Riccardo Schiaffino writes the About Translation translation blog, which was selected by BabLa as one of the top 25 language professional blogs, and also as one of the top 100 language-lover blogs being published today. When you look around the little world of translation blogs on the web (but what a world it is!), it doesn’t take long to discover the About Translation blog, whether it’s referenced, or on somebody’s list of best translation blogs. Oh, and he’s a truly experienced and specialized translator for English into Italian legal work!
We would like to thank Riccardo for interviewing with us, and for providing such insightful and thoughtful answers!
The interview covers a variety of topics, like the future of machine translation, the nature of “word people,” and the differences between the translation industry in Italy and in the USA.
We requested a photograph of Riccardo sweating over a translation on an Olivetti typewriter, but we’ll have to settle for a picture of Riccardo working at his favorite CAT tool and looking like a happy translator!
Q. How did you get into the translation industry?
A. That was many years ago, towards the end of the 70’s. My first year of University (where I started to study geology) had not gone well – my fault mostly, for choosing a subject I was not all that interested in.
I did not know what to do, but continuing with geology was not an appealing choice. A friend suggested that, since I knew English well, I should try the entrance examination for the School for Translators of Trieste University. I tried, and was accepted. A few years later, after intensive studies, I graduated with a degree in translation. After my year of compulsory national service in the Italian Army, I started working as a freelance translator. I have worked in translation, either as a translator or as a translation manager, ever since.
Q. You have a nice translation blog. We’re just starting. Recommendations?
A. Yes: decide who your audience is going to be, and write posts that are interesting for them. Don’t try to make it just a vehicle for your marketing: it would show. For more suggestions about how to start a blog, please see my presentation “Blogging 101” (http://www.aboutranslation.com/p/cta-blogging-101.html).
Follow-up comment: Thanks! We are trying, that’s for sure!
Q. Tell us a little bit about your services and your specialty.
A. We are a small partnership of four experienced translators, and we mainly work for larger translation organizations. We specialize in a limited number of fields: medical and health care (my partners, mostly), legal (myself), business software (all of us: we used to work together in the translation department of J.D. Edwards, a major business software company).
Besides translation we also offer localization testing, technical writing, and translation quality assessment.
Q: If you could do it all over again – becoming a translator, starting a blog, working for yourself, etc., would you? What would you do differently?
A. Becoming a translator: I would do it again – I love this profession! I might have some doubt if I were just starting out: our profession is changing so rapidly that I sometimes think it’s unclear where it is headed, and what the role of most translators will be. But I think in the end I would still choose to study translation and work as a translator.
Working for myself: I’ve worked independently most of my life – the longest stretch where I was an employee were the nine years I worked for J.D. Edwards. Those nine years were a valuable experience: I think I learned and grew a lot, professionally. It was a rewarding period for my personal life: because of that job that I moved to the States, and that has made a big difference both personally and professionally (working here as a freelancer is much better than doing the same in Italy).
Follow-up question: That’s really interesting! Can you explain a little bit about the difference between freelancing in Italy, and freelancing in the USA?
A. The main differences are:
- That in Italy taxes are higher for translators (and for everybody else, of course);
- That translators often charge (or accept) lower rates (despite Italy’s being a country with a high cost of living); and
- That Italian translation companies (and Italian companies in general) pay much later than companies in the States. In Italy you can often see terms of payment of ninety days, and actual payments may be even later.
About rates, here in the States companies ask you what your rates are – in Italy they tell you what they are going to pay you.
But please bear in mind that I’ve been living and working in the States for almost twenty years now, so my information may be out of date.
Starting a blog: Yes, I think I would also do that again – perhaps with a little bit more preparation at the start (for example, I didn’t get a specific domain for my blog until much later).
As for what I would do differently – there is so much that I could have done differently, from studying different languages, to making other choices at various points in my career. If I hadn’t accepted the offer J.D. Edwards made to me back then, I would have had a different career and life. But I’m happy with my life, and so I would not want to change anything major.
Q. Why do some translators seem to embrace technology, and others not? Do you think that it’s related to the industry being made up largely of “word people”?
A. One of the reasons can be that many translators study languages and the humanities, consider themselves “word people”, as you say, and feel some fear of new technology. Our profession has seen great changes in a short time: I started working as a translator using just a typewriter, and progressed from that to a standalone word processor, and then to a growing series of personal computers. We have gone from typing on paper (and white-out fluid), to word processors, and now to computer-assisted translation tools: I was among the first to use them (I still have somewhere my first installation disk for Déjà Vu – serial number # 27), but I know several translators that dug their heels in and have refused to even consider translation tools. In certain fields, it doesn’t really matter. But for most commercial translations the latest tools make us more productive and help us drafting more consistent translations. I believe they help us to be better translators.
Q. Your entry regarding “translation management software,” which is what we do, seemed frustrated. Where does this frustration come from?
A. It comes from some specific experience with an online tool used by a translation company who had contracted me (and a bunch of other translators) for a large software documentation job. After using more advanced tools, working in that online tool felt almost like going back to a mechanical typewriter. The translation memory suggestions were rudimentary, concordance searches did not work properly, the tool was slow and buggy, and though it allowed several translators to work on the same project at the same time, I think its benefits were more than offset by its drawbacks.
I also prefer to keep control of the tools I use, and choose those that work best for me.
I’m also detecting an annoying trend among translation companies, each running their own management software and portal. So you have to keep your own accounting, but then also log in different portals for different customers to create invoices, accept jobs, download files, and so on (and sometimes to perform the job itself). Each with a different tool, different portal, and different ways of doing similar operations. This way, translators might help translation companies save time, but at the cost of wasting more time themselves. Very annoying and very frustrating.
Comment: Yep, that’s us being annoying! However, we would like to interject that we try to make our translation editor user friendly, and translator invoices are made automatically, so there’s no work to be done by a translator there.
Follow-up question: That’s an interesting take – that there isn’t really a time-savings, but rather a shifting of time spent. Do you find that the industry is shifting to direct company-translator relationships, or are translation companies/intermediaries becoming more prominent?
A. I believe that translation companies serve a real need, at least for certain types of projects, for example, large multi-lingual jobs. On the other hand, I believe that any business that is conscious about the importance of quality in their translated documents will, at some point, need to decide for themselves if their interests are served by a large translation company, who may assign jobs to the cheapest translators, or if they shouldn’t instead work directly with a selected group of professional translators.
I’m not saying anything other people have not said before, but I believe that the world of translation is dividing itself between high-end translation, where professionals deliver work of outstanding quality and often work directly with their customers, and low-end commoditized translation, where jobs are assigned most often to the lowest bidder, and nobody seems to care about quality, or able to deliver it. This is the world where translators work for a few cents a word, and feel threatened by machine translation.
Translators who at the moment work in an intermediate zone between these two are feeling the pressure on rate brought about by the spread of cheap translation, and have to decide whether to try to compete on price with the people at the bottom rung of the market, or try to improve their quality and compete for better paid (but much more demanding) job in the upper echelons of the market.
Q. Do you think the industry lacks competency or professionalism in some areas? Which areas?
A. Ours is a profession that most people that know a foreign language think they can do… and usually that’s not true. Knowing a foreign language is just the first precondition – not the only thing needed to become a translator. Most people, for example, don’t know how to write well in their native language, and, as translators, we are writers. And many translators don’t realize what it is that they don’t know, and so they merrily translate in fields for which they are unprepared. It has been said that to develop a real competency in any demanding field takes about 10,000 hours. That’s at least five years of full-time practice – and yet there are so many people that believe they are accomplished translators right out of translation school, or (worse) without having studied translation at all. It is one thing to know a language well enough to speak it fluently when traveling abroad. Quite another to know it well enough to be a writer or a translator.
This lack of competency and professionalism is by no means limited to translators – it goes from the successful freelancer that later fails as translation agency owner, to editors who cannot distinguish between their personal preference and actual errors. From translation software developers who may provide good tools but poor documentation, to machine translation evangelists, who, instead of describing the real benefits offered by their wares, make exaggerate claims, without a care for the real damage they are inflicting to our profession (and even to their own clients).
Follow-up question: Interesting take on exaggerated benefits of wares. Hopefully at Wordbee, we’re not exaggerating, or at least not too much! Is there a machine translation tool that you use sometimes? If so, in what cases do you think machine translation is useful? Any thoughts on “Google glasses” and the future of machine translation?
A. I’ve played a bit with Google Translate and Bing, and worked with a Texas translation company to set up the glossary and train a translation engine for a big project. I could see that an MT engine you could train with specialized terminology and rules can be better than the free engines available on the Internet, but, I think, not by much.
I do think there is a use for machine translation. Certainly for gisting (even without editing or post-editing), and probably for massive documentation projects where quality is not too much of an issue, but delivering a large volume of translated material is.
The problem is that “gisting” level translation is then sometimes used for actual documentation, and “documentation only” post-edited translation is used as actually publishable content. In those instances the results are gruesome.
I also think that the advances made by statistical machine translation are mostly done… Google has fed billions of pages in its engine, and I don’t foresee any further dramatic improvements there. Rule-based machine translation programmers have had decades to refine their products, and I don’t think they are going to squeeze many more improvements from their algorithms. There is still some room for growth in hybrid systems, but I don’t think that machine translation is going to get significantly better than what it is now, in any form.
This is normal, I think. Look at the word processor: the real difference was made when computers took over the work previously done by typewrites. Then it was a process of incremental improvements, with fewer and fewer real ones. For me the big difference was passing from an Olivetti typewriter to WordStar running under DOS. Another big jump (but less important than the first) was the first word processors running under Windows. Sure, MS Word kept on adding features, but has it really improved recently? I think that for 99% of what people do with a word processor, MS Word 2003 is just as good as MS Word 2013.
I think machine translation is in a similar situation, or that it will soon be.
Q. What is your dream tool, as a translator?
A. A new generation of translation tools that:
- Put some intelligence in fuzzy matching (the algorithms used by all the tools I know have not been improved for years), so that more importance is given to matching meaningful words, and less to matching grammatical words;
Follow-up comment: Do you know, this would make a great post for your blog – a comparison of the fuzzy match rates across different tools, to see which program was the most intelligent. If you’d be interested in writing such a post, we’d definitely be interested in participating!
A. I’ve been planning such an article for some time – but to get any meaningful data one would need to prepare the work properly: at a minimum what is needed is a properly designed test file.
- Closely integrate with a wide range of electronic dictionaries, and not only merely with the terminology tool provided with the program;
- Play nice with the file formats used by other tools: ideally, I should be able to receive a translation, translate it in my favorite tool, and deliver it to my customer, without having to either use the tool they want, or jump through hoops to convert files from one format to another.
Follow-up question: Well, Wordbee supports TTX, TMX, XLIFF, etc., and doesn’t generate any incompatible formats, but it’s true that some tools don’t want to be compatible. Do you think that companies who make their tools incompatible get a real market advantage, or do you think it’s mostly just insecurity?
A. I think that at this point they get a real market disadvantage, as translators become more knowledgeable, and try to stay away from tools that won’t allow them to work with formats generated by other tools. In that situation either you are the market leader, or your product is at a real disadvantage. Who wants to use the new super-duper tool X, if that tool won’t work with Trados files?