Here are a couple of stimulating responses that I received weeks ago on my article about illegal software and a general unwillingness among some freelancer translators to operate under common business rules.

Trados expert Iwan Davies shared these insights:

Here in the UK, we pay up to two-thirds as much again for Adobe software compared with what you pay in the US. The list price for Adobe CS3 Design Premium in the US is $1799. Here in the UK, the exact same software is £1409 ex VAT (currently over $2800)! Even allowing for the slippage in the exchange rate since the prices were set, that is a staggering differential. Adobe will put it down to localization costs, and the cost of selling abroad, but when you realize that they maintain a massive distribution centre in Europe (Ireland) and the UK English version is basically identical to the US English version (we still have "colors" in Photoshop, for example), then that argument really fails to hold water. With the price of flights so low nowadays, you can just about fly over the Atlantic, buy the shrink-wrapped package off the shelf and be home in time for tea and scones with change in your pocket from the £1400 it would have cost over here. In fact, I know at least one UK small-business owner who has done just this in the past, having arranged to also meet with a client and do some business at the same time! It's no wonder then that people feel ripped off and compelled to seek alternative solutions, such as those described by your correspondent and those acquaintances of mine in my second point:

Educational licenses. The number of people (especially freelancers) who, when I complain about the point above, respond with "Oh, you can just buy the educational license, you know!" is staggering! When I point out that an educational license explicitly forbids use of the software for commercial gain, they just look at me as though I've just fallen off the tree! How naïve I must be! At least Adobe demands student/teacher ID for the purposes of activation of its educational licenses (though my colleagues simply get a letter from their kids' primary school to confirm that the child is a pupil) -- I could walk up to any PC store here and buy the "educational" version of Office 2007 for around 20% of its retail price, in cash, without having to prove anything!

The FOSS software [Free and Open Source Software] backlash is gaining momentum, and it is certainly worthy of support. Having been a Trados guru for 10 years, I'm not about to ditch Workbench and move in to FOSS solutions, but I'm finding more and more that, especially when it comes to the technical management of localization projects, FOSS is where it's at.

Mario Chávez looked at things from a different angle:

In August 2005, I returned to Argentina for two years to pursue master's courses in translation studies and audiovisual translation. There are large translator communities in the big cities such as Córdoba, Rosario and Buenos Aires. In Córdoba, where I was born, I was very surprised to find out the following:

I will not make a moral case for using only licensed copies of any software program, but I think there are plenty of reasons to make a business case out of it. A software program is a tool, and good tools cost money. From a business standpoint, investing in the right tools will always make me more competitive and productive, not less. (...) What I would suggest for TEnT developers and marketers is simple: invest in translation education along with marketing your tools. For that, we need a strategy to work with translation schools and programs to strike a balance between translation education and profitability. The big companies, such as SDL International and Catalyst, are doing it on a case-by-case basis. Why not the small fry or the open-source community?

A principle of marketing calls for segmentation of the market. Developers, don't just play a zero-sum game of all paid licenses or cracked licenses, because it's a losing proposition. Offer a tiered system of versions according to market segments, subsidize some copies for translation students so that they can find it attractively priced according to the market they live in. Going back to the Argentina example, the current salary for a retail worker is about AR$800-1,000 a month (about $300 dollars). A working translator in Argentina may make between AR$2,000 and AR$4,000 a month (in a good month). I am not suggesting that TEnT software developers should solve the world's economic problems, but do work with the different markets.

I have been quite critical of some of our fellow translators in the past because of their (lack of) purchasing habits of the tools that we need to support our work. But as these comments point out, there are some strategies that the software vendors themselves could adopt to counter this trend. Now, I know that some vendors offer their products at lower prices for developing markets, but maybe this should become common practice.

Here's another business approach that looks at things from a completely different but also very helpful perspective. For corporate and LSP customers, Transit offers a service to convert all their legacy data from competing tools into Transit data, a service that Idiom also offers. To me it seems that this is exactly what we need -- not only for corporate and LSP customers, but also for freelance translators. And if the software vendors themselves don't offer it, maybe it's a new job opportunity for technically savvy translators with the right tool set!

MemoQ -- the very user friendly integrated translation environment which surpasses its competitors with its performance, savvy user interface and integration capability.

What's more, MemoQ offers a real-time document preview for Microsoft Office formats and context matching out of the box.

Download the tool now and join the community!

MemoQ users have more fun.

While we're talking about the legalities of running illegal software, it's not a big step to cracking passwords. . . . Unlike the former, however, the latter can be legal and even rather necessary at times. The three kinds of password encryptions that arguably give translators the most headaches are the ones for Office files -- particularly Word files -- zip files and PDF files. Now, some of those files may be encrypted for good reason (like my ebook if you have not bought it legally . . .), but for others, especially those you are asked to translate, you are better equipped with the right tools to open them.

As with so many other tools, there are a great variety of tools out there that allow you to find the magic word, but the tools that I have been using successfully come from ElcomSoft -- unfortunately in different (paid) versions for the different products. Now, the magic word is not magic for nothing, and it isn't easy to find, even for the smartest software. Plus, there is also a reason for the different levels of "password strength" that you are asked for on the various websites and programs you may be choosing a password for.

Like any of their competing tools, ElcomSoft tools essentially have two strategies. The fast and quick way is the dictionary attack. This is for simple-minded folks like me who know that they would forget their password if it were not the dog's name or something like that. This attack only takes a few seconds and all it does is to run a large list of terms against the actual password until the correct one is found. If that method is not successful, a second method is applied, the so-called brute force method. Typically you can tell the program certain parameters (like only lower- or upper-case letters, with or without numbers/special characters, or the presumed length of the password) and depending on how complex and accurate these are, a successful attack can take minutes or a whole day.

Still, there is an awkward feeling of success when you have finally cracked the code and opened the file. You know it was "OK" to do, but you still have that sheepish "Beagle Boy" feel.

(And if you're not sure who the Beagle Boys are, maybe one of the following terms will help: Rafani, Bjørne-banden, Panzerknacker, Μουργόλυκοι, Golfos Apandadores, Chicos Malos, Penipoisid, Karhukopla, Rapetou, Braća Buldozi, Kasszafúrók, Bjarnabófar, Bassotti, ビーグルボーイ, Suņapuikas, Zware Jongens, B-gjengen, Bracia Be, Irmãos Metralha, Братья Гавс, Buldogi, Bierdnaboját, Björnligan, Karabela Çetesi, 庀兄弟, Gerombolan Siberat, Banda Bassotti.)

Distill This!

The last two big "breakthroughs" for TEnTs (translation environment tools) in the last couple of years have been real-time server-based processing and sharing of translation memory/terminology data and quality assurance features.

Especially this last new feature set has been adopted almost across the board in existing TEnTs, with Trados in particular putting a big emphasis on it. This may not be too surprising: many of the existing third-party tools like QA Distiller, Terminology Matters, and Error Spy have been focused primarily on Trados formats, making it easy for the Trados engineers to pick and choose the best features and integrate it into their own tool (see the QA Checker plug-in in TagEditor). So it's been sort of rough for the third-party tools to justify their existence.

This morning (providing a good excuse for sending this newsletter out late, I hope) I had a look at the latest incarnation of QA Distiller, and I was able to see some very valid reasons why one would invest in that tool.

For instance, there are certain checks that none of the other tools perform (as far as I'm aware) and that in my work as a translator often cost nerves and time, such as non-breaking spaces (between number and measuring unit, for example) or consistent quotation marks in the language you translated into. These and the many other checks can easily be performed on Trados ttx files, but more importantly -- and this was the most convincing feature -- on translation memory exchange TMX files. This is indeed super helpful because it is not limited to one tool -- TMX files are supported by virtually all TEnTs -- and it simply makes a lot of sense for me as a translator to QA and fix an external TMX file before using it for my translation project and possibly multiplying the problems throughout my translation project.

It's important to realize that there are some limitations. For instance, you can't at this point run a QA check on a TMX file that is larger than app. 10 MB, but that can be prevented by splitting larger TMX files into smaller files. In addition, the TMX feature is not available in the lower-priced freelance edition.

I called the developers at Yamagata and asked them whether they would be interested in doing something for Tool Kit readers. They agreed to give away one Professional edition (value $1000) and two Freelance editions (value $250 each), and they left it up to me to decide whom to give this to. So here's the deal: the first buyer of my Tool Box book after this newsletter goes out gets the Pro edition, and the next two will each receive a Freelance edition. If you already own the book, feel free to purchase a copy as a gift for a colleague! ;-)

We've added two Flash-based presentations of TEnTs to TranslatorsTraining.com: Swordfish and Cafetran. Chances are that you haven't seen these two tools in action and most certainly have not had a chance to compare them to the 13 other computer-assisted translations tools that are also covered on our website. If you've already bought a subscription, make sure to come by and view them. And if you haven't had a chance to purchase a subscription, now is the time.

Piotr Graff mentioned in an email that there is a "an even cheaper way, namely free, for those who own Acrobat 7 and 8" to OCR PDF files that are based on images rather than text (see the article on this in the last Tool Kit). You can find that option under Document> OCR Text Recognition, and though I had been aware of this for a while I had never really given it a fair shake. After Piotr's email I ran some tests, and I have to say that I was disappointed. After I pondered it a bit more I realized why the result was disappointing: it really isn't made for translation purposes, and for what it's supposed to do it may work just fine.

Here is what it does: once you select the OCR command, Acrobat re-orients your pages to a top-down orientation if they are sideways and then applies the OCR process internally so that the characters become editable within the PDF. The results in the new version of the PDF look surprisingly good. The frustration sets in when you then copy the text into a text editor or Word or save the PDF file in an editable format, because many characters are then internally mapped to incorrect matches. This is where tools such as Nuance's PDF Converter and Abbyy's PDF Transformer perform MUCH better -- they export the content straight into an external format rather than first performing an internal OCR. One positive thing about Acrobat is that the language selection of the target text is easily selectable (through the Edit button in the OCR dialog) -- something that Nuance's product cannot do.

And to clarify one issue that some readers expressed confusion and concern about after the last newsletter: the two PDF converters mentioned above are OCR applications that apply the same OCR capabilities to PDF files that their full-fledged siblings apply to scanned text. It's just that they do it only to PDF.

I'm sorry -- I realize that I've been writing a lot about Outlook, and that this is relevant only to those who use Outlook. And yes, I know that there are worthwhile alternatives. (In fact, I thought an esteemed colleague of mine put it well when he said the other day that he sees little choice when it comes to MS Office programs like Word, PowerPoint, and Excel, but Outlook is the one program he consciously does not use because there are true alternatives.) Be that as it may, and even despite having vowed to myself that I would not write about the Outlook-plugin Xobni (pronounced zob-nee) after all the mainstream media hype last week, I just have to introduce you to Xobni. It is simply so cool! Really.

Like Lookout (with which it shares its backwards spelling: Outlook - Lookout, Inbox - Xobni), whose technology is now integrated into Outlook 2007, it indexes your emails and attachments for immediate lookup. But unlike Lookout or any other indexing tools that I am aware of, it also analyzes the emails, automatically lists all correspondence with the currently selected person, finds all related users or phone numbers related to that person, shows how many emails you have exchanged and how this person is ranked among your emails correspondents, calculates how long it takes you to respond to emails from that person, etc., etc.

If you live in the U.S. and you've seen the advertisement from the Men's Warehouse, you'll understand this line: You're gonna like the way it looks. I guarantee it.

The only drawback I have seen so far is that Outlook takes much longer to close down, thus essentially amplifying an existing problem.

But then, check out the beautiful price: nada.

You're all familiar with Word's Format Painter (the little paintbrush icon). You can click on or select any text in your document, select the Format Painter, and then copy the formatting of the selected text by highlighting another block of text. As an extension of that functionality, you can also use the same procedure and double-click the Format Painter icon. After being double-clicked, the icon remains activated and the desired format is available to you until you press the ESC key.

Every reader of my Tool Box book knows this little trick, but here is something I just learned about Word last week: the very same feature works with the Highlight button. You can select some text, double-click the Highlight button, and then continue to select text that will immediately be highlighted without having to press that button again. Press the ESC key to exit.

And if, like me, you've always wondered how to batch run Word macros on many files (i.e., perform repeated processes on a number of files all at once), here are the instructions for that. Good stuff.

For most experienced computer users, how to control the programs that are started automatically when you boot your computer is old news. But there is a relatively new and completely free program offered by Microsoft for any version of Windows that gives you a fairly good overview of what these culprits are that make your computer so sluggish: Autoruns for Windows. There are many manual and other ways to achieve the same purpose, but what I like about this program is that it is really very easy to install and use, and it gives some information about what the processes and programs do that are started automatically. And if the explanations sound too cryptic, you can always check this webpage.