Anyone
who is responsible for a company’s technical documents - operator’s manuals,
service manuals, etc. - will be aware that what the end user wants is perfectly
consistent vocabulary, to avoid any
ambiguity. If it said “retaining bolt” in the previous page, why does it
say “retaining screw” on this page? It is the same thing?
Recent
years have seen great progress in the standardisation of vocabulary and
phrasing in technical documents, and a great deal of that progress is due to
the increasing use of translation memories.
You
may have tried out some of the “on-line translation” systems available on the
Internet. You will have seen that while it is often possible to understand more
or less what the original text wanted to say, the language employed is usually
not acceptable. You only have to translate a short text into a foreign language
and then translate the result back into English to see what sort of a mess the
translation is to the foreign reader.
Translation
memory has nothing to do with that sort of “on-line translation”. Translation
memories are used to record accurate translations made by qualified
translators, so that the next time the same words are used in the original
text, the translation memory will automatically produce the same translation.
This provides consistency of vocabulary and phrasing.
Translation
memories are used by translation agencies and freelancers for two purposes:
1. To save time in translation, since the translator
does not have to type the sentence again.
This saves money.
Like
most service companies, translation agencies are selling time – the time that
their translators take in typing out and checking a translation.
2. To ensure consistency, particularly in
repeated sentences. This gives the reader confidence in the document concerned,
since he or she sees identical phrases and vocabulary in various parts of the
document. It also ensures consistency of vocabulary, so that no confusion can
exist as to “Is this the same thing they were talking about on page 33 ?”.
When
various companies started to produce translation memories in the 1990’s, they
were mainly used by freelance translators to save time and to produce
standardised translations.
Translation
agencies, almost all of which use freelance translators when their in-house
staff have too much work, then started to use translation memory systems for
the same purposes. The cost of translation was sometimes reduced, since three
categories of translation were used:
· Perfect match (when the sentence to be translated already existed in the translation memory – no action on the part of the translator – the computer translates the sentence automatically) This category is often paid at 20% or 30% of “per word” price.
· Fuzzy match (when the sentence to be translated is almost the same as a sentence in the translation memory; in this case the translator has to check it for differences). This is often paid at 50% or 60% of “per-word” price.
· No match (when there is no such sentence in the translation memory). This is paid at 100% of “per-word” price.
Among
some of the bigger companies in the world it then became usual to work with
translation agencies who kept such memories on their behalf.
Translation
memories are now often used by the translation agencies for their medium and
small customers and the advantages as regards consistency are immediately
apparent to anybody reading the technical documentation.
Companies
working with translation agencies who use these systems also often enjoy lower
prices for their translations.
No comments:
Post a Comment