GET A FREE CONSULTATION OR SAMPLE TO GET YOUR PROJECT GOING.

Yes, I want my consultaion

How to Check Terminology in Translations with Verifika

Checking terminology in VerifikaAre you unsatisfied with your translation quality assurance process and looking to improve it?

Read this next article in my series about comparing QA Distiller and Verifika to see whether any of these programs could be a better choice than your current solution. In the previous article, I wrote about a useful option in Verifika that lets me treat only shorter words as whole words instead of the entire glossary. This makes it easier to manage a glossary, because I don’t have to add the different word forms for all the words—only for those shorter ones.

While this is quite useful, it’s not the most important consideration when deciding which program is superior in this respect. It is the overall efficiency of terminology checking that matters. With QAD, this check is extremely efficient, thanks to stem-based checking. I just enter the stems of words as translations, even for word combinations. QAD is able to match just the stems, regardless of the inflections. Without this stem-based check, I would have to enter all the possible word forms as translations. For more information, refer to my detailed post about QAD.

Because Verifika seemed superior to QAD in most other respects, I was hoping that it would be at least as good as QAD in this one, perhaps the most important one. Drumroll…

No, Verifika doesn’t support stem-based checking. It does offer a different approach to this problem, but so far I have been unable to reach the efficiency level of QAD.

Checking glossary terms in Verifika

Before checking, I need to enable the Whole words only option for the target terms. Quite expectedly, Verifika returns an overwhelmingly long list of errors. As I review each error, I have three options:

  1. Simply press Space to acknowledge an error. This is easy, but ineffective, because when I do so, I don’t acknowledge other similar errors and will have to go through each of them manually, too.
  2. When the program identifies a term in the translation successfully, but the word form is different, Verifika suggests that I add this word form to its internal database by pressing Alt+A. When I do so, all similar errors are automatically removed from the error list, so I don’t have to go through each of them individually.
  3. When the program fails to identify a word form, I can add it manually. As soon as I do so and press Ctrl+Tab, the errors are removed from the list, just as if I had pressed Alt+A.

Findings

All right, these are the options. The question is whether this approach is better than stem-based checking in QAD. I have yet to find any benefits. So far, what I have found is this:

  1. First and foremost, my concentration takes a hit. When I’m checking a large translation, I want to fully concentrate on my task as an editor. With Verifika, I can’t do that because I’m also constantly adding word forms. And concentrating more on word forms means concentrating less on potential errors. This is a big distraction for me—one that I can’t really afford.
  2. When Verifika fails to detect a word form, I have to add it manually, which is way too inefficient. Remember that in the same cases, QAD wouldn’t even display an error.
  3. Since I’ve been checking Russian translations, a language which is notorious for its numerous word forms, I have a feeling that adding word forms just never ends. For some terms, my database has already reached as many as 30 word forms. I find myself adding one word form after another where QAD would simply find no errors.

Conclusions

Well, maybe I should just persevere? Hopefully, my database will reach a “breaking point” after I add really a lot of word forms. In theory, yes, it will. In practice, I’m not sure how long it will take me to get there. And there’s yet another challenge. For single words, I do believe I can build a comprehensive database over time, because their number is limited. But what about word combinations? There are hundreds of thousands of them; they’ll just keep coming at me! For each new combination, I’ll have to enter word forms all over again. For example, I might have the following word combination as a term in a glossary:

femoral head

головка бедра

I’ll need to enter 10 or 15 word forms before Verifika reaches the efficiency level of QAD.

But as soon as I add another similar term to the glossary:

femoral neck

шейка бедра

I’ll have to go through the same process all over again. By the same token, I’ll end up adding 10 or 15 word forms for every new term with “femoral”—e.g., “femoral artery,” “femoral nerve,” “femoral hernia”—while in QAD, I’d just add one stem-based translation.

I’m afraid that adding word forms will never end. Because I’ll spend more time on checks and my concentration will decrease, I expect that Verifika will decrease my performance in terms of checking terminology. QAD still seems to be a superior solution.

Disclaimer

Now, my opinion is entirely subjective. First, this is a preliminary judgment based on one month of using the program. It’s highly likely that as I continue playing with the tool and see something that I don’t see now, I might change my mind. Especially because I liked almost everything else about Verifika. Second, as I mentioned, I’ve been working with Russian translations. With languages that have fewer word forms, the performance loss might be negligible.

I’ve now reviewed all the major aspects of both programs. It’s time for the final comparison. Stay tuned for the last article in this series, where I’ll provide a table with pros and cons about each tool.

So what do you think about comparing QAD to Verifika? Do you see any winner for yourself yet?

Add comment


About the Author

Roman Mironov
Roman Mironov
CEO & Founder

As the founder of Velior, Roman has had the privilege of being able to turn his passion for languages into a business. He has over 15 years of experience in the translation industry. Roman has helped dozens of clients increase sales by making their products appealing for speakers of other languages.