• localization testing is much more than a linguistic and/or cosmetic review
• multilingual testing is actually a superset of unilingual testing
• outsourcing multilingual testing avoids multilingual headaches
• outsourced testing in general can be cheaper, particularly considering the peaks and valleys of testing requirements (localization vendors and test departments are both used to the fact that development is always late!).

Source language terminology management will continue to grow, finally waking up the C-level to the fact that words (concepts) are an asset. From there, it is a small step towards multilingual terminology management. As "content hypergrowth" occurs, there will eventually be a corresponding "content expertise void." Will this void be filled by traditional documentation departments or by trained terminologists?

Content management will become more widespread and less expensive. More and more systems are paying attention to multilingual content, but none so far have taken over the entire translation process, i.e. the CMS-GMS merge has not yet occurred, and is not looking likely in the near future. With the advent of standard Web Services for translation, it may well be that the CMS will simply provide connectors to such Web Services.

XML will continue to be the format of choice for new content (of any kind). XML, as some have said, is the new ASCII, the ASCII of the Web. In fact, XML is to ASCII as a form is to narrative text. Consider how many millions of forms exist: birth certificates, income tax returns, loan applications, etc. Forms are actually rather dumb in and of themselves, but they allow intelligent processing because the data they contain has been broken down into fields that mean something to the processing layer. Such is the power of XML: simply naming things, giving them structure. As a result, just about every area of human activity can benefit from XML.

Machine translation will continue to be used in more and more niche applications. Here also there may be an "MT expertise void," which may be filled with experts from the GILT industry. MT is no longer a competitor to the translator, but rather a technology that enables new projects that would not have been possible otherwise. MT does not take work away from the translator! Instead, it helps to grow our industry, albeit more in terms of applications than in dollars.

The new OASIS (Organization for the Advancement of Structured Information Standards) committee to define Translation Web Services is working on the right problem: delivering translation work from the customer to the vendor, and back. If the committee moves quickly, it will be very successful. GMS vendors should follow this development closely, as communication between customer and vendor is one of their benefits (and one of the most attractive to translation vendors).

Dreams...

"I say to you today, my friends, that in spite of the difficulties and frustrations of the moment, I still have a dream." - Martin Luther King, Jr.

The entire speech is short, historic and fascinating.

I certainly do not wish to demean the greatness of Martin Luther King's dream by comparing it to any of mine below (dreams come in all sizes, though). What I do want to point out is that the localization industry must be based on the respect of other languages, other cultures, other people. We also have a dream...

I dream of modern ergonomics being put to use to define translation tools that are truly efficient for the translator, for all jobs (not just for the next version of the same thing). The cost of hardware has become so low, it is now intuitively obvious, without an ROI, that it is hardly a factor anymore. Two screens? voice input? graphic tablets? touch screen? or simply better tools for the single screen?

I dream of a complete and modular automation of localization workflow, where each part of the system has well-defined APIs (including Web Services?). I dream of a day when there will a standard set of bi-directional filters for content formats, so our industry can stop writing parsers!

I also dream of the Semantic Web.

And One Hallucination...

In a truly Huxleyian Brave New World, authors and translators actually work together to produce multilingual content, producing better quality results, more quickly and at a lower cost. Better yet, they are all paid as the creative professionals they are. And the rest of the world is pneumatic.

Have another happy pill...