Translation Technology Is Not the Solution ... At Least Not Always

Translation Technology Is Not the Solution

There is no conversation about globalization that does not involve technology. And there can be no conversation about globalization that does not involve translation and localization. But in the same way that translation alone is not always the right solution for your globalization projects, some technology choices can be wrong for your globalization efforts.

Indeed, one of the greater challenges of international brands in meeting the needs of their global consumers is in making smart choices about localization automation — choices that should smooth content work flows without locking them into proprietary translation management systems.

Hard Realities vs Wishful Thinking

If you had a chance to read our Information Orchestration Manifesto, you noticed that we believe that the future of successful, large-scale localization lies not with proprietary solutions but with open solutions.

Yeah, yeah, you may reply. Every sales person talks about open solutions. “Plug and play.” Blah. Blah. Blah.

And you are right: there is more talk than results. Starting in the 1990s, we saw LISA attempt some early standards. And industry organizations like TAUS continue to promote interoperability and industry standards. But, yes, we are still far from having systems that are plug and play.

The reality is simply very messy. Systems that appear on the surface as easily connected usually have a lot going on behind the scenes that have to happen for any semblance of a peak performance.

Moreover, in those systems that are not integrated — where the CMS and TMS operate independently — there are still many tasks that are being performed outside of them as well. Email. Skype chats. Multiple teams in which client, vendor, and translator are using whatever tools they usually use to communicate, manipulate data, or store instructions — all of which are not stored and for which there is considerable duplication of tasks.

The reality, that is, loses a lot of time and opportunities for efficiency.

Making Magic Real

Integrating different technologies and automating localization processes can add up to an optimized work flow that actually enables Agile Localization. Yes, truly integrated systems are complex, and require both a lot of engineers and some time to figure out. And a lot of enterprises continue with unconnected tools and overly complex workflows because this is not easy to achieve.

But it is only those enterprises that invest in making the magic happen — winning the speed and responsiveness of translation automation — that enjoy the success of Agile Localization.

What does that investment look like?

It is standards friendly.

Multiple vendors. Continuous localization cycles. A matrix of products, languages, materials, and content types. Recruiting. Quality controls. Invoicing. All of the inputs and outputs that are handled by complex localization systems would be impossible without some manner of standardization. There may not be any major initiative to create a single standard for everything. But standards such as XLIFF as well as localization process standards give us a chance at interoperability.

It is technology agnostic.

We at Moravia do not believe in imposing specific solutions. An investment in a traditional industry tool is an investment of 10 to 20 years — a serious undertaking when market realities are under constant flux and content demands are rapidly evolving. Clients are using Trados, XTM, Memsource, memoQ, and Plunet as well as Drupal, WordPress, and any other number of content creation and content management environments that require localization. Rather than a single choice, we believe, that language services providers and clients should commit to having their systems communicate with each other.

It is solution seeking.

That said, it is not up to our clients to change their processes to deliver content to us, but for us as language services providers to adapt our standards and tools to work as seamlessly as possible with their content and processes. The goal is not, then, the perfect system but for talented localization engineers to make the workings of the systems transparent, user-friendly, content adaptive, and with as few disruptions to the workflows of diverse and dispersed teams as we can to successfully process and distribute content to meet client goals.


Those are my thoughts on the limitations and opportunities of localization technology. What about yours? Share them in the comments section below.

From Good to Great: How to Beef Up Your Localization Program