After having given you some basic information on the two leading CAT tools in my experience, I’d like to give you some information on one of the dinosaurs of CAT software: STAR Transit.
Developed By STAR Group from 1986 as a 32-bit DOS program, Transit uses a different system than other translation software to manage translation units. Instead of a translation memory of segments with metadata regarding context, Transit’s translation memory engine bases its searches on bilingual reference files.Transit offers a seemingly old-fashioned interface, with few of the modern bells and whistles that define Trados and memoQ: few add-ons, a very poor start to MT, in my language pairs, at least. It does have its advantages, though. Because the engine works in a corpus of bilingual files, it is very flexible and can be used either way in the one language pair, and even offer live-updated matches from another language pair if the target language is the same.
For those who’ve never worked with it, let me give you an example. Let’s say your client sends you a package prepared with Transit NXT (the current version) in the language pair XX to YY. You have already translated many projects for this client from WW to YY. Transit NXT can, as you type your translation in the target window, pull up previous translations from the WW to YY projects previously translated that are close to what you’re currently typing. It will show you fuzzy matches from the reference files in the pair XX to YY, and also fuzzy target segments from any other files you add as a reference with YY as a source or target language. These side-by-side fuzzy matches for source and target languages become more and more useful the more you grow your stock of reference translations. You can also call up the full context of any past translated segment in a separate tab, which isn’t possible with software working with a memory of individual translation units, and correct or update these reference files in context.
The very simple interface reduces lagging to a minimum, even with giant files and references. With my largest client on Transit, for whom I have translations dating back to 2012 (tens of thousands of reference files), I have yet to experience any lags, and have worked with translatable files up to several hundred thousand words long.Transit NXT also allows work in a variety of roles, from translator to terminology expert, and formats, from simple translation, to DTP and subtitling.
Transit works on a 3, 6, 12 month licence base, which can be costly. Certain agencies will allow you to take advantage of one of their floating licences to work with them. As a freelancer, two options are available, Freelance and Freelance Pro, the difference being that the cheaper version is limited to three terminology dictionaries, which some clients (including mine) exceed, and the Freelance Pro version offers compatibility with following package formats: memoQ, XLIFF, COTI, SDL WorldServer and Trados Studio and multiple file formats.
My personal opinion? I wouldn’t have used Star Transit if my clients hadn’t insisted on it. But now I’ve gotten used to it, there are many functions and advantages I wish could be transferred to my favourite tools, the most important ones being its speed, the context references and the ability to mark segments as translated after having created a reference exception so they aren’t automatically applied to the next repetition or pre-translated in the future.
What about you? Do you work with this seemingly outdated but surprisingly efficient tool?Are there functions of Transit I’ve missed and that you love?
Does anyone even work with this software if their client doesn’t expect it from them?
What do you think?
Comments