Spanish-language Facebook

Zia Daniell Wigder | February 08, 2008, 11:54 AM

Yesterday Facebook launched its first non-English language version in Spanish. Though buried within the user's account section (US-based Facebook members must know to click on account, then language, then select “Spanish” from a dropdown menu), users from Spanish-speaking countries will receive the Spanish-language version by default. Facebook claims 1,500 users helped participate in the Spanish translations, while translations for the French- and German-speaking markets are underway.

While it’s not surprising that Facebook chose Spanish as its first language – it’s the third most commonly spoken language by the online population after English and Chinese, and we found it to be the most popular language for US website translations – Facebook’s much publicized decision to rely on a community-based translation model (“crowdsourcing”) is less typical.

Crowdsourcing has garnered much attention in the translation space: Jeff Howe, the Wired reporter who coined the term crowdsourcing, keynoted the Localization World conference last fall. Yet crowdsourcing hasn’t yet been deployed on a wide scale basis. Indeed, the need for a loyal, committed and perhaps most important, skilled base of translators means the crowdsourcing model is going to be effective for only a small number of companies aiming to translate their content.

Facebook presents a confluence of factors that make crowdsourcing a viable model for translation: it has a large and rapidly growing user base (64 million active users), as well as a highly global one (60 percent reside outside the US). Though many of its global users hail from the English-speaking world, it still boasts a substantial number from other regions: almost 3 million from Spanish-speaking countries, for example.

Additionally, Facebook’s users tend to spend a significant amount of time on the site: in December 2007, Hitwise put the average visit on Facebook at 19 minutes, 51 seconds. By contrast, earlier in 2007 Hitwise asserted that the average time spent on broadcast media sites was 7 minutes, 5 seconds and on print media sites was 7 minutes, 6 seconds. Facebook users are spending a good amount of time on the site each time they visit, and may therefore be more likely to spend time assisting with translations.

Facebook is not alone in finding success in the crowdsourcing model: dotSUB, an enabler of community-based translations for online video, highlights the success of community-based translations in a corporate video made by Mittal Steel, for example. Yet companies must be realistic about the commitment and skill set of their user base.

Crowdsourcing is topic that’s coming up with some frequency in my conversations with US-based companies and organizations considering translations, particularly in relation to user-generated content. Indeed, for more information on crowdsourcing and UGC (as well as other translation options for companies that do not have Facebook’s user base), stay tuned for our upcoming report on this topic.