|
I was privileged last week to be invited to do a webinar for the ATA. (While we're talking about translation associations, the FIT, the umbrella organization for most national language associations, is holding a competition that ends on February 4, organized by my good friend and translation technology giant Alan Melby. All you need to do is send an email to FITtrantech@gmail.com and tell him about your favorite free software that you use professionally as a translator. If you win, you'll be featured on FIT's homepage, where you can also go to find more information about this.)
Anyway, I did a webinar for the ATA on translation environment tools last week, and I actually really enjoyed it (and, judging from participants' comments, most of them did, too) and I wanted to use this space in the newsletter to answer some of the questions that I could not get to during the webinar.
One conundrum that I actually did respond to was this:
I'm in a Catch 22: I don't yet have the clientele that requires TEnTs, so I don't use it yet. However, if I don't get started learning to use different TEnT software, I never will get such clients! Better choose one to get started . . .
I liked this statement because it reveals a misunderstanding that is probably only too common. Translation environment tools are not for your clients. They are designed to give you the environment you need to ease and streamline your translation (well, most are). While it's true that some clients require translation environment tools as sort of an entry pass to begin working for them, this should not be the main motivation for you to work with these tools. Why? Because if it is, you'll focus primarily on the client's needs in using the tool rather than your own. Don't get me wrong: I have nothing against considering the needs and requirements of clients. They're important. But it's also very important to mold your work environment into one that really fits your needs. That's why I think it's important to learn these tools with yourself in mind, and then later adapt it so the client might also benefit (if it ever comes to that.)
Here are a few more related questions:
Would you recommend using tools for all types of translation, or only for technical subjects?
Do you know whether translators in the creative fields -- literature, the fine arts, or even the 'softer' academic fields such as art history -- are turning to professional translation tools for help?
These are good questions, ones that we (I) have often answered far too carelessly. I've often said in the past that translation environment tools are helpful for translators of all functional texts, including technical, medical, legal, financial, and various other "functional" disciplines. Now I'm no longer so sure that this needs to be limited by function. Will the translation memory feature be helpful for translators of academic, literary, or marketing texts? Maybe, but probably not. What about the terminology feature in academic texts? Of course. How about the ability to work in file formats like InDesign, Quark, or even Photoshop -- all pretty important for marketing translators? Absolutely. What about the ability to quickly access a corpus of translation of the author you are translating or even that whole literary genre? In that case, TEnTs might be really helpful for literary translators (who, by the way, could decide to forego sentence segmentation and just segment by paragraphs). I could continue this list, but the point is that it's important not to focus on just one TEnT feature (typically translation memory); instead, become familiar with the many other aspects of a good TEnT and then decide whether it would be helpful to use such a tool.
A number of questioners were concerned about language support of TEnTs. There are essentially two different kinds of tools: those that come without any language data (that's the majority of tools), and those that come with some (I can only think of Fluency and Similis in this category). The ones that come with no language data can be used for a translator of any language combination as long as the operating system supports that language. The other tools can also be used for any language combination, but you naturally might not have the same level of language support as languages with existing materials if your language(s) does not fall into that pool.
Is this all there is to it? Pretty much, but specifically if you are working in a right-to left (bi-directional) language or a language that does not have spaces between words (such as Thai, Lao, Khmer, and Burmese), I would try to talk to translators with the same language and find out what they are using. While all available tools are now based on Unicode, there are still differences in the sophistication with which these languages are handled.
The webinar would not have been complete without two other clusters of questions that are always asked:
Can PDF files be processed?
Can TEnTs run on Macs?
I have written plenty about the first question in the past (and devoted a whole big chapter to the answer in my ebook).
To answer the second question about Mac support: yes and yes. There are a good number of tools that either run natively on a Mac (Wordfast Classic) or through Java (Heartsome, Swordfish, OmegaT). Other tools operate independently because they are completely browser-based (XTM Cloud, Wordbee, Lingotek). And "the rest" can be used on a Mac through a virtualized Windows installation. When it comes to the virtualized Windows, I would tend to ask why you're using a Mac in the first place (and then paying for Windows on top of that), but there might be reasons -- such as additional work in a field where Macs are preferred, or simply true love for all things Mac. (And I sure know some people in the latter category, Yves!)
Some people also asked about the use of multiple TEnTs:
Is it advisable to use several TEnTs? Does it make the translation process smoother or more cumbersome? Does it make sense to have several tools?
These are good questions for which there are no easy answers. It depends very much on who you work for and whether there are easy ways to convert between the tools in question.
The file formats generated by market leader Trados and, to an almost similar degree, by Wordfast are typically supported by other tools. The formats that Déjà Vu, memoQ, and Transit generate are much less supported by competing tools. True, the exchange of the translation memories is almost always supported through TMX, but to exchange the translation files itself you'll have to work much harder. You can almost always find a workaround, but often you'll spend a lot of time doing it. And if this becomes a regular feature in your workday (looking for workarounds), you might want to think about purchasing an additional tool and actually making some money rather than doodling around converting between formats.
If you are still looking for a translation environment tool, I would tend to look for one with the widest possible support of file formats that are used in your particular industry, including file formats generated by TEnTs. Will this always be a guarantee that you can work in that tool? Certainly not, and especially not when it comes to projects that are server-based, which usually will require you to work on the tool in which they originated.
Someone else asked this question:
Can you enter documents that you have already translated into all of these TEnTs? How easy is it?
The answer: it's hard, and it's called "alignment." Well, it used to be hard(er), but with specialized tools such as AlignFactory and the AutoAligner is has become much less painful. Still, the old rule holds: only align if you know that you're gonna need it. Don't fill up your translation memory with "stuff" just because you can. Translation memories become powerful when you find matches and terminology in them that are relevant to your current project. If not, they can be more distracting than useful.
And, lastly, a great question on the editing process:
Much has been made about the issue of quality in translation, and yet tools tend to have very poor support for the editing process, which is key to high-quality work. Do you have any thoughts on this?
I can't believe I didn't add this to my list of things that are missing from our tools that I published in the newsletter a few weeks ago. Absolutely true! There are only very few tools that are doing a good job with this, and even for those that do (I happen to think that Lionbridge's Translation Workspace product has a good concept by creating a browser-based editing package), it is often our own process that stands in the way of adding changes to the text in the very last stages of desktop publishing or functional testing that will then not be reflected in the translation memory.
One way of approaching editing with a tool is through what Déjà Vu, memoQ, and now also Trados Studio (through the SDL XLIFF Converter for MS Office) offer: the export of the translation project into a file that can be edited in a word editor such as MS Word or OpenOffice and the subsequent re-import into the translation format. The advantage of this process is that a) all the editing features of Word or OpenOffice are at your disposal, b) if you are your own editor, it's mind-boggling how many errors you see just because it's presented in a different format, and c) if you're not your own editor, your editor does not need the tool in question.
Another new development in browser-based tools such as Wordbee or XTM is version-tracking for individual segments, where you can see a whole long history for every segment and all the changes that have been made. While this helps only indirectly with the editing process per se, it's a good way for you or a third party to have a record of the changes.
Phew. Sorry about the long-windedness.
| 
