Home > Issue25 > Flanagan article

Cause for concern? Attitudes towards translation crowdsourcing in professional translators’ blogs

Marian Flanagan, Aarhus University

ABSTRACT

This paper seeks to identify professional translators’ attitudes towards the practice of translation crowdsourcing. The data consist of 48 professional translator blogs. A thematic analysis of their blog posts highlights three main findings: translation crowdsourcing can enhance visibility of the translation profession, but fails to enhance visibility of the professional translator; ethical concerns are raised regarding translator participation in non-profit translation crowdsourcing, and the shifting of responsibility from the professional to the non-professional translator; professional translators do not openly discuss their motives for differentiating between the various non-profit initiatives, and while there is much discussion on translation crowdsourcing for humanitarian causes, little or no attention is paid to free and open source software projects.

KEYWORDS

Translation crowdsourcing, translator blogs, ethics of translation, thematic analysis.

1. Introduction

The amount of digital content to be translated is growing at an astounding rate since the development of the World Wide Web, but the number of professional translators is not sufficient to meet the demand of individuals and businesses (Kelly 2009a, Vashee 2009). Moreover, budgets to fund this demand for translation do not exist (European Commission 2012: 75). As a result, two options to address these shortfalls have emerged: the use of machine translation (MT)¹ and translation crowdsourcing (Garcia 2015: 19).

The term crowdsourcing was coined by Jeff Howe, a contributing editor with Wired Magazine, when he wrote an article titled The rise of crowdsourcing (2006). Crowdsourcing describes the outsourcing of jobs, typically performed by in-house employees and professionals, to a large, undefined (most often virtual) crowd. In his article, Howe highlights how advances in technology have made the gap between amateurs and professionals smaller, giving hobbyists and amateurs the opportunity to showcase their skills, and companies can in turn use these skills to their own advantage. Crowdsourcing does not automatically mean free labour, but it can mean paying a lot less than employing a professional.

Referring to the open source movement, which advocates free and open source software (FOSS), Howe (2006 n.p.) suggests that crowdsourcing can be considered “the application of Open Source principles to fields outside of software”. Open Source means that anyone is freely licensed to use, copy, and edit the software. Furthermore, the source code is made available, is openly shared and users are encouraged to improve the code. FOSS is an inclusive term for free software (free referring to the price and the freedom to copy and reuse the software) and open-source software (users can access the source code). The FOSS movement, which began in the 1980s, is considered a form of crowdsourcing (Hemetsberger 2013).

Translation crowdsourcing is a translation model that reaches out to a large virtual crowd on the Internet to obtain translations (O’Hagan 2011: 14). Many businesses profit from user-generated translation (UGT), which is when unspecified self-selected individuals translate text for free (O’Hagan 2009: 97). FOSS projects benefit from translation crowdsourcing, as not only can users access software for free, but many can do so in their native language. One form of UGT dating back to the 1980s is the fan-based translation of subtitles (or fansubbing) (O’Hagan 2009: 99). However, while questions could be raised regarding the legality of fansubbing, translation crowdsourcing could be viewed as legitimising UGT (O’Hagan 2009).

Translation as a profession has a long history, but the model as we know it in the European context only emerged in the mid-twentieth century (Garcia 2009, Pym 2011). Throughout this time, the profession has continued to struggle with recognition, status and adequate remuneration (Dam and Zethsen 2008, Jääskeläinen et al. 2011). Pérez-González and Susam-Saraeva (2012: 151) highlight that non-professional translators and interpreters (those who have no formal training and work for free) “have always represented the biggest threat to labour market structures” and “the identity and livelihood of translation professionals”. Now, advances in technology which afforded new business models to emerge, have also allowed for improved visibility of the non-professional translator and translation practice among the general public as Web 2.0 technologies provide these amateurs with a platform to showcase their skills. This has been considered a “potentially significant challenge” to the professional translators’ macro environment (O’Hagan 2013: 506).

To date, little has been written to highlight the possible concerns of translators in relation to the crowdsourcing model (cf. Kelly 2009a, Kelly 2009b, Kelly et al. 2011, McDonough Dolmaya 2012). Also, much of what has been written is published as industry reports, which professional translators commonly deem as questionable sources (see section 4.2.1). This paper seeks to identify professional translators’ own attitudes towards the practice of translation crowdsourcing through a study of professional translators’ blogs. Weblogs or blogs are commonly written by individuals as a narration of thoughts and feelings (Herring et al. 2004), and often used as a personal diary, but blogs may also be used in professional contexts. Efimova (2009: 3) maintains that professional blogs can have a supporting function and argues that “the selected content a weblog author finds interesting enough to link to and to comment on functions as a public record of personal interest and engagement”. It is thus meaningful to study translators’ blogs to uncover their attitudes towards an emerging translation phenomenon such as crowdsourcing. Furthermore, this study builds on previous weblog research (McDonough Dolmaya 2011a, Dam 2013) to uncover translators’ attitudes towards translation-related topics.

2. Crowdsourcing as a translation model

Translation crowdsourcing is applied in different ways, depending on the factors involved. An overall distinction can be made between non-profit and for-profit. Within these two groups, three factors can differ: payment, the crowd, and call for participation. These three factors are elaborated on below.

Payment

Paid crowdsourcing (e.g. cloud marketplaces) remunerates participants financially (Garcia 2015: 18). More commonly, participants are offered non-financial incentives for their participation, including virtual incentives (name on a leader board, badge, mention in the company blog), material incentives (gift tokens or clothing), and invitations to exclusive events (Mesipuu 2012: 48-50; see also Risku et al. 2016 and Pym et al. 2016 in this volume). Moreover, there are situations where participants neither receive any monetary compensation nor other kinds of incentives. This is common when translation crowdsourcing contributes to a humanitarian or social cause (Munro 2010 n.p.).

The crowd

The crowd is one factor which is determined by either the company or the crowd itself. A distinction is made between open community and closed community approaches (Mesipuu 2012: 42). In short, open community means the crowd is not restricted, but participants usually have to register before they can participate. This usually results in a large crowd working on the project. Furthermore, this approach usually employs a voting system, where better translations are voted to the top by the crowd. Conversely, in a closed community participants are vetted before they can join the project. The result is a much smaller crowd. Voting systems are not commonly used within a closed community; instead, volunteer translators work closely together with those running the projects to resolve any disputes in terms of language choice or to improve translation quality (Mesipuu 2012: 42).

Participants in both communities can be professional translators and amateur participants who have knowledge of two or more languages. Collaborators in translation crowdsourcing projects are usually not professional translators (O’Brien 2011: 19). However, this situation is often different when collaborators are translating for a social cause (McDonough Dolmaya 2012: 185-186) or simply if the professional translator deems the project worthwhile (O’Hagan 2011: 13).

Call for participation

The call for participation is another factor where the decision is made by either the company or the crowd. User-initiated translation crowdsourcing is when the crowd makes the call, and it is always grouped under non-profit. The crowd aims to “make content available to others in the language they understand” (Dombek 2014: 27). Therefore, the crowd usually has an interest in the subject domain. One popular example is the translation of Wikipedia content.

Content-owner initiated crowdsourcing is when the company or organisation makes the call (Dombek 2014: 27) and can be grouped under for-profit and non-profit. Within for-profit crowdsourcing, the company can make the call directly, provide the technical platform for the project to be implemented, and recruit participants using an open or closed approach (O’Hagan 2011: 13). Alternatively, a language service provider (LSP) can initiate the call for participation on behalf of the client, when the translation crowdsourcing model is deemed the most appropriate. These two calls ask the crowd to provide translations typically for free (Dombek 2014: 29). On the other hand, a company can turn to cloud marketplaces (Garcia 2015: 26), described as a faster, less expensive option than traditional LSPs. Cloud marketplaces typically conduct paid translation crowdsourcing projects on behalf of clients.

Several NGOs, charities, and non-profit organisations use non-profit translation crowdsourcing to “promote humanitarian ideology, support disaster relief, and to spread information and knowledge” (Dombek 2014: 27). The organisation initiates the call for participation, both open and closed communities are involved, and participants translate for free.

Table 1 presents an overview of translation crowdsourcing elements initiated by non-profit and for-profit companies, with focus on payment, the crowd and call for participation.

	Non-profit		For-profit
Payment	No payment		No payment: Material incentives Virtual incentives Organised events
			Payment: Paid crowdsourcing
Crowd	Open community	Amateur Professional	Open community	Amateur Professional
	Closed community	Amateur Professional	Closed community	Amateur Professional
Call for participation	Content-owner initiated call		Content-owner initiated call
	User-initiated call		LSP makes call
			Cloud marketplaces make call

Table 1: Overview of translation crowdsourcing elements.

For the purposes of this paper, translators’ attitudes on any element of the translation crowdsourcing model (Table 1) are considered.

3. Data and methods

This paper follows the approach taken in previous blog-related studies and uses a convenience sample of blogs as the object of study, as no complete list of translator blogs exists (Trammell and Keshelashvili 2005, McDonough Dolmaya 2011a, Dam 2013). The criteria determining the sample in this study were that the blogs were written by professional translators, and that they were written in English, Danish or German, as the author conducted a qualitative thematic analysis of the texts. For the purposes of this article, a professional translator is one who earns a living by translating (Jääskeläinen et al. 2011). This definition is being used for the sake of simplicity, and can be considered a logical approach, since the people who earn a living from a profession are more likely to have concerns regarding a practice that could develop into a potential threat to their livelihood. However, it also follows that the concerns identified in these blogs cannot be taken as generalising all translators’ concerns or as representing the profession as a whole, since, by nature, the most vocal translators are those who are present on media such as blogs and might be on the lookout for ‘possible threats’ to the profession in the first place.

A list of translator blogs was first created based on lists from previous research on the topic (McDonough Dolmaya 2011a, Dam 2013), but excluding blogs that did not meet the criteria of this study. Others were then added from the Lexiophiles (2014) blog competition website, specifically from the language professional blogs category, and, following a snowball strategy, from the blogroll (a list of suggested blogs) of professional translators already on the list. These three sources generated a list of 66 translator blogs.

To verify the criterion of professional translator, each blogger profile was checked. If no information indicated that the blogger earned a living as a translator, the blog was not included in the sample. Both freelance and in-house translators were deemed suitable but for the case of in-house translators, the blog had to explicitly state that the posts reflected the translator’s views only. No in-house translators or blogs in Danish met the criteria. The final sample contains 48 blogs: English (45) and German (3) (see Appendix A). Within the convenience sample, all bloggers are freelance translators, and have on average 13 years’ experience (experience ranges from two years up to thirty years). Moreover, nearly all bloggers have a translation qualification, are members of several professional associations, and some describe continuous professional development courses they have completed.

Blog posts written about translation crowdsourcing were then extracted from these data, with no date restrictions imposed. The keywords chosen were based on relevant previous research: translation crowdsourcing, crowd, crowdsourcing, crowdsourced, volunteer, volunteer translation, community, community translation, collaborative, user-generated translation, amateur, non-professional, unprofessional, ad hoc, informal, occasional, unqualified, inexperienced, fansubs, fan (Perrino 2009, O’Hagan 2011, McDonough Dolmaya 2011a, Mesipuu 2012, Dombek 2014). Corresponding German keywords verified by a native speaker were used to search the German blogs. In total, 64 blog posts (63 English, 1 German from 26 blogs) out of a possible 10,417 posts (0.6%) addressed the topic of translation crowdsourcing and elements within the model outlined in Table 1 (see Appendix A).

These 64 blog posts were then subjected to thematic analysis to uncover bloggers’ attitudes towards translation crowdsourcing. Thematic analysis allows the researcher to identify, analyse and report patterns (themes) within data (Braun and Clark 2006). Moreover, this method is conducted within a constructionist paradigm, which assumes patterns identified in the data are socially produced, and that bloggers “use language as a form of social action” (Burr 1995: 5). Therefore, the patterns are interpreted by the researcher on two levels: first, interpreting the manifest content (themes directly observable in the text), and second, interpreting the latent content (themes underlying the phenomenon) (Boyatzis 1998). All themes are data-driven, which means that the data were not coded with specific themes or questions in mind. To reduce researcher bias, Creswell (2014: 203) suggests to cross-check codes with a second coder as one way of increasing reliability. This suggestion was applied to this study, and a second coder (one for English and one for German) was asked to code each post as positive, negative or a mix of both, and to assign the post to one or more themes developed by the researcher based on the data (discussed in detail in section 4). This coding phase achieved 92% (58/63 English posts) and 100% (1/1 German post) reliability, which can be deemed acceptable.

4. Findings and discussion

The data contain 10 positive posts, 44 negative ones, and 10 posts include both positive and negative elements. From these posts, several themes have been identified: (positive) ‘subject-matter expertise’, ‘non-profit organisations’, and ‘collaboration’; (negative) ‘free labour’, ‘ethics’, ‘quality’, and ‘authority’. In the following, positive and negative posts are discussed thematically. For each theme, pertinent examples from the blogs are presented and analysed, and these are discussed in relation to current literature in the field.

4.1. Translation crowdsourcing as a positive activity

4.1.1. Subject-matter expertise

Although the posts related to crowdsourcing are predominantly negative, the data also contain some posts that highlight a positive factor. One such factor is related to the benefit of being knowledgeable about the topic or the context in which translation takes place. The translation of the Facebook social networking platform is one such example. Originally available in English only, the platform is now available in 128 languages (April 2015). These translations were done by volunteers who received virtual incentives to participate. Blogging Translator (19 June 2009) argues that “none of the language used on it [Facebook] is highly complex” and therefore its users would “be best placed to understand the terms in context”. In other words, these amateurs have subject-matter expertise within a Facebook context. Furthermore, she highlights that since other Facebook users (predominantly amateurs) can update translations “Wikipedia style”, this would impact positively as a way to resolve “translation errors and stumbling blocks.”

Kelly et al. (2011: 75) believe that, if applied to a translation crowdsourcing project such as Facebook, the translate-edit-proofread (TEP) model used predominantly in the translation industry would have numerous drawbacks. In the TEP model the editor, who is downstream in the process, can singlehandedly veto or alter the translation, whereas the crowd would produce more accurate translations. Indeed, Facebook has presented examples of the volunteers outperforming the professional translators, “due to the former’s familiarity with the inner workings of Facebook” (Losse 2008 cited in O’Hagan 2009: 114). Furthermore, the translation of Web 2.0 terms such as the English Facebook terms ‘wall’ and ‘poke’ (Hosaka 2008) and the English Twitter term ‘unfollow’ (Sanford 2011) may be best left to the community of users themselves.

4.1.2. Non-profit organisations

Sixteen bloggers wrote posts supporting pro bono translation work within non-profit initiatives. The most commonly cited projects are: the Haiti Relief Effort, Translators Without Borders (TWB), Kiva, The Rosetta Foundation, and Global Voices. Some see this as a way of ‘starting out’ as a translator: participation in the project allows the newly-qualified translator to gain work experience and to use the experience to promote their own services. TranslationsDK (10 February 2011) notes:

there is bound to be a position available that will afford you valuable career-related experience […] and of course there is no law that states experience that goes on your résumé has to be paid!

Others see this activity as a way of giving back to a worthwhile community (e.g. Translation Times, 27 October 2011). La parole exportée (28 November 2007) discusses how appalled she is by the low rates being offered for specialised translation work on social networks for translators (proz.com in this instance). In this context she claims that she would rather volunteer with a non-profit translation project than to work for the “impossibly low rates,” because at least volunteering would give her personal satisfaction and her work would be appreciated.

As mentioned in the introduction, non-profit organisations conducting translation crowdsourcing are not confined to humanitarian causes. For example, the Firefox browser and OmegaT, a free Translation Memory tool, were developed within the FOSS framework. To date, OmegaT has been translated into 26 languages. Six bloggers mention that they use or intend to use this tool in the future. However, almost all non-profit translation crowdsourcing projects mentioned by bloggers relate to a humanitarian cause, and none suggest translating software from FOSS projects. It would be interesting to know the reason for this. It could be that professional translators are unaware of the FOSS movement, and hence of the volunteer possibilities with FOSS projects. Data in this study show that some bloggers are aware of FOSS (Thoughts on Translation, 11 February 2008), but it might not be common knowledge among the bloggers in this study or among professional translators in general. McDonough Dolmaya (2012: 176) highlights in her study of participants translating Wikipedia pages that they have a preference for humanitarian and for-profit projects over FOSS projects. Based on this finding, it would be worth investigating how professional translators rank the importance of different types of non-profit translation crowdsourcing projects.

4.1.3. Collaboration

Two bloggers discuss possible opportunities a collaborative translation setting could offer. Naked Translations (20 January 2010) cites two criteria for a successful collaborative project: well-managed projects and engaging people with sufficient skills (professionals and non-professionals). She claims the amount of information to be translated will continue to increase, which will call for more collaborative translation crowdsourcing and generate work opportunities for professional translators. While she never mentions payment explicitly, she implies that the professional translators would be paid like any other professionals engaging in a project that will add value to companies wanting to communicate with their clients in their own language.

Anmerkungen des Übersetzers (2 September 2010) discusses “social translators,” who have come together as a result of Web 2.0 technologies. These non-professionals want to translate because of their interest in the topic, and not for payment. Therefore, social translators could be deemed a threat to the professional translator. This leads the blogger to consider the relationship between social translators and professional translators: are they competitors, colleagues or partners? To help answer this question, the blogger cites Ethan Zuckerman from Global Voices Online (a non-profit citizen media website), who argues that translation crowdsourcing is not a ‘one solution fits all’ approach. This means that there will always be a place for professional translators. Furthermore, the blogger believes that translation crowdsourcing could signal the arrival of new opportunities for professionals including new partnerships, improved translator profiles, more focus on translator specialisation, and higher human translation quality in general.

Munro (2010 n.p.) outlines how collaboration between professionals and non-professional translators working with the text-message based emergency reporting system, following the earthquake in Haiti, “exceeded that of any one individual.” Volunteer translators translated, categorised and geolocated over 40,000 text messages in real-time within the first six weeks. The outcome of the efforts shows that “collaboration among translators was crucial for data-quality, motivation and community contacts.” This example from Haiti demonstrates the potential power of collaboration in crisis situations. But, it could also serve as a template for how collaboration between professionals and non-professionals could be a success.

4.2. Translation crowdsourcing as a negative activity

4.2.1. Free Labour

Sixteen of the bloggers discuss how they do not work for free, unless they decide to do so. Love German Books (5 September 2013) argues that “if someone is going to make money out of the thing they want me to give them for free (e.g. translation) I won’t work for free”. She discusses how some business models lack credibility, as they expect employees to work for free. These comments are clearly directed at the many for-profit companies who engage in translation crowdsourcing. A widely commented case was that of LinkedIn.

In 2009, LinkedIn, a social network platform for professionals, asked its members to complete a survey about their interest in translating the LinkedIn platform via translation crowdsourcing. One question asked to indicate the type of incentive participants would expect, which included ‘because it’s fun’, ‘translation leader board’, ‘upgraded LinkedIn account’, and ‘other’. Of the approximately 12,000 LinkedIn members who responded (with about 50% being professional translators), the most popular incentive was ‘upgraded LinkedIn account’ (5,054), with ‘other’ incentive coming in fourth place (3,280). The majority of these ‘other’ respondents requested monetary compensation (Posner 2009). Many professional translators reacted negatively to this request: a LinkedIn group ‘Translators against crowdsourcing by commercial businesses’ was set up, and Matthew Bennett, a professional translator, wrote a blog post on the topic ‘LinkedIn infuriates professional translators: 10 big questions’. Bennett’s blog post received several encouraging comments from fellow professional translators who were insulted by the request by a for-profit company to work for free. Many professional translators active on the microblogging site Twitter discussed the incident using the hashtag #LinkedInfail. A New York Times (NYT)article (Newman 2009) discussed LinkedIn’s attempt to follow in the footsteps of other social media platforms such as Facebook to translate their site via crowdsourcing.

Seven blog posts in this data set mention the LinkedIn incident. The Masked Translator (30 June 2009) is offended by LinkedIn “both for their presumptuousness and for their cluelessness.” The blogger is also offended by comments from a translator in the NYT article claiming she “didn’t feel cheapened or exploited” and instead considered it a great opportunity. The Masked Translator highlights that this comment only reinforces clients’ views that professional translators should lower their rates or even work for free. Musings from an overworked translator (16 June 2009) writes that she is “not as upset as some” and as she does not pay to be on LinkedIn, she does not expect anything from them. But she is concerned that LinkedIn treated the professional translators as non-professionals, and wonders if “accounting services or PR folks” have also been asked to work for free.

Common Sense Advisory (CSA), an independent language-industry market research company, claims the prospect of ‘free labour’ cited by many professional translators is not the motivation for companies engaging in translation crowdsourcing initiatives (Ray and Kelly 2011). Instead, they do so for reasons of speed, quality and reach (Kelly 2009b). However, bloggers in this study do not always trust these industry reports (see for example, Financial Translator 28 November 2011, 28 May 2012, 8 June 2011; Translation Tribulations 10 July 2012), since it is not always clear to translators how independent the role of the CSA is within the translation industry.

4.2.2. Ethics

Altogether 25 blog posts are thematically linked to ethics. Only two bloggers explicitly discuss the ethics of translation crowdsourcing (5 posts), but latent ethical concerns were identified in the remaining 20. These concerns relate to initiatives by both non-profit and for-profit organisations. Translation Tribulations (2 December 2013) wants to highlight “the creeping deprofessionalization and demonetization promoted by Translators Without Borders (TWB) or similar programs and their well-paid corporate advocates”. He follows up this topic with three more posts (25 October 2014, 29 October 2014, 5 November 2014) discussing possible conflicts of interest for the board members of TWB and the types of projects they undertake, since many members of the board are heavily involved in the translation industry. The then TWB president Lori Thicke (resigned November 2014) and other TWB board members responded to the posts in the comments section, and a heated debate ensued between the blogger, TWB board members and several professional translators. The comments illustrate that many professional translators are in favour of volunteering with non-profit translation crowdsourcing projects, but for some, the apparent conflicts of interest are a concern in relation to obtaining professional translators’ services for free. This point was previously touched upon by Baker (2006: 159), who considered TWB’s humanitarian and commercial agendas conflicting since TWB is an “offshoot of a commercial translation agency.”

Naked Translations (15 January 2010) mentions how she deleted her LinkedIn profile due to ethical reasons. When asked in the comments section of her blog to expand on this point, she argues that the email sent by a for-profit company “whose aim is supposedly to help professionals advance their career showed such a lack of respect for [the translation] profession.” Therefore, she feels uncomfortable using a site “that clearly doesn’t have my best interests at heart.”

Your Professional Translator (1 August 2014) lists crowdsourcing projects translators could volunteer with, and includes Twitter in the list. Two commenters question the reason (and ethical implications) for including this for-profit company. The first commenter is unaware Twitter conducts translation crowdsourcing, but assumes Twitter employs professional translators to review the translations before publishing them. The blogger agrees from an ethical viewpoint that Twitter should do this. Still, she considers the Twitter translation crowdsourcing project as a great way for new translation graduates to promote themselves. In response to the second commenter, the blogger argues that Twitter is included as not all translators want to donate their professional skills to NGOs, it gives translators an opportunity to volunteer and earn recognition, and it will help to further a translator’s career. She adds that this opportunity is probably not of interest to experienced translators. Based on the two comments here, and previous comments from bloggers regarding the LinkedIn crowdsourcing request, some professional translators might consider this an example of what McDonough Dolmaya (2011b: 103) describes as exploitation through various marketing mechanisms. Through translation crowdsourcing Twitter benefits from gaining access to users outside of the current community, and in doing so, increases revenue via advertising.

McDonough Dolmaya (2011b: 106) argues that ethical considerations of translation crowdsourcing projects depend not only on the status of the organisation (for-profit/non-profit), but also on how the project is organised, and how the project is described to the public. Translation Tribulations emphasises in his posts the need for companies and communities to be transparent about their activities, as a failure to explain fully their motivation and the organisation of the translation crowdsourcing may result in an angry and divided rather than a close-knit community.

Transparency of the translation process is linked to the visibility of the translator. Even though many for-profits employ professional translators to check the quality of the ‘free translations’, this is not made clear to the users of the translations. Therefore, while the visibility of translation and translators is increased among the general public, there is also an increased perception that translation can be done by anyone who has knowledge of two languages, to become a translator requires little formal training, and it could be considered more suitable as a hobby rather than a profession (McDonough Dolmaya 2011b: 104). Furthermore, those participating in translation crowdsourcing initiatives do not adhere to (or perhaps are unaware of) a professional code of ethics. Many codes of ethics prohibit translators from carrying out several tasks that are considered an integral part of a translation crowdsourcing project, for example accepting work for which the translator does not have the required competences, and some codes stipulate that members must master the target language like a native speaker, if it is their L2 (McDonough Dolmaya 2011b: 104) (cf. Hunziker Heeb 2016 in this volume).

Ethical considerations of crowdsourcing are identified as a dominant pattern within the blog posts in this study, and they highlight a number of issues that warrant further research. One of these considerations concerns translator responsibility (La parole exportée, 13 January 2008). In crowdsourcing, the responsibility for the translation moves from the professional translators to the amateurs. This could have an effect “on the way translation is viewed, produced and received by Internet users, corporations, and translators themselves” (McDonough Dolmaya 2011b: 107).

4.2.3. Quality

Translation quality is a topic discussed by 18 bloggers. The main point across all posts is that the quality of translations produced via translation crowdsourcing is compromised, due to amateurs carrying out the work coupled with the low payment or free model used. Thoughts on Translation (15 February 2008) believes that Facebook should have hired professional translators to translate its social network platform to achieve high quality translation. She also agrees with comments made by La parole exportée in her own blog post on the topic of Facebook’s translation (12 February 2008), when she claims the “translation of social media tools is even more complex than, say, technical translation, because of the need to convey underlying meanings, slang, cultural connotations etc.” These two bloggers wrote their posts in reference to a video of two professional translators giving their opinions on the English-Spanish Facebook translations (Owyang 2008), pointing out incorrect or sloppy translations from English into Spanish. In the comments section following the video, numerous self-identified professional translators and language specialists continue the discussion, with one translator asking “can’t Facebook afford professional translators or don’t they take non-English speakers seriously enough to think they should spend some money on them?” However, other self-identified professional translators disagree with the opinion of the two professional Spanish translators. In the comments section one highlights that “many of the terms to be translated were new concepts that needed new words, so you can't really use the same norms when judging how the translation was done.”

Financial Translator wrote six posts discussing the topic of low quality obtained via translation crowdsourcing. He states that “translation is something that can be done by any bilingual, with differing levels of success. Professional translation, in contrast, is the product of thought applied to the everyday task of translation” (13 August 2012). He provides examples of low quality translations from Smartlings’ website (a cloud marketplace) and from some of their clients’ websites, and challenges Smartlings’ claim that they employ professional translators (13 August 2012). In another post (6 June 2012), he discusses how Pinterest, a social media platform, published a blog post in Spanish to announce their plans to translate content into Spanish using crowdsourcing. Many Spanish translators (including himself) took to Twitter to discuss the low quality of the Spanish post, which had “faulty punctuation”, “stilted text” and “wooden style”. A Pinterest employee responded, asking what was wrong with the Spanish text. Financial Translator explained that based on the quality of the text, he believed crowdsourcing had been used, tweeting “social media synonymous with low quality”. The employee claimed that the post had been translated by professional translators, but during the exchange on Twitter it transpired that the post had been translated by the employee with the help of her bilingual mother.

Translate This! (13 May 2013) discusses how the Internet is a useful resource for professional translators, but warns that they must “exercise great caution to separate the useful from the ridiculous”. Many web-based companies resort to translation crowdsourcing, with the aim of obtaining high-quality translations for free. This blogger deems this practice “laughable”, claiming that the translation task might be done faster, but not better than if done by professionals. To support his claims, he provides an example of a German text and argues that even though any German speaker could deduce what is meant by the text, this does not mean the translation is acceptable.

Although professional translators worry about the lack of quality within translation crowdsourcing projects, in some of them quality issues are taken very seriously. Facebook for example has been very transparent about the steps it took to implement quality control in its translations (Kelly et al. 2011: 86, Wong 2008). Their quality model relies on two basic components: (1) the votes of a community of users on translations proposed by the members of that same community, and (2) an overview of the entire cycle by professional translators (Jiménez-Crespo 2011: 135). The voting system, coupled with the role of the hired professional translators is in line with the ISO 900 quality standard, which requires “a product or service to satisfy stated or implied needs” (Ørsted 2001:443). Some of these checks are similar to those implemented in the professional TEP model, including the use of professional reviewers, approved translations for technical terms, and automatic checks to ensure the style sheet is being adhered to.

4.2.4. Authority

It is clear that the blog posts in these data focus on translators’ concerns regarding quality and crowdsourcing. However, these calls for quality control could also be interpreted as relating to translator authority. The translators have deemed themselves the most qualified to translate content from the domain of new media, but the companies engaging in translation crowdsourcing do not necessarily agree. Furthermore, many companies are motivated to use translation crowdsourcing as a way to engage their users and to interact with them, which relates to the “principles of sharing, openness and collaboration associated with Web 2.0” (Gough 2011: 195). This motivation is not mentioned by the bloggers. Some posts discussed above under quality claim that non-professionals cannot translate social media platforms to the same high quality as professionals could do. The real issue, however, may be that professional translators believe their authority is being undermined since many high-profile companies seeking to have their platform translated into numerous languages do not use professional translators as the visible translators in the process.

Professional competence is a key requirement within any professional domain, and this means “acquiring the expertise and thus the authority to make professional decisions” (Kiraly 2000: 1). Kiraly (2000: 13) maintains that professional translators should have translation competence (to produce an acceptable target text) and translator competence (to communicate successfully within expert communities considering both language and culture). Many bloggers discuss activities relating to translation crowdsourcing in negative terms, as these activities can be seen as undermining their professional competence, and thus their authority, in both areas.

Twelve bloggers describe the low rates offered by LSPs and cloud marketplaces, commonly referred to as bulk translation agencies. Some bloggers are clearly irritated by the publicity cloud marketplaces receive, often being lauded as the new face of translation in the media, with most of the focus being on the technology rather than the human translators (e.g. Patenttranslator 27 March 2014, 3 April 2014). Other bloggers argue, both implicitly and explicitly, that translators should present themselves as a powerful authority who has control over work and pay conditions. The bloggers therefore argue that professional translators should use their authority to distance themselves altogether from the part of the market occupied by bulk translation agencies (see also Dam 2013: 23).

5. Conclusions and future research

This paper aimed to identify professional translators’ attitudes towards the practice of translation crowdsourcing through a study of professional translators’ blogs. From a convenience sample of 48 blogs (64 posts), seven themes were identified. Reflecting on the title of this paper, the overall low number of blog posts addressing translation crowdsourcing concerns could certainly be an indicator that professional translators do not consider this phenomenon as a significant cause for concern. Yet, this study has also highlighted some concerns of professional translators that are worthy of reflection.

On the one hand, translation crowdsourcing initiatives can enhance the visibility of translation and translators and demonstrate the value of translation to society (see also McDonough Dolmaya 2011b). On the other hand, specifically for-profit translation crowdsourcing initiatives do little to enhance the public’s perception of the skills, training and expertise involved in the translation process (ibid), with professional translators remaining hidden in the background, while the non-professional translators are publicly visible.

The bloggers often discussed the topic of translators’ participation in non-profit initiatives, and several posts were dedicated solely to describe the work being done by the non-profit organisations and how translators could contribute to the various projects. Within this data set, many of the same non-profit organisations were mentioned, and noticeably there was no mention of any non-profit translation crowdsourcing FOSS projects. McDonough Dolmaya (2012: 176) found in her study that survey respondents had a preference for non-profit humanitarian and for-profit projects over non-profit FOSS projects. While the participants in these two studies differ (9 out of 75 participants in McDonough Dolmaya’s study work as professional translators, interpreters or localisers vs. 48 professional translators), it would be worthwhile to investigate why some non-profit projects are prioritised over others.

Regarding the ethical aspects of translation crowdsourcing, two concerns were identified: first, some professional translators have spoken out against a non-profit organisation conducting translation crowdsourcing because of the apparent conflict between the organisation’s involvement in projects that could benefit others financially, while hiring professional translators who work for free. It is currently not clear how divided the translator community is regarding professionals translating for free, even when it is for a non-profit organisation, and this would be an interesting avenue to explore further; second, the topic of professional responsibility was raised by one blogger. If non-professionals are translating for companies for free, the blogger questioned whether these non-professionals also take responsibility for the translations, and what the implications of this shift of responsibility might mean for the translation profession, including professionals and non-professionals. Further research on this topic could shed light on how non-profit and for-profit companies value professional responsibility and how they implement it in their organisations.

Translation crowdsourcing is here to stay for the foreseeable future. What was once perhaps considered “a dilettante, anti-professional movement” (O’Hagan 2011: 11) located on the periphery of the translation profession is clearly occupying a more central position. Based on these data, professional translators involved in non-profit translation crowdsourcing are actively blogging about their experiences and encouraging other professionals to become involved. Those professionals who participate through for-profit initiatives (e.g. quality control, project management, etc.) perhaps do not blog, or are not blogging about their participation. We cannot predict with certainty the final position of translation crowdsourcing within the translation profession. Nonetheless, the discourse among professional translators in this study suggests that professional translators and translation crowdsourcing can coexist without being in competition. This might very well be within different translation markets (e.g. bulk vs. premium), and this decision is one the professional translators need to be involved in making.

Acknowledgements

The researcher would like to sincerely thank Dr Carmen Heine for coding the German language post and for providing valuable feedback on an earlier version of this paper. Sincere thanks also go to Kristina Brun Madsen for coding the English language posts. The researcher would also like to thank the editors of this Special Issue for their invaluable feedback on earlier versions of this paper.

Bibliography

• Baker, Mona (2006). Translation and conflict. A narrative account. London/New York: Routledge.
• Boyatzis, Richard Eleftherios  (1998). Transforming Qualitative Information:thematic analysis and code development. Thousand Oaks: Sage Publications.
• Braun, Virginia and Victoria Clarke (2006). “Using thematic analysis in psychology.” Qualitative Research in Psychology 3(2), 77–101.
• Burr, Vivian (1995). An introduction to social constructionism. London/New York: Routledge.
• Creswell, John W. (2014). Research design: qualitative, quantitative & mixed methods approaches. Thousand Oaks:  Sage Publications.
• Dam, Helle Vrønning (2013). “The Translator Approach in Translation Studies: Reflections based on a study of translators' weblogs.” Maria Eronen and Marinella Rodi-Risberg (2013). Point of View as Challenge: VAKKI Publications 2. Vaasa: Universitetet i Vaasa, 16–35.
• Dam, Helle Vrønning and Karen Korning Zethsen (2008). “Translator status: A study of Danish company translators.” The Translator 14(1), 71–96.
• Dombek, Magdalena (2014). A study into the motivations of internet users contributing to translation crowdsourcing: the case of Polish Facebook user-translators. PhD thesis. Dublin City University. http://doras.dcu.ie/19774/1/PhD_Thesis_M_Dombek_Final_Corrected_Version.pdf (consulted 10.05.2015).
• Efimova, Lilia (2009) Passion at work: blogging practices of knowledge workers. PhD Thesis. University of Utrecht.
• European Commission (2012). Studies on translation and multilingualism: crowdsourcing translation. http://bookshop.europa.eu/en/crowdsourcing-translation-pbHC3112733/ (consulted 02.05.2015).
• Garcia, Ignacio (2009). “Beyond Translation Memory: Computers and the Professional Translator.” The Journal of Specialised Translation 12, 199–214.
• — (2015). “Cloud marketplaces: Procurement of translators in the age of social media.” The Journal of Specialised Translation 23, 18–38.
• Gough, Joanna (2011). “An empirical study of professional translators’ attitudes, use and awareness of Web 2.0 technologies, and implications for the adoption of emerging technologies and trends.” O’Hagan, Minako (ed.) (2011). Translation as a social activity. A special issue of Linguistica Antverpiensia 10, 195–225.
• Hemetsberger, Andrea (2013). “Crowdsourcing.” Russell Belk and Rosa Llamas (eds) (2013). The Routledge Companion to Digital Consumption. Oxon: Routledge, 159–170.
• Herring, Susan C., Lois Ann Scheidt, Sabrina Bonus and Elijah Wright (2004). “Bridging the gap: A genre analysis of weblogs.” Paper presented at the 37th Hawaii InternationalConference on System Sciences. IEEE Computer Society.
• Hosaka, Tomoko (2008). “Facebook asks users to translate for free.” NBC News, April 18. http://www.nbcnews.com/id/24205912/ (consulted 08.04.2015).
• Howe, Jeff (2006). “The rise of crowdsourcing.” Wired 14.06. http://archive.wired.com/wired/archive/14.06/crowds.html (consulted 02.05.2015).
• Hunziker Heeb, Andrea (2016). “Professional translators’ self-concepts and directionality: indications from translation process research.” The Journal of Specialised Translation 25, 74–88.
• Jiménez-Crespo, Miguel (2011). “From many one: novel approaches to translation quality in a social network era.” O’Hagan, Minako (ed.) (2011). Translation as a social activity. A special issue of Linguistica Antverpiensia 10, 131–152.
• Jääskeläinen, Riitta, Pekka Kujamäki and Jukka Mäkisalo (2011). “Towards professionalism – or against it? Dealing with the changing world in translation research and translator education.” Across Languages and Cultures 12(2), 143–156.
• Kelly, Nataly (2009a). “Myths about crowdsourced translation.” Multilingual 20(8), 62–63.
• — (2009b). “Freelance translators clash with LinkedIn over Crowdsourced translation”. Common Sense Advisory Blogs http://www.commonsenseadvisory.com/Default.aspx?Contenttype=ArticleDetAD&tabID=63&Aid=591&moduleId=391 (consulted 10.03. 2015).
• Kelly, Nataly, Rebecca Ray and Donald A. DePalma (2011). “From Crawling to Sprinting: Community Translation Goes Mainstream.” O’Hagan, Minako (ed.) (2011). Translation as a social activity. A special issue of Linguistica Antverpiensia 10, 75–94.
• Kiraly, Don (2000). A social constructivist approach to translator education: empowerment from theory to practice. Manchester: St. Jerome.
• Koponen, Marit (2016) “Is machine translation post-editing worth the effort? A survey of research into post-editing and effort.” The Journal of Specialised Translation 25, 131–148.
• Lexiophiles (2014). http://www.lexiophiles.com/top-language-blogs-and-podcasts (consulted 12.03.2015).
• McDonough Dolmaya, Julie (2011a). “A window into the profession: what translation blogs have to offer translation studies.” The Translator 17(1), 77–104.
• — (2011b). “The ethics of crowdsourcing.”O’Hagan, Minako (ed.) (2011). Translation as a social activity. A special issue of Linguistica Antverpiensia 10, 97–110.
• — (2012). “Analysing the crowdsourcing model and its impact on public perceptions of translation.” The Translator 18(2), 167–191.
• Mesipuu, Marit (2012). “Translation crowdsourcing and user-translator motivation at Facebook and Skype.” Translation Spaces 1, 33–53.
• Munro, Robert (2010). “Crowdsourced translation for emergency response in Haiti: the global collaboration of local knowledge.” Paper presented at the Association of Machine Translation for the Americas Conference AMTA, 2010. http://amta2010.amtaweb.org/AMTA/papers/7-01-01-Munro.pdf (consulted 19.03.2015).
• Newman, Andrew Adam (2009). “Translators wanted at LinkedIn. The pay? $0 an hour.” The New York Times, June 28. http://www.nytimes.com/2009/06/29/technology/start-ups/29linkedin.html (consulted 08.04.2015).
• O’Brien, Sharon (2011). “Collaborative translation.” Yves Gambier and Luc van Doorslaer (eds) (2011). Handbook of Translation Studies. Vol. 2. Amsterdam/Philadelphia: John Benjamins , 17–20.
• O’Hagan, Minako (2009). “Evolution of user-generated translation.” Journal of Internationalisation and Localisation 1, 94–121.
• — (2011). “Introduction: Community Translation: translation as a social activity and its possible consequences in the advent of Web 2.0 and beyond.” O’Hagan, Minako (ed.) (2011). Translation as a social activity. A special issue of Linguistica Antverpiensia 10, 11-23.
• — (2013). “Impact of new technologies on translation studies: a technological turn?” Carmen Millán and Francesca Bartrina (eds) (2013). The Routledge Handbook of Translation Studies. Oxon: Routledge, 503–518.
• Owyang, Jeremiah (12 February 2008) [blog] http://www.web-strategist.com/blog/2008/02/12/video-facebooks-spanish-translation-misses-the-mark-420/ (consulted 04.02.2015).
• Ørsted, Jeannette (2001). “Quality and efficiency: Incompatible elements in Translation Practice?” Meta: Translators' Journal 46(2), 438–447.
• Perrino, Saverio (2009). “User-generated translation: The future of translation in a Web 2.0 environment.” The Journal of Specialised Translation 12, 55–78. http://www.jostrans.org/issue12/art_perrino.php (consulted 12.03.15).
• Posner, Nico (2009). “Translating LinkedIn into many languages.” http://blog.linkedin.com/2009/06/19/nico-posner-translating-linkedin-into-many-languages/ (consulted 11.04.2015).
• Pérez-González, Luis and Şebnem Susam-Saraeva (2012). “Non-professionals translation and interpreting.” The Translator 18(2), 149–165.
• Pym, Anthony (2011). “Training translators.” Kirsten Malmkjaer and Kevin Windle (eds) (2011). The Oxford Handbook of Translation Studies. Oxford: Oxford University Press, 475–489.
• Pym, Anthony, David Orrego-Carmona and Esther Torres-Simón (2016). “Status and technology in the professionalisation of translators. Market disorder and the return of hierarchy.” The Journal of Specialised Translation 25, 33–53.
• Ray, Rebecca and Nataly Kelly (2011). Crowdsourced Translation: Best practices for implementation. Lowell, Mass.: Common Sense Advisory.
• Risku, Hanna, Regina Rogl and Christina Pein-Weber (2016). “Mutual dependencies: centrality in translation networks.” The Journal of Specialised Translation 25, 232-253.
• Sanford, Matt (2011). “Crowdsourcing vs. Community.” http://mzsanford.com/blog/crowdsourcing-vs-community/index.html (consulted 12.02.2015).
• Trammell, Kaye D. and Ana Keshelashvili (2005). “Examining the new influencers: A self-presentation study of A-list blogs.” Journalism and Mass Communication Quarterly 82(4), 968–982.
• Vashee, Kirti (6 November 2009). “Crowdsourcing and translation.” http://www.nakedtranslations.com/en/2009/crowdsourcing-and-translation (consulted 10.03.2015).
• Wong, Yishan (2008) “Facebook around the world.” Facebook, February 13. https://www.facebook.com/notes/facebook/facebook-around-the-world/10056937130 (consulted 10.04.2015).

Biography

Marian Flanagan is an Assistant Professor at the Department of Business Communication, Aarhus School of Business and Social Sciences, Aarhus University, Denmark, where she teaches a range of topics including web communication & social media, web localisation, translation technology, professional writing skills and English language courses. She is a member of the Translation and Interpreting research group. Her research areas include translation technology, localisation, translation crowdsourcing, writing processes, and search engine optimisation.
E-mail: marfl@bcom.au.dk

 

Appendix A

Endnotes

Note 1:
Garcia specifically refers to raw machine translation here, to highlight that some companies use the raw output without any post-editing by humans afterwards (cf. Koponen 2016 in this volume).
Return to this point in the text

Home | Past Issues | Keyword Search | Editorial Board | Events | About | Contact | Help