THE TRANSLORIAL TOOL KIT – THE SHEEP SYNDROME

The latest technology update from the Tool Kit. BY JOST ZETZSCHE

“Sheep syndrome” doesn’t have very positive connotations, but in some cases the result of one following another is a lot more beneficial than its negative image might let on. Take technology, for instance. In a competitive landscape, the innovation of one competitor will inevitably lead to others following suit — if that were not the case, we probably wouldn’t have a competitive landscape to start with.

In the case of translation environment tools, this phenomenon has been repeated over and over with quality assurance features, context matches, concordance searches, and, lately, with AutoComplete features, i.e., the ability of the translation editor to predict or suggest text based on a few typed letters in combination with external data, such as the translation memory or other databases.

The latest tool that has now unveiled this feature is Wordfast Classic, and I have to admit that, somewhat to my own surprise, I like its version of AutoComplete the best so far. More on that later.

Long-standing readers of this newsletter know that Wordfast Classic has always puzzled me to some degree — while truly “only” a combination of Word macros, it has much more power than you might think. Years ago when I wrote about Wordfast, I likened it to “The Little Engine that Could” and the children’s story with the same title. What can you do with a collection of Word macros all wrapped up in a four-megabyte .dot template file? The answer: a lot more than first meets the eye.

Everything in Wordfast Classic is managed in the multi-tabbed Wordfast dialog, which can be accessed with the shortcut Ctrl+Alt+W or by clicking on the button that was formerly an “f” and is now, well, dots on stripes — I’m sure there was an idea behind the design, but it’s not apparent to the uninitiated. Once the dialog is open, you can see all settings concerning

• translation memories (that can now be shared among various translators);
• the “Very Large Translation Memory” — which, unlike the button-that-was-formerly-called-“f”, is a rather descriptive designation (besides being large it’s online and can be accessed by everyone with matching language combinations);
• machine translation (Google, Microsoft, and, if installed, desktop-based systems);
• terminology databases (that can also be shared with others);
• quality assurance settings;
• AutoComplete features;
• and many very advanced features in the ominously named Pandora’s Box.

In the latest version of Wordfast Classic, some of the new features include increased sharing abilities between translators on local networks, enhanced terminology recognition, newly designed separate windows for TM and terminology matches and QA results, as well as the above-mentioned AutoComplete feature.

As I said, it’s this last feature that I really like. As with its competitors, you can use data in the terminology and translation memory databases (including subsegments) that will automatically pop up and give you suggestions based on your first few character strokes, but ironically it’s the fact that you can also use MT matches of the above-mentioned machine translation engines that actually makes it just a little more useful. While in many cases those matches are less than relevant, the pop-ups are unobtrusive enough to just ignore, and they will dynamically change as you continue typing. And if it’s just one in five of these dynamic matches that ends up saving you time and energy, you have reason to be thankful.

So, all is good in Wordfast-Classic-Land? Well, naturally the number of formats are still limited to what can be processed in Word (plus to some degree PowerPoint, Excel, and HTML through a prior conversion), but if those are the formats you exclusively work in, I think you might have a lot of fun with Wordfast.

Trips and Ticks

The online website word-counting tool webwordcount is a quick and helpful alternative for counting words in simple websites. You can use it only for static websites (i.e., websites that are not content management system-driven) and only on text-based files (text, HTML, etc.), but if you need to provide a quote for those kinds of websites, it’ll be helpful. I also like that it actually explains a bit about how it counts words — something that’s missing in most other tools. Note that I have not tested it with non-Western languages.

Skype has become the communication tool of choice in the translation community (I don’t even remember when I last started MS Messenger or another messaging tool), so it’s been annoying to many that the latest versions LOVE to gobble up memory and slow things down significantly, even with high-powered computers. How to fix it if you have that problem? Uninstall any Skype version above version 5 and install Skype 4.2. You can download it from Filehippo. While it’s true that you won’t find many of Skype’s newer features in the older version, I haven’t missed anything and prefer the processing power I’ve regained. (Just make sure that you know your user name and password before you uninstall.)

As of just a couple weeks ago, the WordPress Multilingual Plugin (WPML), a fairly inexpensive suite to manage translation for websites built with WordPress, is now supporting XLIFF, the translation file interchange format. In fact, on one of their help pages you’ll even find detailed explanations for how to process these XLIFF files with tools such as OmegaT, memoQ, and Trados TagEditor.

One thing that always drove me nuts with Microsoft Outlook was its limited non-Western script ability. While it was possible to display Russian, Hebrew, and Chinese in bodies of emails, the text always showed up garbled in the subject lines. Well, it turns out this was my own fault (please tell me I wasn’t alone!). From Outlook 2003 on, the underlyingOutlook .pst repository files have all been Unicode. This includes the ability to display all languages appropriately everywhere in the program (and increased the storage capacity by a few hundred percent). However, I didn’t realize that when I updated from earlier non-Unicode versions of Outlook to a later version by simply using the same .pst file, the repository file remained its old limited self. Only when my Outlook recently balked at me for having too large an email repository did I create a new Unicode .pst file (there are plenty of tutorials on the web that show you how to do that). I copied everything over, and I can now see everything as it’s supposed to be (and now I can receive your emails without Outlook telling me that it just can’t take anymore).

TMbuilder is the “the easiest Translation Memory export creator,” according to its creator. I’m not sure this is true — there are a number of tools that do similar feats and similarly easily — but it’s still a helpful little tool if you have data in Excel or text-based formats that you would like to have in a TM-compatible TMX format. And: It’s free! JZ