Terminology Xtreme Part 1: Beyond MultiTerm 101

Keeping Termbases Healthy

As the localization maturity of companies and organizations rises, so does the number of professionally implemented terminology workflows and centralized termbases. Many of them, we are proud to say, implement our workflow solution quickTerm based on SDL MultiTerm technology (and thus also known as SDL MultiTerm Workflow).

No doubt: Getting your terminology project off the ground was a major and very notable accomplishment. However, once terminology is firmly established in the standard process landscape of a company, an organization, or an LSP, the new levels of sophistication lead to new challenges: Data integrity and quality tracking, batch changing, or transforming of termbase content, dissemination of terminology content into other tools, and, in general, new areas of application for terminology. In order to tackle these challenges, we have developed Excelling MultiTerm and Publishing MultiTerm. Actually, we have just released brand-new versions of both problem solvers. And, by the way, Publishing MultiTerm and Excelling SDL MultiTerm are part of our much larger suite of “tools for language workers”: the expertTools.

Data Integrity and Quality

Just like translation memories or any large content repositories, termbases also get “dirty” after a while:

• Many export and import loops generate inconsistencies in the data, for instance, fields, picklist values, or languages that do not match the termbase definition
• Changes in picklists, database structures, and slight glitches in the migration processes lead to almost unnoticeable but problematic “inconsistent” data
• Annoying details like broken links surface more and more often
• Invalid duplicates that have been entered erroneously but are hard to get rid of make our terminology lives difficult

In our termbase management solution Excelling MultiTerm, we have implemented a great number of consistency checks that enable you to address all these issues and even fix them automatically. So, for instance, if you have German (DE) defined as your language in the termbase model, but some records contain Austrian German (DE-AT), you can automatically map this to German (DE) and add a regional usage label “Austrian” to it. Or you can automatically detect and ideally even fix broken links between entries, terms, or to external sources.

With Excelling MultiTerm, managing your data is easy. For example, batch-deletion can be useful if

 you want to undo an import or to remove a complete language in one step. Another use case is to re-sort all terms in all entries, moving the “preferred” terms to the top of each language.

Batch Editing Terminology

If you have reached a certain maturity in your termbase content, you are bound to run into issues that require batch processing of data. Maybe you need to move a field from term level to language level or you want to batch-replace certain data, run spellchecks etc. One elegant way to master these tasks is to use Excelling MultiTerm that now even supports batch-exporting of Multimedia content. This is extremely useful if, e.g., you want to shrink all images in a termbase at once.

The very flexible and powerful filtering mechanisms of Excelling MultiTerm enable you to filter content down to exactly what you need. These filtering mechanisms let you combine several MultiTerm filters, add term-level filtering, and even programmatic filtering with XPath or externally developed code. These terms are then exported and inserted into MS Excel directly via the API. Due to the non-linear structure of terminology data, this is easier said than done, but we have devised a data structure in Excel that supports the unpredictable and extremely flexible content you might have in your termbase. 

In MS Excel, you can manipulate the termbase content in a structured and convenient way, using all of Excel´s standard features like search/replace, moving entire columns, spellchecking etc.

Getting the changed data back

After the editing is done, you can synchronize the changed data back to MultiTerm. This again sounds easier than it actually is because Excelling MultiTerm has to recognize the difference between new terms and changed terms during the import. Due to our sophisticated stamping mechanism, Excelling MultiTerm is able to detect changes and apply them to the termbase. The standard MultiTerm import mechanism would not be able to tell the difference between a “new” and a “changed” term.

Excelling MultiTerm even enables you to delete terms, languages, or entries in MultiTerm directly from the Excel file, or to select exactly which fields should be read back and which ones should be left alone in MultiTerm. And, of course, Excelling MultiTerm allows you to generate new entries, terms, languages etc. So it can be used to import carefully prepared data in complex structures, or add new languages, for instance.

Disseminating Terminology

Once your termbase has reached a certain size, other applications will want to leverage this data. Software developers want to use it to auto-correct wrong string input. Authoring systems or CMSs want to synchronize it but map it to a totally different logical structure. Even static outputs such as PDF glossaries are sometimes required.

With Publishing MultiTerm, you can filter exactly the content you want to export, and then transform it over and over again, until you reach the required target format in XML, CSV, PDF etc. Once such a conversion task has been defined, you can automatically run it in defined intervals to make sure your target systems are always up to date.

Any Questions?

Interested? We are always happy to answer your questions.

There’s also a free trial of Excelling MultiTerm at SDL OpenExchange!

Or do you have a special request that we have not yet covered in our toolset? Let us know, we´ll see what we can do for you!