I was recently invited to join a few industry colleagues on a webinar sponsored by Memsource, a cloud-based CAT platform provider. As Moravia’s senior project manager, I spoke about the use of cloud-based translation memory and term base tools.
Why do such tools represent a good investment for strong localization programs? And what are the minimum requirements to make any implemented solution a winning one?
At the Speed of Write
The companies that we serve here at Moravia are spread throughout the world. Yes, some concentrate their work in the largest markets in Europe, Asia, and the Americas. And some serve traditionally underserved markets, emerging markets where even finding language services resources is an uphill battle. The real-world project that I discussed in the webinar, for example — a website localization project — is largely localized into 20 languages but includes some parts which are localized into more than 100.
To say that today’s global companies are processing word volumes unheard of before is no exaggeration. Not only is the throughput of the volume in the billions of words each month, the content is more widely diverse than ever before. It’s not just the software strings, but also the related help files, video tutorials, eLearning and traditional training materials, social media posts, and all of the other pieces that, collectively, serve as the brand experience online.
But it is not just the many markets, the high volumes, and the different types of content. It is also the turnaround requirements. Language services providers and internal client localization teams are expected to work at unparalleled speeds to meet consumer content demands. That has of course meant that the content processing flows have changed — no big volume drops in logical, interconnected packages; just a constant flow of single files with their meta information flowing into the system for processing.
All of this means that there is no time to waste on translation management systems and workflows that are sluggish, inflexible, or otherwise unresponsive in meeting global market needs. What’s needed are the boons of automation without additional personnel and unnecessary infrastructure bloat.
What You See Isn’t What You Get
So if these are the conditions under which we are expected to work, what does this mean for the kinds of requirements we would have for the right tools for the job?
- File exchange has to be the top priority. Old methods for file exchange — like FTP, email attachments, or even the emails themselves — these approaches tend to introduce error and miscommunication. Today’s sophisticated translation management systems aren’t off-the-shelf solutions but APIs that allow content flow and related project communication from client, through cloud systems, and on to the language services vendor. And, once processed, it is the interface that allows content and communication to flow in reverse.
- Metadata recognition is important. While the metadata identifies what is to be localized, it also can indicate how, by whom, and when — information that can change over time as the project moves through the production cycle and distribution channels. Effective tools track and share this information and changes.
- Linguistic features are also often required throughout the translation production cycle. Imagine, for example, an online review of an XLIFF file that is exposed to translators via their desktop browsers. Or a similar option for reviewing the translation of image strings. A system that allows for specific features like these help translators meet the project’s quality standards.
- Finally, interconnectivity is vital to cost efficiencies in pre- and post-translation processing as well as in meeting client expectations for visibility. Our own implementation, for example, allows for cloud-connectivity to our financial reporting systems, MT engines, quality assurance systems, and more.
We at Moravia are tool agnostic. More important than a firm belief in any one product’s effectiveness is our belief in discovering tool parts, systems, and processes that can make automation a win for everyone — client, language partner, and project teams. Understanding basic requirements can help you and your selected vendor create the right translation management system for you. (If you’re interested in why, click here to see our video on custom integration and automation solutions.)
Do you have questions or comments about localization engineering? Share them below!