3 Key CCMS Shortcomings Related to Localization — And How to Combat Them
Click here to close
Click here to close
Subscribe here

3 Key CCMS Shortcomings Related to Localization — And How to Combat Them

3 Key CCMS Shortcomings Related to Localization — And How to Combat Them


Last week we discussed the business needs that would drive a technical writing department towards implementing a Component Content Management System (CCMS): decrease time to market, maximize re-use, or single-source “chunks” of content into multiple outputs. While the process may initially be clear and effective for content creation, there are some shortcomings related to localization. In this blog post we discuss these issues and how to overcome them. The challenges are not insurmountable; they simply require strategic foresight and a collaborative approach between content creators and LSPs at every step of the release cycle.

Three key issues stand out:

1. Content is like a puzzle

Authoring in a CCMS can feel like assembling puzzles. It requires a different way of thinking about the content creation process: for example, how to define semantic units of content components and understand their interrelationship and connections.

Then, during the localization phase, a translator has to understand the thinking process that went into creating those specific chunks of content. Both writers and translators have to perform their jobs differently than before. To ensure that all the pieces of the puzzle make sense and will still “work” well together after translation, you:

  • Chunk your content into “puzzle pieces” that represent meaningful semantic units.
  • Train your writers and translators on the concepts of component authoring and localization.

2. Translating collections of scattered puzzle pieces

Translators typically work with XML exports from CCMS systems, which are essentially a collection of scattered puzzle pieces that, when taken individually, don’t flow coherently. The units are not received in the linear, organized order in which they will be published, and the context is lost.

This makes it impossible to understand the big picture — and more challenging to get the translation right. This can also lead to layout and functional issues in the published output, possibly making it necessary to conduct another review post-output.

To alleviate this, you can provide solutions to the context challenges and combat potential quality issues by:

  • Performing pseudo-translation to identify and correct formatting and functionality so that integrity is preserved in localization. XML files intended for translation are “fake translated” and integrated back into the CCMS to publish the “translated” output for debugging purposes. Examining this published output allows the translation provider to adjust settings in the translation memory tool and flush out design and functional issues that need to be addressed, either within the CCMS or within the XML translation process.
  • Providing the big picture with reference materials to get translation right the first time. For example, provide references for translators such as published source-language files, hybrid context documents (which are a mix of already translated and untranslated new/changed text), glossaries and style guides.
    • Published source output. This should be at least one representative published output sample of the new source text (in case you publish the same content to a variety of media output types). Though the translator’s deliverable (XML) may be different, this will give the translator an idea of the big picture into which the translatable components will need to fit.
    • Hybrid published output, containing a mix of previously translated text and new source text. This is especially helpful in an update project, where the XML contains only a few, disjointed components in need of translation. It allows for an in-context view of already-translated text, which will support translation consistency.
  • Allowing time in the schedule to do a post-output check to adjust any layout issues introduced during translation (e.g.: due to text expansion of the translated text).

3. Dealing with consistency issues

Since only new or changed content components are sent for translation, there’s an inherent inability to fix any issues outside of the current translation job. Global terminology fixes and corrections to mistranslations in legacy materials are more prohibitive than they ever were. This puts extra pressure on getting it right the first time and adding additional review layers to ensure consistency. You can best manage consistency challenges by:

  • Writing with translation in mind, which typically means clear, concise sentences with standard grammar and consistent vocabulary.
  • Making sure style guides, translation memories (TM) and glossaries are a) high quality, and b) systematically used on all translation jobs.
  • Adding an in-context review step to focus on the new translated components in at least one representative published output.
  • Defining an effective/low impact process to deal with needed corrections in legacy translations.
  • Adding new product terms to the glossary as they are introduced.
  • Updating the master translation memory (TM) after every translation job.

Reaping the benefits of CCMS on a global scale

These strategies will allow you to stay on top of the challenges of implementing a CCMS and enable you to reap the full benefits of the component content creation paradigm.

Through the use of a CCMS, the mantra “do more with less” can be demonstrated time and time again. Companies are now able to complete source deliverables at the same time as localization occurs, create more product documentation in multiple published output formats, decrease vendor costs and improve time to market — all while maximizing reuse and maintaining the high quality your market expects.

For companies interested in implementing a CCMS, please look for our next blog posts on this topic about features to consider and tips for technical writers.