ToolkitSmall

A computer newsletter for translation professionals

Issue 10-4-163
(the one hundred sixty-third edition)
Contents
1. A New Generation of Tools: Wordbee (Premium Content)
2. A New Generation of Tools: Crowdin.net (Premium Content)
3. Pin the Tail on the Donkey
4. Open Up Your Mouth and Feed It!
The Last Word on the Tool Kit
Translation: The Good and the Bad

Translation is a noble profession, no doubt. And productive, too. Unless you have a week like I've just had: I was told by one of my clients that an over 20,000-word project I had recently worked on (and finished) had been dumped because their client decided that the source (!) was not acceptable (as if I hadn't noticed!); another client sent me a large wrong file -- and only discovered it after my translation was done; and yet another  . . . . well, let's not even get into that. Suffice it to say, it was a week of long nights and of feeling particularly useless as a translator. I guess it's good that these weeks are the exception rather than the rule in our profession!

It's also good that some continue to share the love, such as Translation Guy. As you're checking out this blog posting, have a look at the rest of his blog -- it may be some of the best writing in our industry.

Even more important, the good news last week was the announcement that the three men who were imprisoned for publishing a translation of the Qur'an in Dari in Afghanistan have been released (I wrote previously about their imprisonment).

Whatever our own faith background, much of who we are as translators we owe to sacred translation. This is where concepts of translation were formed, whether in Europe with Biblical translation or in East Asia with Buddhist translation. This is probably also the area where translations had the most profound impact on history as a whole (sometimes in a good way and sometimes not so much). Therefore, it seems fitting to pay attention to those who are still working in those fields.
1. A New Generation of Tools: Wordbee (Premium Content)

In my early college days I lived in the historic German/Roman city of Trier, but I never really warmed up to it as a place. (Perhaps it was the slow, drawling German spoken there. This is where I really started to talk with a strong Hamburg dialect as my own small -- and admittedly immature -- statement of identity.) But I sure liked Trier's mighty neighbor, Luxemburg. (Luxemburg may not often be described as "mighty," but its big city flair in comparison to sleepy Trier seemed to suggest just that.)

That's why I'm happy to report on some interesting translation technology coming out of Luxemburg: Wordbee. Wordbee is a young, completely web-based translation environment tool (TEnT) in two different editions (Team and Freelance) that offers a handful of unique features, some that also exist in other tools but are surprisingly mature in this one, and then some that its makers might need to tweak a little.

The basic concept behind Wordbee is a) the assumption that the future of TEnTs is web-based and b) that the "control over data and process should be provided in one tool" (a quote from one of the developers I talked to), i.e., a TEnT with a translation environment that includes robust project management capabilities which should not be overwhelming and could be learned on the fly. These are fairly high aspirations, but in some ways I can see why they are saying this.

Before we go any farther, however, let's talk about the finances: Since this is a web-based tool, you don't purchase a license but you pay a usage fee according to the SaaS -- Software as a Service -- concept. If you want to use it in a team environment, you can purchase any number of floating licenses that you can pass between your team members. A freelance license gives you space for 250 MB for documents and 250,000 segments in the TM; if you go beyond that you can buy more space or download the data to your desktop. But if you want to just check things out, there is a free trial period available.

In a recent newsletter I mentioned that we should not move faster than technology really enables us, and one area I quoted was cloud-based computing. In some areas of the world we can rely on fairly stable and ongoing connections, but in other areas we can't. And even ninety-nine point something percent uptime sounds great only when you are not on the other side of that number (the point something) -- and especially right before your delivery deadline. Still, the fact is that things seem to be moving in that direction, and it's probably not surprising that many of these tools come from within the EU. Not only are some EU translation organizations interested in finding cloud-based translation technology solutions, but most member states boast a robust Internet infrastructure.

And then there are other aspects that can only be achieved with a cloud-based solution. Think of crowdsourcing -- one area in which Wordbee is trying to make some inroads (thus "Wordbee"). But rather than looking at crowdsourcing as a process with a large undefined mass (à la Facebook or Twitter), it uses the principles of crowdsourcing in a professional environment: In-house translators and project managers subscribe to RSS feeds where new projects are announced and then locked once a translator claims them.

One feature that is amazingly strong in this tool is the translation user interface. So far I have always praised XTM's online translation interface as the best available, but after seeing Wordbee's, I am glad to concede that it should have that badge of honor. As most of us know only too painfully, online interfaces are notoriously hard to build. They often rely on the mouse, have very few or no shortcut menus, or are so filled up with random options and buttons that they are very hard to navigate. Wordbee's translation interface is different. Keyboard shortcuts can be assigned and customized to a total of 26 different commands; you can find right-click access to shortcut menus in many different areas of the program; and the translation view (in tabular format) along with the TM, dictionary (many preconfigured online dictionaries), preview (presently only implemented for HTML), comments, and revision history (!) panes are also freely configurable.

The translation process itself might need some work, however. Sending a segment pair to the TM requires several steps when it should really be an automatic procedure, and I was not able to find a way to add data on the fly to the terminology database (perhaps there is a way, but I wasn't able to ask over the weekend). On the plus side, there is an assemble process of existing database chunks (similar to that of Déjà Vu or memoQ). The list of supported document formats (MS Office 97-2007, InDesign CS3/CS4 [in the INDD and INX export formats], OpenOffice, RTF, XML, HTML-based formats, as well as various other source code file formats, including .resx, c, and Java files) is fairly impressive for a new tool, though I was disappointed that the promised extraction of embedded files in MS Office formats did not work for me.

I looked only at the Freelance version, so I cannot say much about the project management aspect. The file management in the Freelance edition confused me a bit at first but made a lot of sense after I doodled around with it a little bit -- I assume that the same will be true for the larger Team edition's project management components. They indeed seem to have tried to keep the options rather minimalistic to avoid overpowering the user.

One intriguing aspect of the product is this: You might argue all day about the clever name, but Wordbee's makers wouldn't care. The product is specifically geared toward customizability for whoever uses it, so you can even "rename" the product and add your own label -- meaning that we might see this tool within all kinds of third-party products and incarnations in years to come.

Oh, and before I forget: As cool as the user interface is, it wouldn't hurt to have a native English speaker go through it. Though it might lose some of its Lëtzebuergesch charm, it could also provide a couple fewer stumbling blocks.

ADVERTISEMENT

What is all the fuss about MultiTrans Advanced Leveraging Translation Memory (ALTM) technology?

  • Granularity - Get more matches by searching and retrieving full and fuzzy segments, paragraphs, terms, and even sub-segments

  • Context - View how ambiguous terms, sentences and even paragraphs were previously translated by accessing the full context

  • Automation - Automatically create TMs faster and with a higher alignment success rate than a Conventional TM

For more information visit www.multicorpora.com/ALTM
2. A New Generation of Tools: Crowdin.net (Premium Content)

On to the next tool: I would like to introduce a man whose name has the most occurrences of "y" of anyone I have ever met: Sergey Dmytryshyn. At least the Latin version of this name is quite impressive! Sergey is one of the folks behind a project that in itself is rather impressive as well: Crowdin.net. Before I tell you more about it, just one more fact about the Crowdin.net team: They are located in Ukraine. You may remember that I have pointed out many times that Ukraine seems to be a hotbed for language-industry related things (AIT, Practiline, TranslationDirectory.com -- check out links to some of TD.com's surveys in my Twitter feed -- and now Crowdin.net).

I had a long chat with Sergey about his site and tool, but let me tell you first about the tool itself.

Crowdin.net is a (so far) free platform for the translation of a number of software file formats in a crowdsourced environment. As of now, GNU gettext .po, iPhone/MacOS X .strings, Qt .ts, Android, BlackBerry, Symbian, .resx, Java .properties, and Delphi DKLang files are supported, but there are plans to support other formats also (you can actually vote for other file formats). The concept is this: After you register, you can create a project in which you can upload any number of the supported file types and create groups of translators (there is a page where you can invite them). You can also mark your project as open for everyone, in which case others can find your project and send you a message asking you for permission to translate. Once the accepted translator logs in, he or she can see the list of files in his or her language combination, open them (the software code is hidden and only the translatable text will be displayed), and start to translate. During the translation, the translator will be shown three different kinds of matches: MT matches from the Google and Microsoft MT engines (at least in my language combination, the MS engine performed notably better -- not surprising since IT data makes up a lot of what that engine is trained from) and translation memory matches.

The TM concept is quite a bit different than in most other tools. Rather than having individual TMs for every project, there is only one large TM for every language combination (if applicable) from which all data is read and into which all data that is translated is written into. Currently there are about 150 languages supported, many of which already hold very large amounts of software translation-related data. (English has more than a million strings; Arabic, Catalan, Czech, Danish, Dutch, Finnish, French, German, Greek, Hungarian, Italian, Japanese, Polish, Russian, Turkish, and Ukrainian have between half a million and a million; and even languages like Xhosa, Zulu, Afrikaans, and Macedonian have at least 100,000 strings. It's impossible to tell what languages those languages are matched with, but in many cases it is naturally English.) The data right now comes from translations of open-source software (I specifically asked whether they also used the Microsoft TMs, which Sergey emphatically denied) so the quality is a mixed bag, but that's what you get for now. When you translate you can select one of the TM (if available) or MT matches or type in your own translation and then "suggest" it. All this is possible with the help of (non-configurable) keyboard shortcuts. When you go back to a segment that you already translated the target field will be empty again but you can see your own translation in the Suggestions field where others now can vote for or against it. A file can be exported at any stage and the translations used will be the ones with the highest votes. As the project owner you will continue to be informed about the project in the form of an RSS feed you can subscribe to.

Interesting? I sure think so.

Here are some of the questions that I discussed with Sergey.

Presently the main target audience is developers of shareware or open-source applications for whom it is unreasonable to go through the traditional localization process (from the list of supported file formats you can see that there is a particularly strong support for developers of mobile applications). This group will always have free access to Crowdin.net, but eventually (presumably in the summer) the tool will also be marketed to service providers who will then have to pay a fee for its use. (Sergey: "I think they are a solvent audience.")

While right now everything goes into one large public TM, there are some thoughts about adding options to prevent the sharing of TMs -- which might be an important feature if language providers are indeed going to use this.

Also, there is presently no way to filter data (such as filter OS-specific terminology: Windows vs. Linux vs. Macintosh), and though this was not in the plans before we talked, it may be now. ;-)

Presently, the only file types that are supported are software files (see above), which of course is not going to be sufficient when other data such as documentation, websites, or collateral materials needs to be translated. While it is possible to convert text-based formats to .po files (in fact, the site offers a host of conversion utilities for this), this also will not suffice for everything so there are two options around this: translate within Crowdin.net and export a project-specific TMX file (planned for the near future) or for Crowdin.net to support XLIFF and translate all kinds of format that way (vaguely planned).

I would encourage you to go to Sergey's site and poke around. It has a very attractive user interface, and while it's not quite as streamlined in the translation interface as Beetext (and there are some real hiccups in the project management interface), it works just fine.

(Oh, and just to be fair, while someone is proofreading Beetext, it would not hurt to do the same for Crowdin.net.)

 

So are these the tools of the future? No idea. I do think it's fantastic that our industry continues to be interesting enough for developers to continually come up with new solutions, and in these two cases solutions that respond with a very creative approach to current developments, such as crowdsourcing.

 

And while we're speaking of current developments in crowdsourcing, Twitter just announced its leading crowdsource translators for German, French, Spanish, Italian, and Indonesian -- and you will be hard-pressed to find any professional translators among them.

3. Pin the Tail on the Donkey

I am sure that those of you who are not using Windows 7 get a little tired of "Windows 7" tips left and right. The fact is, though, it's a pretty good operating system, and if you are still using Windows now and plan to use it in the future, you will at some point also use Windows 7.

One of the areas that has been completely overhauled is the taskbar area. The QuickLaunch area -- the little strip to the right of the Start button -- has been sort of merged with the taskbar: you now "pin" shortcuts to applications directly on the taskbar. At the same time, the usefulness of these shortcuts has been extended significantly.

Windows 7 also introduced the "jump lists." These are accessible through most of the application icons (the shortcuts) on the taskbar or icons on the Start menu if you right-click the taskbar shortcut or hold your mouse over the Start menu link. With some applications these jump lists give you quick access to application-specific properties, but most applications just offer you a list of the most recent documents or websites you had open -- very, very handy. And there is nothing easier than pinning an application shortcut to the taskbar. Right-click on any .exe file and select Pin to Taskbar (alternatively, you can also select Pin to Start Menu); or you can right-click the taskbar icon of any open application and select that same command; or you can just drag it to the taskbar. To unpin, just right-click again and select, you guessed it, Unpin the program from Taskbar. Why do I spend all this space talking about this feature? I've recently worked in a number of projects with an uncommon combination of necessary tools and I found it super-helpful to create these little shortcuts just for the duration of the project and delete (unpin) them once the projects were done. So there.

And while we are talking about those buttons on the taskbar: the most logical way to open them may be to use the mouse, but there also keyboard shortcuts: WinKey+T selects the first of the buttons and then cycles through the rest; alternatively, the WinKey + <number> selects the icon that corresponds to the number (1 for the first, 2 for the second -- you get the picture).

Oh, and one more thing: If you need to open a second instance of a program (like a second browser window or a second instance of Excel) but you only have that icon that has now been replaced with an "open-look" icon, you can click on it with the middle mouse key (typically the wheel) or, if you don't have one, press Shift and click.

Oh, and ONE MORE thing: Nah, let's do that in the next newsletter.
ADVERTISEMENT

Increase your translation quality with the SDL CCM Methodology bundle

Receive all the software you will need to ensure consistent and efficient translations - at a great price!

Buy SDL Trados Studio 2009 SP2 and get 10% discount + SDL MultiTerm Extract 2009 for free.

Hurry, offer ends 30th April - www.sdl.com/toolkit_163

4. Open Up Your Mouth and Feed It!

I'm sure it would be fun to play with an iPad. But after reading several reviews I was depressed to realize that the one truly distinguishing factor between the iPad and a laptop is this: the iPad is to consume, the laptop is to produce (and to consume). I think this is a rather sad commentary on society: We look at computers increasingly as consumers (of games, multimedia, music, etc.) and tend to forget that this really is more or less incidental to computing -- which, as the term itself implies, denotes something proactive and productive.

So, on that note, let's get right down to two productivity tools, one of which is only in the concept stage while the other actually exists.

The flip-flop mouse, literally a mouse that is worn like a flip-flop and is clicked with your toes, was developed (in concept) for folks who can't use their arms -- but I think it would be a fantastic productivity tool for just about anyone who is tired of taking his or her hands off the keyboard all the time.

And, yes, there are also some foot-mice (and, I hate to admit it, but "mouses" would be grammatically correct here as well!) that actually are for sale. These, which Naomi de Moraes pointed out to me (no pun intended), are actually NOT pointing devices but foot pedals that can be programmed to "replace repetitive keystrokes or mouse clicks with the action of your foot. Enter, Tab, Shift, Ctrl, Alt... anything you can type on the keyboard can be memorized," including keyboard shortcuts. I sometimes struggle with using my hands in certain ways -- these might be good productivity tools for those times. (Or for times when I just want to wiggle my toes a little more.)

The Last Word on the Tool Kit

If you would like to promote this newsletter by placing a link on your website, I will in turn mention your website in a future edition of the Tool Kit. Just paste the code you find here into the HTML code of your webpage, and the little icon that is displayed on that page with a link to my website will be displayed.

This reader just added a link:

kv-emptypages.blogspot.com (the best-informed blog on machine translation)

© 2010 International Writers' Group