Fluency Translation

Linguists or Translators?

2012-02-21T14:20:00.000-08:00

One thing all good translators knowis that accurate terminology is reallyimportant. It's why translators usually list what domains they have experiencein - because the ability to use the correct terminology is of utmost importance.Inside a particular domain, what otherwise would be normal words/phrases in alanguage can be claimed for precise terminology. For example, a few months agoI was talking to a business development guy at a company about translationsoftware. He came from a military background and referred to himself as alinguist. I mentioned that I was also a linguist. I was referring to the factthat I have a B.A. in Linguistics. He, however, thought that meant that I hadbeen a military translator/interpreter. In his domain, linguist and militarytranslator/interpreter were synonymous. Usage, I think, tends toward makingevery linguist either a accent guru, or a translator who picks up languages forfun. Familiarity with the fields of syntax, semantics, morphology, phonetics,pragmatics, etc. are not what people think of when they think 'linguist', justsomeone who knows other languages.

This terminology domain overlap bringsme to issues that we run into all the time here and, I think, are commonterminology errors or conflicts in our domain of translation. Let me enumeratea few. Please let me know if any need to be modified or additional ones youthink are often mischaracterized.

Translation Tool (translation software)- In our field, this term is too ambiguous. We either need to refer to CAT(computer aided translation) or MT (machine translation).

Translator - This is understood insideour world as a document translator. Outside our domain, I think the tendency isto assume this means interpreter, and while these skills overlap, they are notthe same.

Fluent - There is a whole spectrum ofpotential adjectives you could assign to your non-native language. 'Fluent' Ithink is the most abused/ambiguous. It's useful to communicate general competence;it's too nebulous to base a reputation on.

Linguist - As I stated above, this meanssomeone with knowledge in the fields of syntax, semantics, morphology,phonetics, pragmatics, etc. A translator is not necessarily a linguist nor a linguist a translator.

PDF - This is a curse word. Youshouldn't use it nor allow others to. If you must, please specify whether it isa 'live' (created from an electronic document) or 'dead' (scanned orimage-based) PDF.

Terminology/Glossary - Some people getvery religious about how to label a list of words/phrases with other languageequivalents. Some of the problem here is that Terminology and Glossary havemonolingual meanings as well that overlap in many ways. The jury still is outon what we should go with on this one - Wikipedia describes glossary andterminology VERY similarly as do their dictionary definitions. Glossary on the whole seems to connote lessstructure, terminology more.

Localization - This term has largelybeen commandeered by the software industry. Localization andInternationalization (L10n and i18n) have specific meanings, but have now cometo mean just translation of software (and occasionally websites) . But the word'localization' also means to translate any content for a specific locale.

Computer literate - One of the mainreasons for the burgeoning CAT market is that clients expect translators orLSPs to understand everything about electronic content regardless of format,structure (or lack thereof), etc. As unfair as this might be, a good rule ofthumb is to error on the side of conservative about what you know and canhandle for computer skills.

Let me know what other terms I’mmissing.

Virtual LSP

2012-01-23T15:24:00.000-08:00

One of our longtime users once told us that there is a "growing morass of daunting technology faced by translators." One place where this could not be more evident is in networking a small LSP or group of translators together. For a long time, networking was a thing only available to the few. If you were a large LSP, you could have your translators connect in and share TMs and terminology seamlessly. You could share projects without sending countless emails. But that was only for the big-boys. With more and more options in tools, this is changing. Networking is something everyone is coming to expect everywhere. And that's probably a good thing. But as a group of translators or small LSP, that means more questions for you to answer, more things to become familiar with, and all with an already overloaded schedule. So what do you do?

You certainly have options, but it can all get very confusing very quickly. Should you use a web based tool? How does that work if I need to work with someone that doesn't have reliable internet? What about data security? Do I need to sacrifice tool quality to enable better networking? Should you use a server version of a desktop tool? Will that allow real-time sharing of TMs and terminology? What's a server? What's a desktop? And how much should this all cost me?

IT, security, and tool-related headaches can pile up quickly - not to mention bills. With all of this in mind, we are offering something that I think is...for lack of a better word...game-changing. We call it the Virtual LSP. We've had the pieces for a while, but now we're putting them together. Essentially, we offer hosting for all your tool needs in a secure environment. Project management, TM/Terminology sharing, real-time collaboration, and the tool packaged together. You're not going to be sacrificing tool quality for networking capability, you are not going to have the IT and security related headaches, and you're not going to be breaking the bank. I'd love to quote you a price right here because I think you'd be impressed, but we are offering some custom discounts based on needs so send us an email asking for pricing or give us a call.

Translationese

2011-12-20T14:21:00.000-08:00

Translators come from various walks of life - former computer programmers, long-time linguists, stay-at-home mothers, former government officials and everything in-between. Because of the diverse skill set required by the translation profession, many translators do not actually ever study translation. They grow-up bilingual or learn languages as a hobby, have a particular aptitude with words, and are familiar with a particular subject matter. In-depth knowledge of a particular genre of text is usually paramount among qualifications. So as much as studying translation is a legitimate pursuit in higher education, it is by no means a prerequisite for becoming a legitimate professional translator. Consequently, many of the long researched questions in translation are not widely known among translators. The findings of the majority of this research are often not influential on the day-to-day work of translation. While it is not a surprise to me that the intellectual pursuits of academia usually do not guide the business world, I do find it lamentable in some cases.

One case in particular is that of translationese. Translationese is the term for the characteristics of language that manifest themselves more abundantly in translated text as opposed to a text generated natively in that language. Researchers have long hypothesized which characteristics or features are unique to translated text, both dependent and independent of source or target languages involved. Here are a few of the features hypothesized: text length, sentence complexity, vocabulary, etc. Essentially translated text is usually more explicit, simpler in vocabulary, and normalized to perceived cultural and language tendencies. Language specific characteristics of translated text vary widely.

There were many early studies on the matter that seemed to validate the hypothesis. Recently, more complex tools and algorithms have been applied to the question and validated it beyond question. Modern tools can classify a text as either translated or original with accuracy greater than 90%.

So how should/could this academic finding influence day-to-day translation? Imagine with me a CAT tool that can identify instances of this "translationese." One that can show a translator, a project manager, a client, areas of translated text that identify the translation as a translation. Imagine a tool that could suggest a revision that might be less "translationese" and more natural. Sound interesting? I think so.

But there isn't anything like that...yet. Here at Western Standard, we would like to pursue this technology and include it in Fluency, but we want to know if it would be useful for our users/perspective users. How does it rank on your wish list? The speed with which we undertake this endeavor will be commensurate with the interest expressed, so if you think this would be useful, let us know.

PDF Translation and OCR Services

2011-11-23T09:03:00.001-08:00

PDFs are meant to be end-of-the-road files, not editable evolving documents. Despite this fact, translators often are handed PDF files and told "translate this". As much as you'd love to say "translate it yourself, buddy," there are many scenarios where PDF translation is a must. If you have direct access to the client, I would recommend asking if there is a version of the file that is NOT PDF. But in most cases, you don't have much recourse. The first thing to do in those cases is determine whether you have a "live" (created from a Word document or something of that sort) or "dead" (originating from a scanned document) PDF. Then you have a couple options: for live PDFs, you can convert the file to a Doc using either a PDF to Doc converter (such as the one in Fluency) or use one of the many freely available tools to extract and translate the text and send it back for the client for reformatting. For "dead" PDFs, you really only have one option: Optical Character Recognition.

None of these options are great. Often, just extracting the text and relying on the client to reformat is not acceptable to the client. PDF to Doc conversion is an imperfect science and has led many translators to charge extra for these types of translations. And then you have the dreaded OCR scenario... Fluency is one of the only tools on the market that will do anything with these types of files and the only tool that does it locally to your computer. Fluency's built in OCR technology does a good job for your basic scanned documents in about 20 languages but only retains text, not formatting. If you need to retain formatting from the scanned document, Fluency provides an interface where you can do that. For automatic format retention, you will need other software such as Abbyy FineReader which you would need to purchase separately. And even in the ideal conditions, OCR is even more of a black art than PDF to Doc conversion and will often require significant proofreading.

With that in mind, we here at Western Standard recently realized that we could help the translator/LSP out even more. While providing OCR technology in Fluency is going above and beyond the scope of most every translation tool, OCR is not your bread and butter. While in many cases you might be able to raise the rates on these types of documents, often the process is not as streamlined. You either don't have professional OCR software for format retention like Abbyy, you aren't familiar with the OCR tools, or you just don't like the time and effort it takes to produce the source document that you then have to translate.

Western Standard has two divisions - translation technology and publishing. We have been in the business of electronic publishing for well over a decade now working with World Book Encyclopedia, Encyclopedia Britannica, and others. In this area, we have significant experience and abilities in OCR technology and services. We have processes that make OCR processing much more streamlined and affordable. In that light, we decided that we could offer our services to any translator/company that has OCR projects and would like them done professionally, quickly, and at a reasonable cost. If you have any interest in these types of services, please contact us at info@westernstandard.com.

Conference and Beyond

2011-11-09T11:20:00.000-08:00

Thanks to everyone that stopped by our booth and participated in our give-aways. We had a great showing at the conference and we excited to see our users and talk with you. If you missed us at the conference but want to know more about the software, we are always willing to do a demonstration of our software so just send us an email and we'll get it set up.

We got great feed-back at the conference and are currently working on implementing a lot of useful new features including auto-correct. We're also pushing forward on our project management software as there was a significant interest in that as well. While we can't be everything to everyone, we feel we have good options for freelancer and LSP alike. We understand the pain the translator feels when required to use a tool that they don't like in order to get a job which is why our first and ongoing focus is to make Fluency useful for the translator. On the other hand, organizations/LSPs also have detailed needs on which we feel we give an innovative perspective.

Going forward, we will keep our ear to the ground and make sure we stay ahead of the curve with translation technology while remaining grounded and accountable to our users. Thanks again to everyone for the great conference!

ATA Conference

2011-10-17T14:43:00.000-07:00

The annual ATA conference is coming up next week, and we will be exhibiting. This event is always a lot of fun and a good chance to network/get the word out about Fluency. This year, we'll be giving away a netbook. Anyone that purchases Fluency at the ATA conference has the chance to win. Come swing by our booth - number 69 in the exhibitor hall. We'll also be giving a sneak peek on our FluencyFlow project management software. We'll be giving a tool tutorial on Friday afternoon - here's the information:

Title: Getting to Know Fluency 2011
Type: Hands-on Workshop
Day and time: Friday, 2:30PM-3:30PM
Location: Vineyard, Fourth Floor

Please tell your friends and come say hi!

CAT prices

2011-09-22T20:32:00.000-07:00

Money is always a touchy subject, and CAT tool prices are no exception. A couple months ago on ProZ, there was a thread demanding ultra cheap tools. There were strong feelings on both sides - tools should be free, or tools have a right to be really expensive. So let me explain to you our philosophy.

First of all, there are many ways to price a translation tool. Online tools often let you buy access up to a word limit. For desktop tools, there are really two options - renewal based licenses and perpetual licenses. Then, of course, there's free. We've considered these options extensively. Developing an online version of Fluency hasn't proved economically prudent as of yet, so the online pricing models are mostly irrelevant. And free means no paychecks which I think you'll understand is not an option. Our main pricing model is the perpetual use license. We do a have a hosted version of our TM/Terminology server that is renewal based, which includes server access and a number of user licenses, but everything else we offer is one-and-done. We also offer free support and training. Our prices are, however, not sky-high. For the most part, we are cheaper than our competitors, and in many cases, dramatically so.

This all sounds great, but one translator that was somewhat familiar with our pricing asked "Is this sustainable?" Our answer - yes. Some translators are jaded from the perpetual license model because of Trados. We understand. We do feel though that the idea of making you pay for our software yearly to support our development of features is...weird. If we develop a new version of the product in the future, it had better be good enough and priced correctly for you to upgrade/purchase. That's incentive for us to do a good job. We understand that in the renewal model there are also incentives to keep up development, but in that case, you also have incentive to keep paying - your don't want your license to expire. Software as a service renewals make sense when your software IS a service. But even good desktop software desktop software isn't really a service (unless it has a hosted server or comes with free support).

We released commercially over a year ago now and our prices haven't changed dramatically, but our product has. Our support is still top-notch, and we'll train anyone who asks. As far as our price-point, while driving around in a Mercedes and laying on the white sand beaches of Tahiti while the money rolls in (...I'll let you guess who I'm talking about) sounds nice, we believe in hard work and fair prices. When freelance translators are all making millions, maybe we will too. But until then, we'll keep things priced reasonably and accessibly, without compromising quality. If you need custom pricing, please let us know and we'll see what we can do.

The Virtues of MT

2011-09-13T11:32:00.000-07:00

I recently ran across an article extolling the virtues of Google MT - http://www.independent.co.uk/life-style/gadgets-and-tech/features/how-google-translate-works-2353594.html. While I agree with many of the ideas in the article, a few of the points and the whole tone of the article seemed out of line with reality. First, the idea that MT should focus on statistics more than extracting meaning I agree with...at least for now. But lets at least concede that that is fundamentally different than what we do as humans. I do believe in statistical theory, and have in my linguistic background studied the role that statistics plays in human language but I do NOT believe that word sequences and alignment statistics are the only determining characteristics of acceptability for a sentence. I DO look at meaning. So given this fundamental difference in processing, we have to assume the introduction of errors. So while the article praises the virtues of statistical based processing, let's temper our enthusiasm as that is only part of the puzzle, and probably not the most important one for real fully automated high-quality machine translation.

Which brings me to my next point. The most inexplicable part of the article is where the author, David Bellos, discusses how human translation errors are usually more dangerous than MT errors. I'm completely lost on the reasoning here. He says when Google MT makes an error, it's obvious, but the human translators make an error, it's not, so human translation errors are more dangerous. Analogously then, as a business owner, the data entry guy that 10 to 20% of the time spews junk is preferable because I know I can ignore his garbage vs. the guy that makes an error once every 100 or 200 entries. What? Isn't the reason I'm getting data entry (analogously translation) because I WANT to understand, not disregard, the output? I'm baffled. The only reason that human errors would be more dangerous, is because the output is actually useful - and that's kinda the idea.

Finally, the name of the article is "How Google Translate Works". While I understand that he's probably trying to write for a broader crowd, it doesn't really go into any technical detail beyond "it uses human translations" and "it uses statistics". No equations, no specifics. And then he makes the assumption that because our desires and needs are the same, the premise of everything-has-been-said-before MT should work. Once again, while I don't outright disagree, linguistic nuances go a little deeper than that.

Whether or not you like Google, the virtues of a good MT engine (which Google's system is) are numerous. Many would say that we're just beginning to tap the MT-as-a-professional-translation-resource well, but let's stay grounded here in reality.

Alternate Endings

2011-08-23T14:57:00.000-07:00

The only way this post is at all related to translation is that the movies discussed are global phenomena. My boss was spouting off alternate endings to popular fantasy/sci-fi movies to me earlier today and I think they're worth repeating. Here we go:

Star Wars Episode 4 - A New Hope - Obi-Wan Kenobi is fighting Darth Vader with Luke looking on. Obi lifts the light saber above his head to go for a vertical downward blow. Darth lifts his light saber to block on the horizontal. Obi-Wan deftly switches his light saber off and on as he swings down avoiding the block and killing Darth. The End.

Lord of the Rings - Fellowship of the Ring - Gandalf returns from his research about the ring and does the creepy "is it secret? is it safe?" thing. He then calls his eagle friends that he uses later in the movies and gives them the ring with instructions to drop it in Mount Doom. The Nazgul realize what's going on but are too late because they are in the shire. Sauron doesn't have any other flying creatures to help so he stares with the fire-looking eye helplessly as the birds drop the ring in Mount Doom. The End.

Harry Potter and the Prisoner of Azkaban - Hermione uses her time-turning device with Harry to go back to 25 years before and find adolescent Tom Riddle eating lunch. Harry points his wand at Tom and yells avada kedavra. Tom Riddle dies and Voldemort never exists. Harry is never famous but gets his parents back. The next 4 books/movies are combined into one called "Hogwarts - Teenage Angst and Quidditch." The End.

Hope you enjoyed.

What's in a native?

2011-08-19T13:58:00.000-07:00

I recently read a portion of a ProZ forum thread that was discussing translation agencies that request translators to translate into their non-native language. This is, as mentioned on the thread, at a minimum unusual and in some cases, outright ill-advised. The thread progressed (or digressed?) into a discussion of native language vs. 2nd language (or nth language) and what really makes up a native. Would a native by any other name smell as sweet?

The questions/opinions expressed on the thread brought me back to my days in linguistic classes - in particular, psycholinguistics. There is a theory in linguistics known as the Sapir-Whorf hypothesis that language shapes thought. It's a highly debatable idea, but one that certainly bears relevance when discussing qualities in natives that make them indispensable in the translation process. Concrete examples of this phenomenon are discussed in this paper: http://www-psych.stanford.edu/~lera/papers/gender.pdf As discussed in this article, these findings hold largely true for true bilinguals as well - those that grew up speaking two languages. The prejudices that arise from one language seem to be carried over into the second language. For example, in the study a group of bilingual English/Spanish children who had grown up in America were asked to describe a bridge in English. The descriptions were masculine. Likewise, a group of bilingual English/German children who had grown up in America were asked the same question in English. For them, a bridge displayed more feminine characteristics. This, of course, mirrors the gender associated with these nouns in the respective non-English language.

How does this all play into translation? Beyond the obvious fluency issues when a translator translates into their non-native language, there are various cultural and cognitive nuances that are really inaccessible to non-natives. This fact however begs the question: if a non-native can't access these nuances, how are they going to be true to the original material? Fact is, they probably won't, but at least the target language will be natural and culturally/cognitively plausible. While these more nuanced issues with translation are certainly more relevant to literary translation than to technical documentation, there are many cases in normal translation where this is an issue. Marketing and advertising to name a few.

Of course, "bilingual" and "native" really have gradient values which compounds the complexity of these issues. As far as linguistic theory goes, the acquisition of a language before adolescence is certainly preferable in order to claim true native or bilingual abilities. But who am I to say that to a friend of mine from the Netherlands who learned English at age 19 and now has no discernable accent and has lived in America for 30 years? Is he less native in English then another friend of mine who grew up speaking Dutch and English? The relevant translation question would then be: which should either of my friends translate into if they abide by the rule of only translating into their native tongue? Is it just the one that they feel more comfortable in? Or where they've lived the longest? Or most recently?

Well, putting the unanswerable questions behind us, I agree with the generalization that translators should only translate into their native language. If you claim you're native in a language, you should speak it with 100% fluency. Just beware that aside from fluency, natives are not created equal, cognitively and culturally speaking.

MT and reality

2011-08-01T07:57:00.000-07:00

The question of MT post-editing for professional translation work is pretty much an unavoidable question nowadays. It's also a powder keg for some people. For some MT is a godsend with potential to speed up translations like nothing before. For others it's a crude hack that should be avoided at all cost. I think most would agree that, just like MT in general, it really depends on the language pair and the subject matter in question. The debates surrounding this question get quite heated in many cases, but what I've noticed over the past couple years is a surprising lack of data to back up any claims. Sure, there's anecdotal evidence all over - "I use it and my translations are better than ever," vs. "MT slows me down and messes me up ALWAYS." But where's the data?

In search for an answer to this question, I stumbled upon this article:

http://www.mt-archive.info/AMTA-2010-VanEss-Dykema.pdf. It appears that someone is actually looking into this empirically. The difficulty in this empirical approach to this question of course lies largely in translation metrics - how do you really determine translation quality? There's no question that using MT and doing post editing will give you different results for those you would get if a translator did it the old fashioned way, but how do you determine quality of the one versus the other? And when does speed increase trump linguistic accuracy?

The article mentioned above is a case study proposal from last year that doesn't seem to have had results published yet. I'll see if I can get a hold of any preliminary findings, but it's encouraging to me that in the near future this question won't just be he-said-she-said. Without the data, I can see both sides of the argument - if post editing of MT gives you quick, intelligible results then I can see many cases where it would be very useful. On the other hand, MT doesn't think. If you are familiar with the Vauquois Triangle (a often used graphic to demonstrate the different levels at which meaning transfer for MT can occur), anything that doesn't use an interlingua to translate is not going to the level of a human and thus probably misses nuances that would be left out of a MT editing processes. And while MT can stay try to word, syntactic, and some semantic constructs through statistical leveraging of enough data and some rules, translators understandably balk at numbers and rules governing something both innately artistic and cerebral.

So until we have the data, live and let live. If professional translators are using MT to increase their speed without compromising quality, then great. And if you are one of those that believe for whatever reason that MT is never a viable solution for professional translation, then ponder and paint away.

Fluency and the Mac

2011-07-18T12:40:00.000-07:00

We often run into translators that work on a Mac that are having issues finding a CAT tool that they can work with. Then there are the translators that are investigating whether to buy a Mac or PC. Fluency runs on .Net (like most of the main tools out there) which is a Windows based technology, so we want to take this chance to explain how to run Windows software on a Mac. There are a couple ways that you can run Fluency on a Mac, but they don't come without some additional work. We aren't opposed to developing a Mac version, but it isn't exactly a straightforward process. So, until we have a native Mac version, we have to run Windows on a Mac a use Fluency. Currently for prospective users, there are advantages to PCs but Windows based software can be run on a Mac as well - including Fluency.

The easiest option to get Fluency on your Mac is to install Parallels which costs about $80. This will also require a license of Windows. Parallels is one of the most popular methods of running Windows programs on a Mac, so this solution allows you more flexibility beyond just installing Fluency. Boot Camp is another option, which will allow your Mac computer to boot in either the OS X operating system or Windows. Boot Camp is free and you still need a Windows license, but in this case you have to reboot if you want to move back over to your normal interface. VMware is another option which is also free (except for the Windows license) and similar to Parallels, but not integrated in the same way. Once you have one of these options on your Mac, you can install Fluency without any issue.

As far as a general decision between Mac and PCs, everyone has an opinion. There are good reasons to buy either. In our company's experience in the translation business, PCs natively give you more options for CAT tools and fewer interoperability issues because PCs still have the majority of the market. For many translators, if you don't want to try one of the Parallels/VMware/Boot Camp options, it might be worth investing in a PC even if you use a Mac normally if only for saving your time avoiding work-around issues between PC/Mac.

This is still in many ways a Microsoft business world (they have the vast majority of the market) so that should be a consideration when looking at computer options. As far as mainstream translation tools go, only online tools and WordFast run natively on Macs (Swordfish, OmegaT, and some others also are Java based, but they aren't as big). You can, as already documented above, use Boot Camp or Parallels or VMware, but those are just Windows-on-a-Mac options, not true Mac options. And as much as online tools will tell you all the advantages of the "cloud" for interoperability, online software just doesn't have the resources that desktop software does because it's in a browser.

If you are interested in running Fluency on your Mac and need help, let us know and we will help you out. Getting any of the Windows alternatives up on a Mac isn't always easy so we're here to help. Just contact us and we'll make it work for you.

Speeding Up

2011-06-23T12:50:00.000-07:00

One of the main premises behind translation tools is an increase in speed. The trick with any tool though is that in order for it to make you more productive, you've got to learn how to use it. That's one of the main reasons we offer free training when you purchase. The faster you learn how to use our tool, the faster it will...make you faster. In these training sessions we discuss pretty much anything you want. The sessions are one-on-one with a trainer, in most cases, so we can do that. The problem is that the sessions usually revolve around high level training - here's the TM, here's how to use it, etc. Unfortunately, the lower level stuff is often where the real speed increase is. So here are a few of the things you should know.

First, keyboard shortcuts. These are invaluable. This is true both for Fluency and for Windows in general. the Windows key + D minimizes all of your windows at once. Windows key + E opens an explorer window. Windows key + R opens a run window that you can start any program from. In Windows 7, moving a window to the top of the screen will maximize it. Moving it to the left or right will set up a 50% side-by-side view with another window. Shaking the open window will minimize all the other windows. A lot of your probably already knew these. In Fluency, maybe you're not as familiar with the shortcuts. The whole list of pre-defined shortcuts can be found under Help > Keyboard Shortcuts. You can also assign key combinations to particular characters in the preferences menu. Some pre-defined shortcuts you might useful: CTRL + M will merge the current sentence with the next one. CTRL + # will change the active pane (1 = source text, 2 = target preview, 3 = target text, 4 = glossary). Keying up and down in the glossary pane will initiate the extended glossary search for whatever word is selected, and enter on the selected word with put it in the Target text and let you continue translating. Enter and CTRL + Up or Down will move you from segment to segment. CTRL + W sequentially copies a word from the source for that segment. CTRL + G will initiate a terminology lookup on characters from the last space up to the cursor and let you auto-complete from the list. There are a lot more, so I recommend looking them up and printing them out for quick reference.

Second, Regular Expressions can come in quite handy, not only in Fluency but in dealing with documents in general. Say you get a document that has been converted from a PDF that you import into Fluency and you come to find out that the "columns" that it shows are just fake - they end with new line characters. Which, of course, makes the segmentation very bad. What do you do? Well, open the document in Word and do this find and replace - "^13([a-z])" replace with " \1" (Word's wildcard characters are not your "regular" regular expressions. In any other software, "\n([a-z])" would be your search string). These expressions are often not perfect but do a much better and quicker job than you going through and trying to find problems manually. The "\1" above is a capture group which preserves your original text when you are doing a find and replace. Without it, you would not know what to replace the [a-z] with and the search would be useless. Fluency has capture groups available in its Find/Replace function as well. Regular Expressions give you the flexibility to look for a term that could have any number of morphemes appended to it, so they are available in various other areas in Fluency besides Find/Replace.

Third, a comparison tool lets you see how a document has changed from a previous revision. Fluency has a TM profiling tool that shows you the number of x% matches in a document, but sometimes something with a little more granularity can be helpful. I'm sure most of you are familiar with Word's Track Changes ability which can be useful for editing but can often get messy. There are many free comparison tools that will show you changes from a revision of a document, but the best that I know of is Beyond Compare. It's not free, but it only costs 30 dollars and it has a lot of flexibility that other free tools don't. It can't do PowerPoint or Excel that I know of though. For comparing those, you can try WinMerge which has a PowerPoint plug-in, I'm told, or DiffDoc which I've used myself but that costs $400 dollars.

Lastly (ironically, right?), take a look through the Help manual. You might find a function or two that you really could use and hadn't thought about, or something you wanted that you didn't think was available. Often, the documentation/help is a last resort, and we believe to use Fluency, you shouldn't need to spend an hour reading how, but in order to make use of some of the more advanced/complicated features in any software, a quick glance over the Help file can't hurt.

As always, if you want something to speed you up in Fluency, let us know! We often add features for translators (usually weekly) that help with their processes and speed them up. And while we continually try to innovate, we can't help with your specific situation if we don't know what you need.

In Context

2011-06-07T09:12:00.000-07:00

First of all, thank you to all of you for bearing with us last week. We were overwhelmed last week by the amount of interest our ProZ ad generated and have had about 1700 people download our trial! Please, if you have questions or need support, don't hesitate to contact us.

Last week I was looking around on my alma mater's website and saw an article about context and translation that had recently been published by an old professor of mine and family acquaintance, Alan Melby. He's done extensive work in the translation field and is serving on the American Translators Association Board of Directors. Needless to say, I thought it would be worth my time to read the article. It was an interesting read. If you want the full text, it's available here http://www.trans-int.org/index.php/transint/article/viewFile/87/70

The point of the article is to claim that there are 5 aspects of context: co-text, chron-text, rel-text, bi-text, and non-text. If you want an in-depth explanation of these, I refer to the paper. In short, co-text is the text surrounding the word or phrase. This is the definition that most of us are familiar with. Chron-text is the context of different versions of the document in which the word/phrase is found. Rel-text is the related information available in other resources. Bi-text is the context available from translation memories. Finally, non-text is what Melby refers to as 'paralinguistic information'. The context of a word/phrase in the language and culture as a whole.

There are many interesting lines of discussion this could generate, but for the sake of brevity, I'll just pick two. First is machine translation. I said a few posts ago my background is computational linguistics, and as such I'm familiar with the algorithms used in machine translation. Most good MT systems now are largely statistical, with some systems making use of linguistic rules as well with a hybrid approach. In either case, one of the big questions is "what is context?" The problem that Melby's context definition poses for MT is two-fold. Namely, with the first 4 "-texts," how far is far enough? What portion of the surrounding text needs to be looked at? And, the last "-text" is currently intractable for MT. MT doesn't know culture. Given Melby's definition of context, it's easy then to see why MT struggles sometimes and why, despite the care given to the sophisticated statistical algorithms, MT will never replace the human - unless they can understand culture and "paralinguistic information".

Which brings us to the next line of discussion - how do you as a translator cope with the all the variations of context for a translation in the limited time you are given? This is where the TEnT comes in handy. You don't have the time or the resources without a tool to look at everything you could to inform you on a translation. Even with a tool, you won't have time. But with a tool, you should be able to quickly figure out what context you most need to reference and do your research. This is where one of the strengths of Fluency is manifest - you have quick easy access to lots of resources. Online resources, integrated terminology, TMs, etc. Online resources allow you access to rel-text and non-text to a degree without wasting your time. All you need to do is highlight the phrase in the source window (or just hover over a particular word) and click on the appropriate tab, and off you go. You also get an intuitive approach to co-text, where each segment is still part of a whole document, not cells in a table, each an island, entire of itself. Not to mention any images and formatting that could give you further "paralinguistic info" on the document.

Anyway, I recommend you look at Melby's article. It's an interesting exploration of one of the main challenges that face you as translators.

Marketing, paraphrase, and translation

2011-05-25T11:58:00.000-07:00

Déjà Vu released DVX2 recently. I was perusing their feature list earlier today as I try to stay current on what the competition is doing. At the last ATA conference, our booth was stationed next to Atril's and we were impressed with the friendliness of their representatives to a competitor, so the following is not meant at all to be derogatory to them or their product. But as I read, I was struck that many of the features had a lot in common with things that we released in the last 6-8 months. It's not that surprising that given the same problems, we came to similar solutions, but it is the case that DVX as one of the big tools out there has the budget for a lot of hype and marketing that we don't. Sure they polished some of these features up and spun them in new ways, but the real genius is in marketing. Marketing makes these features sound amazing. Now don't get me wrong, I'm sure some people are going to love and really use these new features, but in the variations of these features that we have had in Fluency for a while, we haven't really seen much use. We released a term miner with no real response. We have AutoTM and term lookup as you translate for an autosuggest and while our users like the concept, only a small percentage actually use it. But that's why it's all about marketing. It's the flash of a cool coined term for a feature that often sells it more than its usefulness.

That got me thinking about marketing in general. I'm not a marketing guy but in many ways, marketing reminds me of paraphrasing. If I say "Fluency is the first tool that can translate publisher files," that's boring. If I say "In an industry first, Fluency allow you to seamlessly translate Microsoft Publisher files," it says essentially the same thing but sounds more exciting. The thing about marketing is that instead of staying as close to the original meaning as possible, your goal is sound as good as possible without straying too far from the original meaning.

Like I said, marketing doesn't really interest me. But paraphrase does, as does translation. Paraphrase and translation have even more in common. In both, you need to be as true to the original meaning as possible. Where marketers want to author the paraphrase, translation and linguistic paraphrasing want to be accurate to the original author. While the means are the same, the ends are a world apart. So it would seem that if there is a group of individuals that would shudder at marketing's ends, it would be translators. But like I said, marketing sells, even in the translation space. So all this is not to say that we here at Western Standard aren't going to market our product (though it won't be me). What I'm saying is that on the home front here at Western Standard, we know that when it comes right down to it, the real worth is what's behind the pretty paraphrase.

CAT got your tongue?

2011-05-19T13:21:00.000-07:00

The Fluency Translation Suite is what many people refer to as a CAT tool - a Computer-aided (or assisted) translation tool. In other circles they are referred to as Translation environment tools, or TEnTs (coined I believe by Mr. Jost Zetzsche). The idea behind the technology has been around for over 20 years now: save previous translations and give translators resources when they translate. It began with DOS based tools, with the most familiar one known as TRADOS, a clever derivation from the words "Translate" and "DOS". Since then, these tools have evolved substantially into comprehensive translation solutions. There are many competitors for TRADOS now, some of them quite good.

Enter Fluency. About 5 years ago, our parent company--The Western Standard Publishing Company--was awarded a government contract that involved language technology development. This contract included some translation work. Hence the development of a translation division of Western Standard. As we began this translation work, we surveyed the available CAT tools and got the distinct impression that these tools were not geared for what we needed - fast and accurate translations of unique material. So we built our own tool which we named Fluency.

At the end of the government contract, we opted out of pursuing Language Service Provider work in favor of pursuing the technology side of translation. After a few months work on the Fluency tool, we brought the tool to market. In gearing up for marketing, we visited multiple language conferences. One of the banners we brought claimed that Fluency was the world's smartest translation software. One translator questioned me particularly about this claim in context of the market full of CAT tools that have been around for decades. It was a legitimate question.

The answer is in the question. There's a tool disconnect. After having been around for decades, tools have language and translation so wrapped in technology that you can't see the meaning for the words. This standard and that standard, this file format and that file format, this feature and that feature. One of our clients told us after we responded to a technology question that there is an "ever-growing morass of daunting technology faced by translators." That's where Fluency is different. We try to strike a balance between cool technology (I'm personally a computational linguist interested in pretty much all things techy that deal with language) and good-ol' translation. We know that to you, time is money, and because of this we get out of your way and let you translate. We'll provide TM matches, terminology, and a wealth of other resources, but in a simpler, more friendly way. Proverbially, we don't want the CAT to get your tongue.

To us, features need to be motivated not by "coolness" but by actual usefulness in the translation process. Will it speed the translator up or help accuracy? Is it practical? Will it get in the way? If it doesn't pass these questions, why add it? Features for the sake of features just muddies the water.

So that's a little about us. We are around to stay, and would love to hear your feedback. Please let us know what you think Fluency needs, or what it does well. We'll post information about cool technology, but also cool language tid-bits.