I attended Localization World London both as a guest speaker on what I call an upsurge in machine translation, almost a “transition frenzy” towards...
3 min read
07/01/2010
Sometimes one comes across some free advice that solves old problems, throws new light and confirms the fears. As SDL software becomes as ubiquitous in the translation industry as Microsoft in the OS world, it really takes effort and encouragement to yell to the thousands of users (newcomers as well as established corporations) that there are other possibilities out there, away from the forces that drive the markets (and our pockets). We are going to deal with a CAT tool which, in my opinion, does not get the limelight it deserves. Let's see how Deja Vu can be used for Trados projects. Trados/SDL software has become ever so present in our everyday translating lives (through very powerful and very effective marketing, among other techniques) that it is hardly imaginable why translators and corporations would use anything else unless they were forced to, or unless they were serving direct, small clients who really still make up a large share of our industry. Even then, fear of the unknown or a lack of a research spirit that might actually benefit the LSP itself keeps the traditional tools as the monolithic point of reference. I often wonder why LSPs (and translators) do not experiment with the wider choice of tools of customizable tools in the market. Perhaps one of the reasons is the learning curve. Linguists have spent an awful long time learning how to use the standard tools, so why bother with minority, geeky alternatives. Another is the lack of support or at least the lack of a user community and forums where you can exchange tips and tricks. Again, SDL Trados Forums (with parallel learning programs, "qualified status" and webinars) are ubiquitous compared with other tools. Nothing wrong with it, as it must be acknowledged that Trados and later SDL brought order to an otherwise chaotic and fragmented industry lacking investment and a drive for new technologies and applications. The downside, however, was years of a TM-dependency syndrome. One sometimes wonders what might have happened if the push for technology had gone to the perfection of machine-translation applications, rather than waiting for the next version of a software which solved problems with tags and compatibility with 3rd-party applications. Too frequently, I still find clients (even large corporations) fixated with the idea of Trados as a money-saving tool, disregarding all other options. Interestingly, they always have one more than one vendor for safety or for capacity reasons. But all the vendors are either forced to use the same tool or put in a situation where it makes no sense to use another tool. If we add the fact that QA tools and processes (not recommended) are basically reduced to the ones integrated in the translation tool or a handful (really a handful) of tools in the market, we see a chain where innovation is hardly nurtured. From a corporation point of view, this is nonsensical. I spent a large amount of time working for and around the automotive industry, where corporations choose suppliers for many reasons: quality systems, delivery times, prices, flexibility. Each one achieves this in different ways and all technologies are welcome as long as the final product (doors, tyres, springs, glass, handwheels) meets the company standards. There is a pressure from the car manufacturers towards the suppliers in terms of performance and price of course, but the former would never dream of imposing the best tools to manufacture the items. Quite the opposite, it is up to the suppliers to come up with better ideas and efficient processes - it is expected from them. Thus, you find a number of specialist firms manufacturing several parts for car companies and for several of them, too. One might argue that modern cars share a large amount of common parts and that the push is precisely towards the use of standard parts, but this still an objective rather than a reality. The case in the translation industry is practically monotheism. Let me quote Jost Zetzsche from his edition 152 - it cannot be expressed in better words
we always have a choice not to become part of new developments in technology or processes, but our decisions carry consequences that might influence the way this technology or this process develops in the future -- almost independently of whether we embrace it at some later point.This is the catch 22 we face as language vendors. We can opt out of the mainstream tools and technologies at our own risk (being left behind, being incompatible). Self-developed solutions and technologies, or the use of alternative technologies are often discouraged by the translation buyer, who is weary of experiments. Compliance with TMS and CMS is the norm. With the exception of machine-translation (and MT had been around for a long time, in different shapes until the recent Statistical eclosion), little has changed in the TM approach to translation. In reality, we must know and use the mainstream technologies (from PC-based applications to translation software), even if we just do so as a point of reference whence to build other solutions. By doing so, we are both supporting the mainstream and depending on it. That is why it is very refreshing to see sites offering alternative knowledge of how one can use other tools (perhaps more efficient in context) without disrupting the client's workflow. Some freelancers are supporting Deja Vu as a substitute of Trados. I would recommend anyone who is serious about CAT tools to get in touch with Atril and ask them how they can save money and time with their tool, and even package Trados and Studio files perfectly.
Next time you think languages, think Pangeanic
Your Machine Translation Customization Solutions