Extended Property for Entries plugin (serendipity_event_entryproperties) is required for this theme, which is not installed or is inactive. Please install the plugin to fully utlize this theme.
- November, 2014
"What do you mean 'it's identical'?"
One of the main marketing arguments for buying a CAT tool is that it is capable of "translating" files that contain a lot of repetition very quickly, thus saving you time. This statement is basically true, but it ignores the fact that it may still be necessary to check the "100% matches" and other very close matches to see if they fit the context, which also takes time and calls for great care on the translator's part.
Kilgray has made this easier for users by introducing the concept of a "101% match" in memoQ, which simply means that the segments directly adjacent to the 100% match are also taken into account; if they match up, too, then you have a highly dependable 101% context match.
Matches get copied – or "propagated" – throughout the text you are translating if you have set memoQ to do this. The setting is accessible via "Translations" > "Auto-Propagation..." in the main menu bar at the top of the screen when you have a document open in a project, as the following screen shot shows (auto-propagation is the fourth item down):
If you click on the item, the following dialogue will appear:
As you can see, you can set memoQ to propagate translations of identical segments in a particular direction: (a) "only forward", i.e. downwards from the cursor or (b) "backward and forward", i.e. upwards as well as downwards (= throughout the document you're translating). Various other settings can also be made.
Translations that have been propagated, i.e. copied to "identical" segments, are indicated by a dotted green arrow pointing downwards:
3. No go!
Before you choose an option, though, you might want to ask yourself if auto-propagating translations of "identical" segments is actually a wise idea; there are cases where it isn't, namely if a source segment needs to be translated in different ways, depending on its position in the overall text (= the context).
I encountered this situation in a recent translation in which a short source segment was used as a heading in one place and as a bullet point in another, which called for capitalisation in the first case, but not in the second one. It's also in cases like this that you might want to pick different translations for the same source text. But beware! If you set memoQ to propagate "identical" segments automatically upwards and/or downwards, it will insert one translation wherever the source text occurs, regardless of its contextual function.
To prevent this from happening, you need to de-activate auto-propagation altogether, after which you can handle seemingly identical segments on an individual basis rather than as a group. To do this, just unclick the box at the top, next to "Allow auto-propagation":
When you decide to re-activate the function afterwards (don't forget!), just go up to the menu again, select the item, click on the box again and a tick will re-appear. Close the dialogue and then memoQ will be set to propagate identical translations automatically again. (It would be nice if auto-propagation could be switched on or off with a single click on an icon in the main toolbar one day...)
images: my own screen shots & courtesy of Kilgray
Related information on the topic
- Kilgray's online help documentation on auto-propagation
- More details about the auto-propagation settings dialogue
And a quick look over the fence...
- Changing SDL Trados Studio 2014's auto-propagation behaviour (a blog post by Jayne Fox)
- Auto-propagation and "one-to-many" matching in OmegaT
- September, 2014
Over the last few weeks, I've been busy working on a big translation and editing project for a customer. There were a lot of Microsoft Word™ files to translate, and during the project, I wanted to measure the length of the translations in order to issue a preliminary bill for the work that had been completed by then.
Obviously, I could have opened each file in Word and then measured it using the program's own word-count feature, but that would have taken ages to do for so many files. What I really needed was a dedicated program for counting the amount of text automatically in a number of files, and more or less simultaneously. What's more, I wanted it to go a step further and calculate the figures I needed for invoicing: the net amount to be invoiced, the amount of VAT due and the grand total.
This is where dedicated text-counting programs come in. One of these tools is AnyCount, a program with three different versions made by Ukrainian software producer Advanced International Translations, aka AIT (also the developer of Translation Office 3000, in which it has been incorporated). As its name implies, AnyCount can count a large number of file types (I made it 36 in the latest version of the Enterprise edition, v8.0.8, including images, HTML, XML and exchange formats like .mif for FrameMaker™ files). Needless to say, this range also covers Microsoft Office™ formats (PPT, DOC, XLS and the more recent XML-based ones), but PDF and OpenOffice™ formats have been included as well, which makes AnyCount very versatile. Even Microsoft Visio™ and compiled help files (CHM) can be processed. A full overview is available on AIT's website (click here).
The workflow with AnyCount
Before you measure a file with AnyCount, you can make a number of settings in it to specify what gets counted and what doesn't. Under "Settings" in the menu bar, you can say whether headers and footers should be included and even how often (once per page or once per section). If there are footnotes or end notes in the file you want to count, you can also specify which of these it should cover, as the screen shot on the left shows.
Objects that are embedded in files, such as Excel tables in Word or PowerPoint documents, can be measured, too, along with a number of other (albeit rarely needed) options like shapes and hidden text.
Under "Settings", you can also choose the unit of text you want to be measured, a feature that includes words, characters with (or without) spaces, lines and pages, and define how many units make up a line or page, for example:
Select the files you want to measure under "Files", check you've made the right settings and then count the files by clicking on the "Count!" icon in the menu bar.
The whole workflow is summarised by this screen shot from the interface:
Putting it to the test
It took AnyCount less than two minutes to count the number of characters in 16 Word docx files and calculate the overall figure and resulting cost. You can either process this data yourself or get AnyCount to export it to a file format like Excel to make a printout. If you have configured a template for invoicing in the program, you can even get AnyCount to create an invoice automatically using the data. It can save you a lot of time by automating such data-processing steps.
What I like about version 8 of the tool is that...
- it's simple to use
- it measures the amount of text in files quickly and seems accurate (I want to check this more closely, though)
- the interface is attractive (a nice shade of blue!) and intuitive to work with
- you can configure a good number of settings yourself to suit your own way of counting
- and it lets you output the results as reports in a number of common formats such as Excel, HTML or Word (these can also be printed out without much editing).
It's also reasonably priced at 49 euros for the basic edition, called 'Personal Standard', (or less during promotions), 79 euros for the 'Professional' edition (which supports more file formats) and 95 euros for the full-feature 'Enterprise' version (which is the one that counts text in images – four different kinds!). AIT's technical support is also friendly and responsive. AnyCount 7.0 is available in 15 different language versions, while version 8.0, which is relatively new, is currently available in 11, English being the default. You can pick another language for the interface during the installation process:
PractiCount is another rival product I also use, but the customer service its maker (Practiline Software) provides is not as good in my experience (they are slow to respond to enquiries) and the tool doesn't include the option of counting text in images, which AnyCount does. Another program I know of on the German market (where I'm based) is called TextCount, which is available in English and German. You'll find other packages on the internet, too, which have interfaces in other languages. You can find out more about these alternatives by reading the posts on word-count software on one of Proz.com's user forums, for instance (see the link at the end).
Try them out and see what you think. Tools like these are practically a necessity for a busy translator.
images: my own screen shots from AnyCount 8.0.8 and by courtesy of AIT
- July, 2014
Some of you may already have heard of Heartsome, a firm that produces computer-assisted translation software. Its two main products these days are TMX Editor and Heartsome Studio, a suite of translation tools. The company is based in Hong Kong and has partly been catering to the Chinese-language market in Asia and partly to speakers of English. What's unusual about it is that it's about to close its doors for good. And it's making its products available for further development in an Open Source environment.
Heartsome says it's doing this in the hope that other software developers will pick up where its own developers left off and make the two tools more helpful to translators and agencies (see the farewell note on their website [which may get discontinued]).
Development and technical support at Heartsome is going to stop tomorrow, on 31 July. This is an unusual move in the CAT-tool industry, or at least one that doesn't happen very often. But like any business, if the sales figures just won't offset your business costs, you can only keep going for so long before drastic action is called for. Other companies in the field have seen bad times, too (like Atril, the makers of Déjà Vu), but they may manage to consolidate their business and carry on (which Atril did with a French financier's help).
I'm sorry to hear about this step as the more producers there are on the CAT-tool market, the better, at least in terms of competitive product development and pricing in the interest of users. It will be interesting to see what becomes of Heartsome's products in the future. I also hope the Open Source move is a positive one and that competent developers will step in and not let development stagnate.
Anyone who wishes to try out Heartsome Studio 8.0 and/or TMX Editor 8.0 now (à la Heartsome) can do so by downloading them from the GitHub site Heartsome has set up (see below) or from other trustworthy software platforms (even so, please make sure your PC is well protected from web-based viruses before downloading the packages; the software links are sound as far as I can tell, but you never know...).
The products can run on several operating platforms: not just Windows and Mac OS, but also Linux, for example.
I haven't tried out Heartsome Studio yet, but I know it can create bilingual XLIFF files, which can be easily modified for importing into other CAT tools (such as memoQ and SDL Studio 2011 or 2014). For more details about the Heartsome Studio suite, check out the features listed on GitHub.
I'll have a go when I get a chance and see what it's like in terms of usability. As for TMX Editor, that's a useful program you can use to edit and modify TMX files created by other CAT editors if their own TMX editing features are lacking. Why not try them both out and see what you think?
images: © Heartsome
- blog post in German by Torsten Rox-Edling: "Heartsome wird Open Source"
- Dominique Pivard's short e-training video on how to translate using Heartsome Translation Studio 8 (it lasts 8.5 mins)
- June, 2014
Recently I came across a reference to a book on editing that caught my attention, partly because I hadn't encountered many comprehensive guides on editing at that point and partly because this one was specifically aimed at translators. It turned out that the work had been around since 2001 and was now in its third edition (issued by Routledge in early 2014), so it was obviously popular and had been updated, too).
"Revising and Editing for Translators" is written by Brian Mossop, a Canadian who worked for the Canadian Government's Translation Bureau for many years and now teaches editing/revision and translation at university level. The author's considerable experience of revising translations and teaching students and teachers alike about revising and editing is reflected in the clear structure, real-life examples and broad scope of this work.
Routledge's edition of the book is 244 pages long and divided into 14 chapters. These are followed by six appendices (e.g. on assessing quality and grading texts) and a list of bibliographical references and other books and articles for further reading, plus a helpful index. There are a number of practical exercises and tips for further reading at the end of each chapter, which relate to the subject matter covered. This way of presenting material makes the book suitable for self-study as well as classroom use.
What I like about this work is its clarity: the language the author uses is straightforward and lucid (not academic and dense), the chapters are structured well and he employs plenty of examples to make his points understood. He also illustrates different kinds of attitudes and approaches to editing/revising, i.e. proscriptive v. liberal, without dictating the stance the reader should actually take.
I also like the amount of differentiation Mossop uses, which makes it clear how many different levels there are to editing and how many factors play a role in the choices editors make (cf. chapter 2, "The work of an editor", chapter 3, "Copyediting" and chapter 4, "Stylistic editing", for example); basically, chapters 2 to 7 all make this point.
Chapter 8 is particularly interesting in my view as it's concerned with software tools that editors and revisers can employ:
- internet searches to check terms and phraseology using search engines like Google
- looking for definitions of terms online
- using bilingual databases like Linguee and WeBiText and online translation-memory programs
- using editing features that word-processing programs offer (spelling and grammar checks, find & replace, displaying changes, adding reviewer's comments, comparing different versions of documents, etc.)
Mossop also makes a clear (albeit personal) distinction between editing and revising at the beginning of the book and consequently divides the work into two sections on each area. Chapters 2 to 7 are on editing, while 9 to 14 are on revising. In a nutshell, he takes editing to mean "reading a text which is not a translation in order to spot problematic passages, and making any needed corrections or improvements" (p. 29). As for revising, he regards this as a task "in which [translators] find features of the draft translation that fall short of what is acceptable, as determined by some concept of quality" (p. 115).
In chapter 10, he discusses 12 parameters that play a role in revision, including accuracy, completeness, logic, facts, page layout and even typography (i.e. the use of bold, italicised or underlined text, capitalisation and colouring). Chapter 11 covers degrees of revision (from "intelligible" to "polished"), whether or not full or partial checks should be done and the risks inherent in spot checking. Chapter 12 is about the actual revision procedure (e.g. which steps to take and in which order) and what you can do about any unsolved issues.
There's a lot more to the book than I can write about here. In short, I'd say it's essential reading for any translator, not just for editors and copywriters, since every translator has to read their own work through and edit (or "revise") it themselves before sending it off to the customer. I'm surprised I only discovered the book by chance, but that may be because it used to be published by a very small specialist publisher (St. Jerome Publishing); perhaps word will spread faster now that Routledge is backing it. (Click here for details about the book.)
image: my own screen shot of the title page taken on Routledge's website
- May, 2014
Kilgray, the maker of memoQ, my main CAT tool, is going to be holding a series of virtual events that might interest you on Wednesday 28 May (just five days from now). In fact, this is their third "Virtual Conference" on Proz.com, and like past events, it's free for anyone to attend. All you have to do is register for it beforehand; you don't need to be a paying member of Proz.com to do so.
Here are some of the benefits of attending at least some of the events in Kilgray's view:
- learn first-hand about the firm's most recent development, memoQ 2014, which will be released after the conference [more on this in a moment]
- attend various sessions and download training videos [to watch at your leisure on your own PC]
- learn the ins and outs of memoQ
- hear about some useful tips & tricks
- chat with Kilgray’s support team
- take advantage of special offers on memoQ
- get some hints on how to become a well-employed translator [presumably from LSPs that employ memoQ] and on dealing with customers.
The programme of events includes various presentations (live and on-demand webinars), discussion panels and live chats with staff at Kilgray in which you can ask questions of your own about memoQ and the firm's other products for translators. The on-demand presentations are likely to be particularly interesting for intermediate and advanced users of memoQ:
The highlight at this conference, however, is definitely going to be the preview you can get of memoQ 2014, a revamped release which includes a number of new features and productivity enhancements. These include:
- project templates, which will save you a lot of time when setting up projects [this might sound old-hat to some, but it's new to memoQ]
- automated steps in a workflow, which can be set in your project templates to simplify certain operations during a project (e.g. pre-translation, X-translate or export to MQXLIFF or a two-column RTF)
- productivity improvements such as better tag handling, grammar checking (not just spelling checks) and presenting several documents in a single view
- automatic assignment of files to specific translators working on online projects
- the ability to process embedded files and localise images
- easier terminology import from third-party packages (Studio and STAR Transit)
- project-management integration with Language Terminal (tracking projects)
- and many other productivity boosters (e.g. a built-in stopwatch feature for gauging the time it takes you to edit assignments).
The Virtual Day is due to take place between 10 a.m. and 5 p.m. GMT (12 noon to 7 p.m. CEST), but may run a bit longer because of the live chats. Those participants who attend it will be able to watch the recorded webinars again afterwards for a certain period of time.
More on memoQ 2014 in a later post; like memoQ 2013, this version really does seem to help translators and make many operations easier. (You can also read about memoQ 2014 on Kilgray's blog and Kevin Lossner's excellent Translation Tribulations.) As mentioned earlier in this post, the version is due to be launched after the virtual conference (in June).
Bye for now,
P.S. A free webinar dedicated to the new features that memoQ 2014 includes is going to be held by Kilgray on 11 June (at 3 p.m. GMT/5 p.m. CEST). All you need to do is register to attend it (click here).
As of 10 June, you can download a copy of memoQ 2014 from Kilgray's website. Try it out for free for 45 days or simply upgrade to it if you already have a current licence.
To hear what users of the new version have to say about its strengths and weaknesses, it's a good idea to follow the discussions about it in the memoQ user group hosted by Yahoo! (You need to sign up to use the list, but it's free and can be very informative.)
images: courtesy of Proz.com and Kilgray; sources of information: Kilgray's recent announcement and pre-launch briefing
- learn first-hand about the firm's most recent development, memoQ 2014, which will be released after the conference [more on this in a moment]
Despite us being well into the computer era, dictionaries are still an essential language resource for any translator. The dictionary-making industry has had to move with the times, which has rocked the boat quite a bit at established publishers such as Langenscheidt, but even so, it has managed to come up with electronic alternatives to paper dictionaries – and equally importantly, various ways of paying for them. Electronic dictionaries are now available as apps for mobile phones and portable e-readers like the Kindle, as PC software and as Web-based applications, for example. This post is about the latter, which are offered as paid services to which users subscribe (software as a service, or "SaaS").
One of the online dictionary services I use is MOT dictionaries and is run by a Finnish company called Kielikone. This firm provides various linguistic services via the Web that cater to people with an interest in translation and proof-reading, including Web-based dictionaries (e.g. via MOT mobile), machine translation (MOT translation) and online proof-reading (MOT proofing):
In addition to these three groups, Kielikone also offers its customers a small number of language guides, e.g. on English and Swedish grammar and on writing and spelling Finnish correctly, which caters to needs in their local market in Scandinavia.
Rather misleadingly, MOT translation is actually concerned with machine translation – from one of nine European languages and (Mandarin) Chinese into English and vice versa. Kielikone realistically admits that its MT system is only capable of making a "quick" translation of a text and is best suited to translating technical texts, instructions and news, i.e. material in which the focus is on content rather than style. In other words, post-editing is necessary to polish up the engine's translations (another task we human translators get lumbered with).
Rather than relying on MT's statistical algorithms, however, I prefer to use several of Kielikone's online dictionaries to help me translate my work. A wide range of monolingual and bilingual works are available, each of which costs a relatively small amount of money to subscribe to (around 15 euros for three months or roughly 49 euros for 12, for example). I've signed up for three German-English dictionaries, viz. MOT Collins German Dictionary, MOT Langenscheidt Muret-Sanders Großwörterbuch Englisch and MOT Oxford German Dictionary, and have an annual subscription, which is rather better value than a quarterly one. These electronic dictionaries are all based on well-known printed versions and have been licensed to Kielikone. (Installable versions for Windows PC users are also available, incidentally.)
To use the dictionaries you have chosen, you just log on to the MOT server via your Web browser using the user data that go with your subscription. It's very fast (and the connections I've set up have always been totally reliable). Set the language for the browser interface when you first start using MOT – it can be in English, German, Swedish or Finnish at the moment – and then pick the dictionary you want to use from a drop-down list:
If you set the system to search all of your dictionaries at once (by picking "Multiple dictionary selection" and clicking on the round plus symbol and selecting the works to be used), then when you enter a search term on the right, the results of your search will be displayed in a vertical list of all the dictionary entries that were found. The hits actually look much like those in a paper dictionary, with the entry being on the left and translations of it on the right together with sample sentences showing how the words are used (i.e. with contextual information):
Different kinds of searches are possible, depending on which option you select:
Like CD-ROM dictionaries, searches for terms can be done very quickly online, but I'd say the real advantage of subscribing to a number of MOT dictionaries is that you can see which dictionaries come up with hits instantaneously and the hits all appear on your monitor at once; there's no need to make individual searches in each dictionary as a universal search is done. So a system of this kind can save you time. Plus the fact that no maintenance is needed at the user's end, no system updates are called for and there's no need for any troubleshooting if anything goes wrong – this is all taken care of by the company providing the dictionaries over the Net.
So there's no hassle at all. What's more, you can try out additional dictionaries whenever you like, with no obligation to take out a subscription for them. If you find them useful, you pay for them and they get included in your personal software "suite" for immediate use. And if you decide you want to stop using your suite for any reason, you can simply cancel your subscription.
I thoroughly recommend "MOT dictionaries" as a translation resource as it's convenient to use, flexible and very helpful in many cases. Apart from including general dictionaries, the range also covers several commercial, technical and specialist dictionaries and encyclopaedias. Here's a sample:
If you'd like to try out the MOT dictionaries, get in touch with the Sales team at Kielikone and ask for a free trial.
- More on Langenscheidt's restructuring programme (in German)
images: screen shots from Kielikone's website; logo and other website images by courtesy of Kielikone Oy
- April, 2014
The other day I happened to stumble across a job advert by SDL plc, a well-known producer of software tools for translators, one of which I reviewed here briefly last autumn. The position was being advertised via monster.de, the big job portal on the internet, and looked as though it was in nearby Munich, which was why it initially caught my attention. It turned out it was actually an internship as an English to German translator. On closer reading, I found that the post was over in Sheffield (in the UK) and that SDL expects the person who is fortunate enough to get the job to work for what I would call peanuts – the National Minimum Wage in Britain, a stunning £6.50 an hour (7.85 euros at today's exchange rate).
Here's the top half of the ad, showing an appealing picture of a young team of professionals around a laptop, presumably to put you in an enthusiastic mood:
Looking at the job description, it sounds like an interesting position for a qualified translator. You would be working with a variety of CAT tools (obviously, since that's what SDL produces) and doing translating and editing work that can be demanding (post-editing machine-translated texts). You'd also have a degree of responsibility because you'd be checking software to find any mistakes or problems with it, testing websites (presumably how well they work) and doing terminological work, which means building up and possibly verifying and editing terms in a terminology database, no doubt for other members of staff (= translators) to use.
You need an educational background in translation or terminology for this job, as SDL specifically requests, and very good skills in two languages as well:
What do you think about this advertisement? My first reaction was one of disbelief; I asked myself why anyone with suitable qualifications should want to undertake work like this for a pittance? After all, it's a full-time position, even if it's only for six months (apparently with no option of being taken on properly afterwards).
Is it reasonable for an employer to expect this kind of work to be done by an intern, do you think? I could understand them offering the internship for a shorter period, say three months, but six seems a bit steep to me under these circumstances. On the other hand, it's become quite common for young people, including recent university graduates, to try and get some work experience in business fields that interest them even if it means them not being paid anything at all – it's the experience that counts for them rather than their earnings. (Think of the media, for example – TV and advertising.) In that respect, this particular position could be regarded as a particularly attractive one since you'd be getting paid all the time in addition to being given the opportunity to learn a lot about modern translation technology.
Still, I must say I find it unsettling to see low-paid positions being advertised that require skills that not everybody happens to have. Although the person's manpower would be tied up for half a year and the work they'd do would benefit the employer's own web marketing, they'd only be offered a token salary. Something's not quite right there, is it?
- March, 2014
Seit dem 27. März läuft ein neuer Film über den Einsatz der Bundeswehr in Afghanistan in den deutschen Kinos. Besonders interessant für uns Übersetzer und Dolmetscher ist, dass er die Beziehung eines deutschen Soldaten zu seinem afghanischen Dolmetscher in den Mittelpunkt stellt.
"Bundeswehrsoldat Jesper (Ronald Zehrfeld) meldet sich erneut zum Dienst in das krisengeschüttelte Afghanistan und erhält mit seiner Truppe den Auftrag, einen Außenposten in einem kleinen Dorf vor dem wachsenden Einfluss der Taliban zu schützen. Dabei wird der junge Afghane Tarik (Mohsin Ahmady) als Dolmetscher zur Seite gestellt.
Jesper versucht mit Tariks Hilfe, das Vertrauen der Dorfgemeinschaft und der verbündeten afghanischen Milizen zu gewinnen – doch die Unterschiede zwischen den beiden Welten sind groß.
Er steht immer wieder im Konflikt zwischen seinem Gewissen und den Befehlen seiner Vorgesetzten. Als Tarik, der von den Taliban bedroht wird, weil er für die Deutschen arbeitet, seine Schwester in Sicherheit bringen will, geraten die Dinge außer Kontrolle."
Die Regisseurin Feo Aladag realisierte den Film an Originalschauplätzen in Afghanistan.
Der Film ist von aktuellem Interesse, da die deutschen Schutztruppen dieses Jahr – nach mehr als 10 Jahren vor Ort – aus Afghanistan abziehen. Sie hinterlassen ein unbefriedetes Land und zahllose lokale Mitarbeiter, die von den Taliban als Kollaborateure der Besatzungsmächte gesehen werden. Auch lokale Dolmetscher, die für die Bundeswehr gearbeitet haben, stehen jetzt in Gefahr. Manche von ihnen versuchen mit Mühe, nach Deutschland zu kommen, um wieder ein sicheres Leben führen zu können.
Weitere Infos dazu gibt es vielerorts im Internet, u.a. auf der Website der tagesschau und bei arte. Auch Übersetzer- und Dolmetscherverbände wie der BDÜ haben dazu Stellung genommen und gefordert, Dolmetscher aus Afghanistan aus dringenden humanitären Gründen aufzunehmen.
Der Film ist bestimmt sehr sehenswert. Einen Trailer finden Sie hier.
Bilder: © Wolfgang Ennenbach / Majestic Filmverleih. Textabschnitte: Pressemappe von Majestic Filmverleih.
Page 3 of 7, totaling 51 entries