of the same term (mapa del web/site map) or use different formulations to make reference to the same section (Acerca de nosotros / Sobre nosotros).
1.2.3 Lack of terminology precision
The lack of terminological precision generates confusion in the receiver and causes a bad impression. In Spanish, a link called Carreras in a website of an enterprise linked to the world of sport could take us both to a section related to athletics and to a section in which employment is offered. Paying attention to this type of ambiguities and solving them will increase the quality of the final product.
1.2.4. Incorrect use of initials and acronyms
As Jiménez Crespo (2008) states, style manuals often discourage the use of acronyms on websites due to their lack of uniqueness. However, in the case that it is decided to use them, it is required to develop them, and English-speaking acronyms must not be used together with a translation into Spanish that does not correspond to the acronym (*See PDA (Ayudante Digital Personal)).. In addition, they must be pluralised correctly depending on the language to which they are being translated: therefore, the Spanish plural for PC would be PC while in English PCs or PC’s could be used.
1.3.1. Syntactic calques
The calque of syntactic structures of the source language that are unnatural in the target language: *ir atrás (go back), *es operado por (is operated by) should be avoided.
1.3.2 Respect of grammatical rules of the target language
Possible errors of agreement and gender (sometimes caused by the lack of context during the translation): *is one brands of…; *necesitamos poder contar con un equipo de expertos should be checked; misuse of reflexive verb forms (*contáctese con…), etc.
1.3.3 Calques of source language structures
One example of this error occurs when the personal pronouns (usted, tú) are explained more frequently than what is idiomatic in Spanish, tracing the use of other languages such as English, French or German. Another example that is usually given in legal texts of websites is the recurrent use of the determinant in expressions such as *estos Términos y Condiciones o *estas reglas, while the most usual in Spanish original texts would be the form los presentes… (Términos y Condiciones / reglas).
Localisation also involves adapting the text to the stylistic conventions of the target language. In this sense, one must attend, for example, to the general tendency of the language in which it is being written with respect to the length and complexity of the sentences. In this way, it would be an error to use long sentences with a lot of subordinates in a language such as English that tends to the use of simple and brief sentences.
1.4.1 Inappropriate register
As Bolaños (2003) describes, the predominant register on the Internet, influenced by the North-American culture influence, is the colloquial one, while in Spanish this is usually higher and it often demands the use of formal treatment. The following translations into Spanish are examples of texts that do not reflect the proper register, which affects their idiomaticity: *Usted está sujeto a los siguientes Términos y condiciones, que podemos actualizar de vez en cuando sin notificarle a usted de ello; *Cualquier cosa en este site tiene copyright. It would also be a mistake to alternate between different register levels or between formal and informal treatment (tú/usted).
The orthotypography of the text must follow the rules of the target language. In this section, aspects such as the following must be taken into account:
- The correct use of capital letters or small letters, as required to the language, for titles of works or part of works, months of the year or languages.
- The use of the most appropriate type of quotes according to the language (in Spanish, for example, it would be the angle quotation marks («») before the English ones ("") or the simple ones ('').
- Opening question and exclamation marks should not be omitted in Spanish (*Obtenlo ya!; *No se lo pierda!).
It should also be checked that the typography of the text is consistent.
1.5.1 Badly written words (cacographies)
For this aspect, it is advisable to use a typographic checker in the last phases of localisation.
Using the right linguistic variety will not only make a better impression on the users but it will also have a positive impact on the number of visits the site gets - the user will only be able to access a certain site if the keywords they use when searching on a browser are the same as the ones that are used on the site. Thus, it would be advisable for a clothing company to use the word 'suspenders' for the US version and 'braces' for the UK version of their website.
The information detailed in this section, except where indicated otherwise, is based on Microsoft (Dr. International 2003).
It is necessary to adapt prices and other mentions of amounts of money to the currency of the locale. Regarding the position of the currency symbol, which may be a symbol such as ‘€’ or a three-letter code (‘EUR’), the convention of the locale must be followed: in the case of Spanish of Spain (ES_ES), the currency symbol is postponed to the figure and is separated by a space: 15 €. On the contrary, in the Spanish of America, the symbol is usually preceded and is not separated by a space: $15,429. With respect to the three-letter codes, they are always separated by a space from the figure, and in America it is acceptable that they are preceded: USD 50,82 (RAE y ASALE, 2010).
The date format must also be adapted to the locale. Depending on it, for example, the DD/MM/YYYY format (the usual one in Spain), MM/DD/YYYY (typical for example, of the United States) or YYYY/MM/DD will be followed. Attention must be paid to the symbol that is used to separate days, months and years: for example, the slash, the dash or the full stop (typical, for example, of the German: DD.MM.YYYY) can be used.
You can also use a long format combining numbers and letters (February 22, 2018). In that case, the Spanish norm, unlike other languages such as the English, indicates that the name of the month is always in lowercase. Also, in Spanish the years don’t use a full stop (it would be wrong, therefore, to write in Spanish *2.018) (RAE y ASALE, 2010).
The locale will also influence the time format used, which may be the one of 12 or 24 hours. In the case of the 12-hour format, depending on the language, it is possible that the abbreviation AM/PM goes before or after the hour and that the letters that form the acronym are separated with full stops (A.M./P.M.).
The two points (:) are the most used character to separate hours, minutes and seconds, but are not the only possibility, as is the example of some Asian languages that use ideograms.
The character used to separate the thousands can be a comma (1,000; United States) or a full stop (1.000; Spain, Germany). The same happens with the character that separates decimals (its use is the reverse in the examples of locale cited). The position of the % symbol can be preceded or postponed to the figure, with or without space, or it can be represented by the letters pct.
The measurement units constitute another aspect susceptible of variation depending on the locale. Regardless of the measurement unit used, it is essential that it is explicit so as not to give rise to confusion (38 ℃, 75 kg, 1 sheet of paper A4).
They should always include the international prefix to facilitate communication with international users (00 34 123 456 789 / +34 123 456 789).
In their creation, you must pay attention to some aspects:
- 'Province', 'autonomous community' or 'state' fields: it is recommended that their completion is optional, since this may not proceed depending on the country of origin of the user.
- 'Postal code' field: the number of digits that can be introduced must be flexible and admit letters.
- 'Telephone' field: it must admit a very variable number of digits.
- 'Surname' field: in a website that will be used by people of different sociocultural origins, the user should be offered enough flexibility to indicate one or more surnames. Consequently, also the name of the field must reflect that flexibility (Surname or surnames, for example) (Jiménez Crespo, 2008).
It is important to consider the possible cultural charge implicit in images, colours and icons that appear on the web page. Flags should be avoided in areas such as the language selector, since they rarely represent all the speakers of a language (a British flag does not represent the speakers of the United Stated, for example). On the other hand, especially in the case of distant cultures, the cultural significance of the colours used in the site must be taken into account, for which Andreu-Vall and Marcos (2012) recommend consulting the graphic https://informationisbeautiful.net/visualizations/colours-in-cultures/. In the same work, there is a warning about the possibility of images having negative connotations for the target culture or that it would be necessary to adapt some icons ("an overused example in usability is the mailbox icon to receive mail, so different between the Spanish one, usually square and flat, attached to a wall or a door, and the American one, represented as a post in front of the entrance of a house with a garden, with a more cylindrical format").
Whenever possible, links to other websites must redirect to their version in the target language of our website. If this version does not exist, it is advisable to inform the user about the language in which the context of the link is written: www.ideal.es (in Spanish).
Ideally, the localisation should cover all aspects of the interaction with the user, including customer service. There are different ways of offering customer service in the target language, which include from both telephone and email attention to the use of conversational bots or chatbots.
When there are differences in the shared knowledge of the source and target culture, it will often be necessary to provide explanations that facilitate the understanding of the information. For example, in a web page about tourism in Spain localised for the Chinese public in which several Spanish cities are mentioned, it could be very convenient to explain the location of these cities in the map: Oviedo, a city in the north of Spain… It would be a mistake, however, to add incorrect information or whose knowledge is presupposed to the receiver. For example, on a website intended for the health sector, it would probably be unnecessary to define terms such as apoplexy or rhinoplasty.
Although, as a minimum, the news section should be translated just like the rest of the website, ideally, relevant news should be published for the target audience; for example, the celebration of events that will take place in the corresponding region or the opening of a new branch of the business in the country in question. In the case of websites selling products, the most convenient would be that in the version of the website adapted to a locale and a particular region, only products available in the region of the locale would be offered. This is how the IKEA company does on its website: www.ikea.com.
In social networks, it is highly recommended to have different profiles for each locale. On the other hand, it is necessary to consider the possibility that specific social networks or research engines especially popular among the target audience exist. In Russia, for example, VK is one of the most used social networks and Yandex is the preferred search engine, while Yahoo is popular in Japan. According to these examples, if our website is localised to Russian or Japanese, we must check and enhance the positioning of the website in these search engines.
Previously, the Ikea site has been mentioned as an example of website where products adapted to the region in which the consumer is located are offered. This is possible thanks to the fact that on its home screen the site asks the user to indicate his region and his language: for example, the Belgium site can be visited in both Dutch and French. This is an excellent example of how to build a website taking into account regional and language differences; although one possible improvement would be to add more languages per country or region (for example, it would probably be interesting for a lot of consumers who live in Germany but do not master German that the IKEA site was localised in English).
On the other hand, it is highly recommended that the choice of language selection is accessible from any page, and that when choosing a new language the site does not redirect to the home page (Andreu-Vall and Marcos, 2012).
In addition to working properly, the user should be able to choose between searching in all languages or only in one.
As Andreu-Vall y Marcos (2012) state, a target context tag must be included in the section head of the code of the website indicating that the coding would be UTF-8 (8-bit Unicode transformation format). This type of coding allows that the browser interpret the characters of the vast majority of alphabets. One example of this type of tag would be:.
On the other hand, if any of the target languages to which the website will be translated uses right to left writing (Arabic, Hebrew) the web developer must take it into account so that it does not cause problems at the moment of the localisation and the final result presents a correct display.
The different versions of the website must offer the same information as the original site, as long as that information is relevant to the target audience.
Downloadable documents (PDF forms, .doc files) must also be subject to the localisation process.
Likewise, at the beginning of the HTML document, the language in which the page is written must be said, as well as the variant of the country. For example, the code for Spanish of Spain would be <html lang=”es”>. If any part (word, sentence or paragraph) within a web page appears in a language different from the rest of the page, it must be indicated in the code. For example, a paragraph in Spanish within a page in English would be marked with the paragraph tag followed by the language attribute:
(Andreu-Vall y Marcos, 2012).
If you use a content management system for the creation of the website (WordPress, Joomla), it should be checked that this is compatible with the target languages to which it will be translated. In this regard, it is recommended to learn beforehand about the different plugins that allow to incorporate translations to the website as well as check that the subject used to configure the site appearance bear several languages (De la Cova, 2016).
Responsive web design «comprende una serie de técnicas y pautas de diseño que permiten adaptar sitios web al entorno de navegación del usuario, entendiendo como entorno de navegación la multiplicidad de dispositivos, móviles o no, por medio de los cuales los usuarios pueden acceder y navegar en internet» (González y Marcos, 2013). If the website is developed using a responsive web design, the web page will offer a good display both in computers and in other devices (mobiles, tablets).
3.9.1. Localised pictures, audios and videos
To localise the content inside images, image-editing programs such as GIMP or Adobe Photoshop are useful. To obtain a full study about this aspect, Mata Pastor (2009a, 2009b) can be consulted.
Audios and videos that are offered in the web page must be translated into the different target languages, either by the use of subtitles or dubbing. In this point, attention also must be paid to the cultural aspects: for example, in Muslim countries the usual is the use of male voices for dubbing (Martin, 2002).
Web accessibility constitutes a very wide field. Some key points that must be taken into account are:
- Alt and title tags of the images must be translated, because they do not only benefit the web page’s positioning in the search engines, since they index the content through tags, but also they allow people with visual disabilities identify the content of the images through special reading programs (Andreu-Vall and Marcos, 2012). In the same way, another reason for which the use of flags in the language selector is not recommended, besides the ones mentioned in the previous section, is because they impose accessibility barriers when a textual alternative is not offered that allow people with visual disabilities identify the language that they want to select (Rodríguez Vázquez, 2015).
- There are computer programs that allow to verify the accessibility level of a website. One example is aDesigner.
- The World Wide Web Consortium (W3C) offers a series of guidelines for accessibility of web content. These guidelines can be consulted here: https://www.w3.org/TR/WCAG20. It can also be used the reference fast guide that is offered in https://www.w3.org/WAI/WCAG20/quickref/?showtechniques=11.
The way in which the information is structured on the website must be taken into account, and if necessary, make the convenient modifications. From the textual point of view, it should be understood that users tend to scan the information on the screen instead of reading the text from beginning to end as when it is printed on paper, so both structure of the texts and style must be especially clear, with brief and concise sentences. Regarding the different elements of the website, the ideal is that its layout is simple and quickly recognisable.
There are several factors that affect the positioning of the website in search engines. Andreu-Vall and Marcos (2012) mention the following:
- Translation of the title tag (head): is the title of the web page, the one that appears at the top of the browser and the search results when the web page is retrieved in a search engine. It is important to translate it not only so that users can identify the page but also so the search engine robots index the title in the corresponding language.
- Translation of the ‘description’ and ‘keywords’ tags: they are not visible from the browser because they are in the headsection of the source code, but search engine robots do read and index them. In addition, the ‘description’ tag is used by search engines on many occasions to show the summary they offer in the results list.
- Translation of the ‘alt’ and ‘title’ tags: see section 3.9.2.
- Translation of the ‘title’ tags of the links: it is the one shown when the cursor is over the links. The search engine robots can also access them.
A web localisation project necessarily requires a phase in which it is check that the final product has a satisfactory quality. To do this, it will be necessary to carry out a series of checks such as those suggested by Sandrini (2008):
- Check that the localised website works in a correct and identical way to the original.
- Check that the source text has been translated and that the target text has a satisfactory linguistic quality.
- Check the appearance of the website localised: all the text must be visible, the text of the images must have been translated and the format must be the correct one. It is recommended to carry out these checks using different browsers and, if possible, different devices.
Andreu-Vall, M.; Marcos, M.C. (2012). “Evaluación de sitios web multilingües: metodología y herramienta heurística”. El profesional de la información, 21 (3), pp. 254-260, DOI: 10.3145/epi.2012.may.05.
Bolaños Medina, A. (2003). “Dificultades y estrategias de la localización de sitios web comerciales del inglés al español”. Posteguillo, S. et al. (eds.), Internet in Linguistics, Translation and Literary Studies. Castelló de la Plana: Publicacions de la Universitat Jaume I, p. 241-262.
Cadieux, P.; Esselink, B. (2002). “GILT: Globalization, Internationalization, Localisation, Translation”. Globalization Insider, 11 (1.5), pp. 1-5.
de la Cova, E. (2016). “Translation Challenges in the Localisation of Web Applications”. Sendebar, 27, pp. 235-266.
Dr. International (Microsoft Corporation) (2003). Developing International Software. Redmond: Microsoft Press, ISBN: 0-7356-1583-7.
Esselink, B. (2001). A Practical Guide to Localisation. Amsterdam-Filadelfia: John Benjamins.
González, D.; Marcos, M.C. (2013). “Responsive web design: diseño multidispositivo para mejorar la experiencia de usuario”. BiD: textos universitaris de biblioteconomia i documentació, 31, DOI: 10.1344/BiD2014.31.19.
Gutiérrez-Artacho, J.; Olvera-Lobo, M.D. (2017). “El uso del método Delphi como herramienta de evaluación consensuada en la didáctica de la traducción: el perfil del traductor-localizador”. Del Valle, M.E. (coord.), Congreso CUICIID 2017: Congreso Universitario Internacional sobre la comunicación en la profesión y en la Universidad de hoy: Contenidos, Investigación, Innovación y Docencia, 25-26 de octubre, en línea, p. 226-240.
Hurtado Albir, A. (2001). Traducción y traductología: introducción a la traductología. Madrid: Cátedra.
Hurtado Albir, A. (ed.) (1999). Enseñar a traducir: metodología en la formación de traductores e intérpretes. Teoría y fichas prácticas. Madrid: Edelsa, p. 120.
Jiménez Crespo, M.A. (2008). “El proceso de localización web: estudio contrastivo de un corpus comparable del género sitio web corporativo”. Tesis Doctoral. Universidad de Granada.
LISA (2003). The Localisation Industry Primer. Fry, D.; Lommel, A. (eds.). Ginebra: Localisation Industry Standards Association.
Martin, M. (2002). “La localización de software: una especialidad incipiente en los Estudios de Traducción”. Bravo, J.M. (ed.), Nuevas perspectivas de los estudios de Traducción. Valladolid: Universidad, pp. 295-313, ISBN: 84-8448-187-5.
Mata Pastor, M. (2009). “Algunas pautas para el tratamiento de imágenes y contenido gráfico en proyectos de localización (I)”. Entreculturas: revista de traducción y comunicación intercultural, 1, pp. 513-532.
Mata Pastor, M. (2009). “Algunas pautas para el tratamiento de imágenes y contenido gráfico en proyectos de localización (II)”. Entreculturas: revista de traducción y comunicación intercultural, 1, pp. 533-569.
Nida, E. (1945). “Linguistics and Ethnology in Translation-Problems”. WORD, 1:2, pp. 194-208, DOI: 10.1080/00437956.1945.11659254.
Pierini, P. (2007). “Quality in web translation: An investigation into UK and Italian tourism websites”. The Journal of Specialised Translation, 8, pp. 85-103.
Real Academia Española y Asociación de Academias de la Lengua Española (2010). “Expresiones numéricas específicas”. Ortografía de la lengua española. Madrid: Espasa Calpe, ISBN 978-6-070-70653-0.
Rodríguez Vázquez, S. (2015). “Exploring current accessibility challenges in the multilingual web for visually-impaired users”. Proceedings of the 24th International Conference on World Wide Web, 18-22 de mayo, Florencia (Italia). New York: ACM, pp. 871-873, DOI: 10.1145/2740908.2743010, ISBN: 978-1-4503-3473-0.
Sandrini, P. (2008). “Localisation and Translation”. Gerzymisch-Arbogast, H.; Budin, G.; Hofer, G. (eds.), MuTra Journal, Vol 2. LSP Translation Scenarios. Selected Contributions to the EU Marie Curie Conference Vienna 2007. Saarbrücken: ATRC, pp. 167-191.
World Wide Web Consortium (2008). “Web Content Accessibility Guidelines (WCAG) 2.0”. https://www.w3.org/TR/WCAG20/.