The Challenge of Using MT in Localization

We live in an era where MT is translating more than 99% of all the translation being done on the planet on any given day.

However, the adoption of MT by the enterprise is still nascent and still building momentum. Business enterprises have been slower to adopt MT even though national security and global surveillance-focused government agencies have used MT heavily. This adoption delay has mostly been because MT has to be adapted and tuned to perform better with very specific language used in specialized enterprise content.

Early enterprise adoption of MT was focused on eCommerce and customer support use-cases (IT, Auto, Aerospace) where huge volumes of technical support content made it a necessity to use MT technology to allow any possibility of translating the voluminous content in a timely and cost-effective manner to improve the global customer experience.

Microsoft was a pioneer who translated its widely used technical knowledge base to support an increasingly global customer base. The positive customer feedback for doing this has led to many other large IT and consumer electronics firms doing the same.

The adaptation of the MT system to perform better on enterprise content is a critical requirement in producing successful outcomes. In most of these early use-cases we see that MT is used to manage translation challenges when the content volumes were huge, i.e., millions of words a day or week. These were “either use MT or provide nothing” knowledge-sharing scenarios.

These enterprise-optimized MT systems have to adapt to the special terminology and linguistic style of the content they translate, and this customization has been a key element of success with any enterprise use of MT.

eBay was an early MT adopter in eCommerce and has stated often that MT is key in promoting cross-border trade. It was understood that “Machine translation can connect global customers, enabling on-demand translation of messages and other communications between sellers and buyers, and helps them solve problems and have the best possible experiences on eBay.”

A study by an MIT economist showed that after eBay improved its automatic translation program in 2014, commerce shot up by 10.9 percent among pairs of countries where people could use the new system.

Today we see that MT is a critical element of the global strategy for Alibaba, Amazon, eBay, and many other eCommerce giants.

Even in the COVID-ravaged travel market segment, MT is critical as we see with Airbnb, which now translates billions of words a month to enhance the international customer experience on their platform. In November 2021 Airbnb announced a major update to the translation capabilities of their platform in response to rapidly growing cross-border bookings and increasingly varied WFH activity.

“The real challenge of global strategy isn’t how big you can get, but how small you can get.”

Dennis Goedegebuure, former head of Global SEO at Airbnb.

However, MT use for localization use cases has trailed far behind these leading-edge examples, and even in 2021, we find that the adoption and active use of MT by Language Service Providers (LSPs) is still low. Much of the reason lies in the fact that LSPs work on hundreds or thousands of small projects rather than a few very large ones.

Early MT adopters tend to focus on large-volume projects to justify the investments needed to build adapted systems capable of handling the high-volume translation challenge.

What options may be available to increase adoption in the localization and professional business translation sectors?

At the MT Summit conference in August 2021, CSA's Arle Lommel shared survey data on MT use in the localization sector in his keynote presentation. He noted that while there has been an ongoing increase in adoption by LSPs there is considerable room to grow.

Arle specifically pointed out that a large number of LSPs who currently have MT capacity only use it for less than 15% of their customer workload and, “our survey reveals that LSPs, in general, process less than one-quarter of their [total] volume with MT.”

The CSA survey polled a cross-section of 170 LSPs (from their "Ranked 191" set of largest global LSPs) on their MT use and MT-related challenges. The quality of the sample is high and thus these findings are compelling.

The graphic below highlights the survey findings.

CSA Survey of MT Use at LSPs

When they probed further into the reasons behind the relatively low use of MT in the LSP sector they discovered the following:

• 72% of LSPs report difficulty in meeting quality expectations with MT
• 62% of LSPs struggle with estimating effort and cost with MT

Both of these causes point to the difficulty that most LSPs face with the predictability of outcomes with an MT project.

Arle reported that in addition to LSPs, many enterprises also struggle with meeting quality expectations and are often under pressure to use MT in inappropriate situations or face unrealistic ROI expectations from management. Thus, CSA concluded that while current-generation MT does well relative to historical practice, it does not (yet) consistently meet stakeholder requirements.

This apparent market reality validated by this representative sample is in stark contrast to what happens at Translated Srl, where 95% of all projects and all client work use MT (ModernMT) since it is a proven way to expedite and accelerate translation productivity.

Adaptive, continuously learning ModernMT has been proven to work effectively over thousands of projects with tens of thousands of translators.

This ability to properly use MT in an effective and efficient assistive role in production translation work has resulted in Translated being one of the most efficient LSPs in the industry, with the highest revenue per employee and high margins.

Another example of the typical LSP experience: a recent study by Charles University done with only 30 translators using 13 engines (EN>CS) concludes: "the previously assumed link between MT quality and post-editing time is weak and not straightforward." They also found that these translators had “a clear preference for using even imprecise TM matches (85–94%) over MT output."

This is hardly surprising, as getting MT to work effectively in production scenarios requires more than choosing the system with the best BLEU score.

Understanding The Localization Use Case For MT

Why is MT so difficult for LSPs to deploy in a consistently effective and efficient manner?

There are at least four primary reasons:

1. The localization use case requires the highest quality MT output to drive productivity which is only possible with specialized expertise and effort,
2. Most LSPs work on hundreds/thousands of smallish projects (relative to MT scale) that can vary greatly in scope and focus,
3. Effective MT adaptation is complex,
4. MT system development skills are not typically found in an LSP team.

MT Output Expectations

As the CSA survey showed, getting MT to consistently produce output quality to enable use in production work is difficult. While using generic MT is quite straightforward, most LSPs have discovered that rapidly adapting and optimizing MT for production use is extremely difficult.

It is a matter of both MT system development competence and workflow/process efficiency. 

Many LSPs feel that success requires the development of multiple engines for multiple domains for each client, which is challenging since they don't have a clear sense of the effort and cost needed to achieve positive ROI.

If you don’t know how good your MT output will be, how do you plan for staffing PEMT work and calculate PEMT costs?

Thus, we see MT is only used when very large volumes of content are focused around a single subject domain or when a client demands it.

A corollary to this is that it requires deep expertise and understanding of NMT models to acquire the skills and data needed to raise MT output to useful high-quality levels consistently.

Project Variety & Focus

Most LSPs handle a large and varied range of projects that cover many subject domains, content types, and user groups on an ongoing basis. The translation industry has evolved around a Translate>Edit>Proof (TEP) model that has multiple tiers of human interaction and evaluation in a workflow.

Most LSPs struggle to adapt this historical people-intensive approach to an effective PEMT model which requires a deeper understanding of the interactions between data, process, and technology.

The biggest roadblock I have seen is that many LSPs get entangled in opaque linguistic quality assessment and estimation exercises, and completely miss the business value implications created by making more content multilingual. Localization is only one of several use-cases where translation can add value to the global enterprise's mission.

Typically, there is not enough revenue concentration around individual client subject domains, thus, it is difficult for LSPs to invest in building MT systems that would quickly add productivity to client projects. 

MT development is considered a long-term investment that can take years to yield consistently positive returns.

This perceived requirement for the development of multiple engines for many domains for each client requires an investment that cannot be justified with short-term revenue potential. MT projects, in general, need a higher level of comfort with outcome uncertainty, and, handling hundreds of MT projects concurrently to service the business is too demanding a requirement for most LSPs.

MT is Complex

Many LSPs have dabbled with open-source MT (Moses, OpenNMT) or AutoML and Microsoft Translator Hub only to find that everything from data preparation to model tuning, and quality measurement is complicated, and requires deep expertise that is uncommon in the language industry.

While it is not difficult to get a rudimentary MT model built, it is a very different matter to produce an MT engine that consistently works in production use. For most LSPs, open-source and DIY MT is the path to a failed project graveyard.

Neural MT technology evolution is happening at a significantly faster pace than Statistical MT. To stay abreast with the state-of-the-art (SOTA) requires a serious commitment, both in manpower and computing resources.

LSPs are familiar with translation memory technology that has barely changed in 25 years, but MT has changed dramatically over the same period. In recent years the neural network-based revolution has driven multiple open-source platforms to the forefront and keeping abreast with the change is difficult.

NMT requires expertise not only around "big data", NMT algorithms, and open-source platform alternatives but also around understanding parallel processing hardware.

Today AI and Machine Learning (ML) are synonymous, and engineers with ML expertise are in high demand.

MT requires long-term commitment and investment before consistent positive ROI is available and few LSPs have an appetite for such investments.

Some say that an MT development team might be ready for prime-time production work only after they have built a thousand engines and have this experience to draw from. This competence-building experience seems to be a requirement for sustainable success.

Talent Shortage

Even if LSP executives are willing to make these strategic long-term investments, finding the right people has gotten increasingly harder. According to a recent survey by Gartner, executives see the talent shortage not just as a major hurdle to progressing organizational goals and business objectives, but it is also preventing many companies from adopting emerging technologies.

The Gartner research, which is built on a peer-based view of the adoption plans of 111 emerging technologies from 437 IT global organizations over a 12- to 24-month time period, shows that talent shortage is the most significant adoption barrier to 64% of emerging technologies, compared with just 4% in 2020.

IT executives cited talent availability as the main adoption risk factor for the majority of IT automation technologies (75%) and nearly half of digital workplace technologies (41%).

But using technology early and effectively creates a competitive advantage. Bain estimates that “born-tech” companies have captured 54% of the total market growth since 2015. “Born-tech” companies are those with a tech-led strategy. Think Tesla in automobiles, Netflix in media, and Amazon in retail.

Technology has emerged as the primary disruptor and value creator across all sectors. The demand for data scientists and machine learning engineers is at an all-time high.

LSPs need to compete with the global 2000 enterprises who offer more money and resources to the same scarce talent. Thus, we even see technical talent migrating out of translation services to the “mainstream” industry.

There is a gold rush happening around well-funded ML-driven startups and enterprise AI initiatives. ML skills are being seen as critical to the next major evolution in value creation in the overall economy as the chart below shows.

This perception is driving huge demand for data scientists, ML engineers, and computational linguists who are all necessary to build momentum and produce successful AI project outcomes. The talent shortage will only get worse as more people realize that deep learning technology is fueling most of the value growth across the global economy.

Thus, it appears that MT is likely to remain an insurmountable challenge for most LSPs. The option for an LSP to start building robust state-of-the-art MT capabilities in 2021 is increasingly unlikely. 

Even the largest LSPs today have to use “best-of-breed” public systems rather than build internal MT competence. Strategies employed to do this typically depend on selecting MT systems based on BLEU, hLepor, TER, Edit Distance, or some other score-of-the-day, which again explains why there is <15% MT-in-production-use.

As CSA has discovered, LSP MT use has been largely unsuccessful because a good Edit Distance/hLepor/Comet score does not necessarily translate to responsiveness, ease of use, adaptability of the MT system to the production localization use-case needs.

For MT to be useable on 95%+ of the production translation work done by an LSP, it needs to be reliable, flexible, manageable, rapidly adaptive, and continuously learning. MT needs to produce predictably useful output and be truly assistive technology for it to work in localization production work.

The contrast of the MT experience at Translated Srl is striking. ModernMT was designed from the outset to be useful to translators and created to collect the right kind of data needed to rapidly improve and assist in localization project-focused systems.

ModernMT is a blend of the right data, deep expertise in both localization processes and machine learning, and a respectful and collaborative relationship between translators and MT technologists. It is more than just an adaptive MT engine.

Translated has been able to overcome all of the challenges listed above using ModernMT, which today is possibly the only viable MT technology solution that is optimized for the core localization-focused business of LSPs.

ModernMT is the creation of an MT system optimized for LSP use. It could be used quickly and successfully by any LSP as there is no startup setup and training needed, it is a simple "load TM and immediately use" model.

ModernMT Overview & Suitability for Localization

ModernMT is an MT system that is responsive, adaptable, and manageable in the typical localization production work scenario. It is an MT system architecture that is optimized for the most demanding MT use-case: localization. And it is thus able to handle many other use-cases which may have more volume but are less demanding on the output quality requirements.

ModernMT is a context-aware, incremental, and responsive general-purpose MT technology that is price competitive to the big MT portals (Google, Microsoft, Amazon) and is uniquely optimized for LSPs and any translation service provider, including individual translators.

It can be kept completely secure and private for those willing to make the hardware investments for an on-premise installation. It is also possible to develop a secure and private cloud instance for those who wish to avoid making hardware investments.

ModernMT overcomes technology barriers that hinder the wider adoption of currently available MT software by enterprise users and language service providers:

• ModernMT is a ready-to-run application that does not require any initial training phase. It incorporates user-supplied resources immediately without needing upfront model training.
• ModernMT learns continuously and instantly from user feedback and corrections made to MT output as production work is being done. It produces output that improves by the day and even the hour in active-use scenarios.
• ModernMT is context-sensitive.
• The ModernMT system manages context automatically and does not require building domain-specific systems.
• ModernMT is easy to use and rapidly scales across varying domains, data, and user scenarios.
• ModernMT has a data collection infrastructure that accelerates the process of filling the data gap between large web companies and the machine translation industry.
• Driven easily by the source sentence to be translated and optionally small amounts of contextual text or translation memory.

ModernMT’s goal is to deliver the quality of multiple custom engines by adapting to the provided context on the fly. This fluidity makes it much easier to manage on an ongoing basis as only a single engine is needed.

The translation process in ModernMT is quite different from common, non-adapting MT technologies. The models created with this tool do not merge all the parallel data into a single indistinguishable heap; separate containers for each data source are created instead and this is how it maintains the ability to adapt to hundreds of different contextual use scenarios instantly.

ModernMT consistently outperforms the big portals in MT quality comparisons done by independent third-party researchers, even on the static baseline versions of their systems.

ModernMT systems can easily outperform competitive systems once adaptation begins, and active corrective feedback immediately generates quality-improving momentum.

The following charts show how ModernMT is a consistent superior performer even as the quality measurement metrics change over multiple independent third-party evaluations conducted over the last three years. 

None of these metrics capture the ongoing and continuous improvements in output quality that is the daily experience of translators who work with dynamically improving ModernMT at Translated Srl.

Independent evaluations confirm ModernMT quality improves faster with COVID data set on English > German in the chart below.

ModernMT was also the "top performer" on several other languages tested with COVID data.

In Q4 2021 the COMET metric is widely being considered a "better" score because it is more aligned with human assessments and also incorporates semantic similarity, and again ModernMT shines.

 If the predictions about the transformative impact of the deep learning-driven revolution are true, DL will likely disrupt many industries including the translation industry. MT is a prime example of an opportunity lost by almost all the Top 20 LSPs.

While it is challenging to get MT working consistently in localization scenarios, ModernMT and Translated show that it is possible and that there are significant benefits when you do.

This success also shows that when you get MT properly working in professional translation work, you create competitive advantages that provide long-term business leverage.  The future of business translation increasingly demands collaborative working models with human services integrated with responsive adapted MT. The future for LSPs that do not learn to use MT effectively will not be rosy.

A detailed overview of ModernMT is provided here. It is easy to test it against other competitive MT alternatives, as the rapid adaptation capabilities can be easily seen by working with MateCat/Trados or with a supported TMS product (MemoQ) if that is preferred.

ModernMT is an example of an MT system that can work for both the LSP and the Translator. The ease of the "instant start experience" with Matecat + ModernMT is striking when compared to the typical plodding, laborious MT customization process we see elsewhere today. Try it and see.

The Translation Market – Is it Really Understood?

I saw some interesting comments to a blog post by Kevin Lossner that I thought would be good to share with the community that reads this blog, as it raised some cogent points I thought. The comments basically talk about a larger more complex translation market than many of us might believe exists based on market research available. I do not claim to have real insight or knowledge of this larger translation market, but I am definitely aware that the largest translation initiatives in the world are generally overlooked by traditional market research e.g the many branches of the US government (DoD, NSA, CIA, FBI, DIA, State and even Commerce), the EU and I expect many of the clandestine “intelligence” operations around the world, especially amongst the G20 governments.

I would also bet that the really big, almost nation-like, Fortune 100 corporates also have captive and hidden translation operations that are buried and invisible within PR, Marketing and Investor Relations somewhere to translate the stuff that really matters or is really secret. (I would not be surprised if the people in these departments did not even know if a localization team exists elsewhere in the corporation.)  If it really matters, why would you ask Lionbridge or SDL (or any other large LSP) to translate it? definitely is something to ponder upon.   Surely it would be more likely to go to internal subject matter experts, or to trusted and elite boutique services that actually understand the subject matter of the material, and can protect the information with the same zeal and protective assurances as those who create it.  Imagine you are an oil company called ABCP and want to make sure that you look less culpable for a major accident caused by management insistence on moving ahead with a risky drilling project. I think the odds are high that the translators chosen to translate critical memos and communications and "put the right spin on it" before it is shown to regulators are going to be different from the ones that work for Lionbridge since it might save a few billion in damages that will have to be paid.

I also generally expect that specialists, i.e. translators with demonstrated subject domain expertise, will have a much brighter future than those who will translate anything that is within arms reach. Specialization means building subject matter expertise, which I think will matter more and more, and I for one would stay away from LSPs who do not specialize or have long-term demonstrated competence in a few select domains.

I find this discussion interesting also because I think that repetitive, low-value, short shelf-life, bulk (high volume) content is eventually going the way of PEMT or even raw MT, but there is a huge world of high value content that is unlikely ever to head that way until we reach the Star Trek Universal Translator levels of quality, which are not expected to be available till the 24th century. I actually think that IPO and many SEC filing documents (10K, Registration documents) and user manuals of any kind including nuclear machinery and medical equipment are fair game for competent and very specialized PEMT initiatives, but I would not use MT for anything that requires linguistic finesse or reading between the lines e.g. wedding vows, great literature, letters to the board/stockholders or poetry. Even in those areas where you have high volume and lots of repetitive and highly similar content, MT can work well only when real expertise is applied, and there is a real and active collaboration with translators and linguists who all want to produce an engine that will reduce future efforts.
 
These are some of the excerpted and unedited (by me) comments made by Kevin Hendzel at the blog post referenced above written in a more visceral style than the more careful elaboration in much greater detail on his own blog. I don’t agree with everything Kevin says about MT, but I think his views are generally based on deeper observations than “MT is crap” and I can appreciate that we have different views on this issue. (Excerpts printed here with his and Kevin Lossner’s permission.)

From my own viewpoint, it does seem that the localization industry/bulk translation market has long suffered from a “we’re the only game in town” problem. There’s an amusing story about SeaWorld (an aquatic theme park in the US) that goes a long way toward illustrating this exact echo-chamber problem that the localization industry and pure bulk-market providers seem to be perpetually trapped in. Occasionally you’ll see protesters outside SeaWorld holding up signs that declare: “It’s not SeaWorld, it’s PoolWorld.” The corporate entity SeaWorld telling tourists that these tiny, familiar pools constitute “the sea” does not make them the sea. The sea is immensely, incalculably larger and more complex.

The same is true of the translation market. Referring to the tiny pool you are familiar with (low-end bulk localization and translation) as “the sea” (the whole rest of the market) tends to distort one’s sense of the enormity of the sea, the complexity of sea life, not to mention how damaging it can be to trap sea life in unfamiliar and hostile surroundings. There may also be value in dispensing with a couple of misconceptions.

Myth #1: There are two market segments (premium and bulk) that are easily delineated and the premium market is dramatically smaller than the bulk market.
Reality: There’s a very long continuum that encompasses all market segments, with raw bulk free MT at one end and $25,000 tag line translations of 3 words at the other.

It’s far more accurate to characterize the continuum in terms of gradual and consistent gradations of shade rather than in terms of clear differentiating boundary lines. The “premium vs. bulk” dichotomy is a form of shorthand only. That also applies to price and quality, since the correlation between the two is not always linear. The premium sector includes commercial segments that are fiercely guarded and (often) shrouded in secrecy to prevent additional competition. Many of these are boutique translator-owned companies that deliberately fly under the radar of “research” companies like Nonsense Advisory (itself shamelessly in bed with the large companies it purports to “cover,” and stubbornly resistant to acknowledging its own 50-kilometer-wide blind spots) to avoid alerting other companies to their profitable businesses. There is an astonishing amount of money in these premium sectors. Pure translation alone in the high-end expert pharmaceutical, medical device and IP litigation as well as the premium legal, financial and marketing sectors across all languages and in all countries dwarfs the entire global IT localization industry by about two to three orders of magnitude. There are some years where one single IP pharmaceutical litigation case in Japanese-English alone will run into the $10 - $20 million range – about 10 times the “savings” that TAUS preaches are available to localization companies and their end clients that embrace their “translation as a utility” model in localization. That’s one single translation project in one single language pair. And the net profit margins are considerably higher.

Myth 2: Price is the key differentiator between the premium and bulk market.
Reality: While it’s true that the premium market tends to operate at higher prices, the market really operates on a completely different value proposition than does the bulk market. That proposition is that the cost of failure is dramatically higher than the cost of performance.
So in the premium market, the cost of translation errors – liability, regulatory failure, loss of life, damaging publicity or significant loss of prestige – far outweighs the cost of “getting it right.” Paying whatever cost premium for translation that is necessary to PREVENT the cost of failure is viewed as a wise investment.

In the bulk market, those two are reversed. The cost of failure is low, so there is no corresponding push to invest in getting it right. This can be tested by comparison to the dynamics of other industries, too. The cost of failure for a Walmart product is very low – the consumer almost expects the damn thing to break. It’s the same with cheap online localization and “just good enough to understand it” bulk translation. But a fractured fuel pump on a Boeing aircraft in flight has an enormous cost of failure, so several layers of review, ongoing maintenance and testing as well as regulatory enforcement are built around it in an effort to ensure that does not happen, a process which drives up fuel pump manufacturing costs dramatically.  When the failure of an IPO or the collapse of a deal due to a translation-related regulatory failure or when nuclear weapons are improperly dismantled or lost to unknown people – yeah, that’s a very, very high cost of failure. Wallets open up to pay a premium for translation in these cases. Of course, translators who want to play in this market must be Boeing quality, though, not Walmart. (If any serious person considers this view “elitist,” I will contemplate the validity of that charge when that person agrees to fly on Walmart-manufactured jet aircraft that fly without regulatory approval or oversight.) :)

Myth 3: The largest translation company in the world is Lionbridge, crowned once again by Nonsense Advisory.
Reality: It isn’t. It may be the largest localization company that openly shares public financial data in an easy-to-read format and hence is trivially “researched,” but it omits huge operations that just don’t advertise their existence in quite the same way. For example, there are Global Linguist Solutions and L-3 Inc. just in the US alone. Never heard of either, right? GLS won the original US Army contract to support Iraq ops worth about $4.64 billion over five years after L-3 had the original one pre-Iraq. Perhaps more to the point in terms of current size, the U.S. Army recently awarded a huge US Army contract referred to as DLITE valued at $9.7 billion to 5 companies including those two. Those are JUST the U.S. Army contracts. The open, unclassified ones. This does not include all the other U.S. federal open spending on language services for all the other agencies that these same companies along with DynCorp and McNeil and Booz Allen and a dozen others that have never been to an ATA or any other translation conference compete for and win. It also omits all U.S. classified and confidential contracts. It omits all other governments’ outsourced classified and unclassified language spending. It’s like omitting the Indian Ocean and half the Pacific from your "research."

It’s a vast, complex, cloudy and immensely varied translation sea out there.

I know that those who have dealings with the US government around translation technology at least have an inkling that this is true. It is sort of like the discussions on the Deep Web which contains much of the highest value information available in the world that is not indexed or accessible by the search engines that we all use. This is the part that is private, gated and contains the really important high value content that can only be seen by people who are properly authenticated and authorized. I can’t say for certain that the proportions in the graphic below are true for the translation market but based on what I directly know about the data volumes processed in the clandestine communities it certainly would not be impossible.

Anyway I thought this subject was interesting and worth more exposure. Also, it was easy to do as Kevin Hendzel wrote the bulk of this post.   

P.S.  I thought it was worth adding this post-script here since Luigi Muzii has also made extended comments on his blog on this subject and so I add his Twitter comment to the main body of this post.

From @ilbarbaro
My comments to @kvashee latest debated post can be found in goo.gl/Zpu6Mh, goo.gl/OfZRsB and goo.gl/bN1HJB

Highlights from ELIA Munich ND - Translation Pricing & PEMT Process Management

My view of a conference is usually determined by the quality of the sessions related to MT and translation automation, or sometimes other sessions  that may trigger new thoughts on innovation and business process evolution. The ELIA conferences I have attended, stand out for me because I think they have better content in general than most, and one actually learns new things. To me it is clear that business translation is evolving beyond a focus on software and documentation localization (“the SDL mindset”) and I look for content that recognizes and addresses these emerging issues and market imperatives.

One of the most interesting sessions and perhaps the only one by a translation buyer was entitled “How Cloud TMSs are Changing the Relationship Between a Translation Buyer and LSPs” by Elina Lagoudaki of Turner Broadcasting. She described how cloud-based technology is used to manage a growing stream of digital media localization projects. Turner is a good example of a translation customer who has many small jobs (micro translation), often involving social media content and usually also closely  linked to dynamic web content that needs to go out in 15 languages. Elina presented her very organized and structured process to identify, administer and supervise translation projects and also provide final quality feedback to translators on an ongoing basis. Some things that she pointed out about her process included:

• A preference for a SaaS or Cloud-based TMS solution (WordBee in her case) over inflexible, costly, arcane and management-heavy onsite solutions
• The need for a management dashboard that allowed high level and job-specific status monitoring
• A translation management environment that allows and facilitates collaboration between translators
• A translation management environment that allows and facilitates online review and content sign-off
• A translation management environment that allows and facilitates ongoing feedback to translators
• A translation management environment that allows and facilitates that enabled terminology and TM collection and centralization
• A translation management environment that allows and facilitates that facilitates vendor comparison and selection

For those who still have doubts about how much sense cloud-based solutions make for many customers, Elina presented a very clear and articulate view on how her cloud solution was superior (to archaic client-server solutions) not just in terms of cost, but also in terms of scalability and ease and speed of customization, for her unique requirements and needs. 

Some of the things that stood out from her presentation in my mind included:

• Half the translation work was done by agencies and half directly with individual freelance translators sourced via ProZ  - and it was interesting that she used the phrase “trusted translators” to describe how a subset of the freelancers had risen to this status because they had tuned in to the writing style of the company, were reliable, and thus favored on an ongoing basis.
• Elina also showed a slide (shown below) that showed the large variance in rates for the same language pair. This variance will of course raise questions in a buyer’s mind about whether there is a trade-off in quality or reliability of some kind, or is it just what they think the market will bear. This slide shows why the buyer should be wary and do the due diligence to understand what trade-offs they make if any,  with higher and lower prices.  LSPs should also take great care to properly understand their costs, define prices and link them to well defined quality/service deliverables, as collaboration tools like WordBee will make these comparisons easier and easier to do.

The most striking point she made to my mind was when she showed a slide of how different LSPs responded to an RFQ request where every agency was given the exact same job specification and was also promised that they would get 20+ more projects of this kind over the coming year. The translation task involved translation of 3 flash banners, which means there was very little text (5 –10 words at most per string) to translate but the translated text had to be placed in a Flash banner. So we are talking about maybe 30 words to be translated into 14 languages and delivered in a multimedia format. It is kind of shocking that she received quotes that ranged from $310 to $10,430 for the exact same job description. The actual price quotes she received are listed below in the slide she showed to show the wide variance in price quotes.

This points to several problems in the translation industry that range from completely random pricing practices, lack of understanding of multimedia content and translation tasks, price gouging, business model mismatches to sheer unprofessional behavior. She characterized this as “a wild west approach” in the market where anything goes. There were clearly some in the room that were upset at being exposed and I heard that some complained that too much information was shared. I think we will increasingly see more work involving multimedia content, coming in steady dribbles but critical to building trust and credibility with a customer. It turned out that the agency with the lowest quote also had a track record of success and reliability  with Turner, and thus probably understood multimedia issues much better, and so did not impose huge price penalties for  simply putting text into Flash. The companies with the highest price quotes clearly did not understand the complexity of the job or perhaps simply lacked scruples.

This is related to some extent, to an interesting UnSession discussion that I also attended where a group of LSPs (plus Elina and me), discussed how one could respond to a potential customer who said that they already had a translation agency they were working with. Much of the discussion focused on identifying “problems” with the current  vendor and thus displacing them, and to my mind only one of the LSPs had a compelling differentiation story. The session made three things clear to me:

1. It is very easy to displace an LSP that is previously engaged with a customer if you can identify problems the customer is having with their current vendor.
2. That quality and “service” are repeatedly used as differentiators but nobody can define either, in a way that is understandable or clear to a buyer.
3. That very few LSPs understand the business of the customer and thus have great difficulty building trust.

The best strategy that I heard in the UnSession, was from Alinea, an LSP who had a clear domain expertise & focus and who ONLY focused on building customer relationships in that domain, with a long tenured in-house team that were expert in the subject domain and thus could add overall business value in the translation process. I would bet that that particular agency is very hard to displace, and can charge premium prices, and are often viewed as real trusted extensions of their customer's organization. 

Building trust is a critical foundation for long-term success in a service business, and this requires that there is real transparency, clear communication and a collaborative and cooperative business approach.

  Post-Editing from the LSP Perspective highlighted many issues around the LSP experience of PEMT in the market today. The session had a strong focus on the management of the PEMT process which included things like managing quality and cost/price expectations with the customer, selection and training of post-editors, and ensuring source material quality is good, as this is an area that LSPs understand and action here can have a large impact on MT quality. Some highlights from the presentation:

• Post-editors need to have a positive attitude (to MT), be flexible and  be “system-oriented” to provide constructive feedback,
• The technical issues that the session focused on included capitalization, punctuation and there was much talk about the issues in handling tagging with MT which is as messy with MT as it is with humans,
• Several examples of MT output with various error types were shown so that others could understand the nature of the problems and the challenges,
• The problematic issue of proper compensation was discussed and most felt it was easier to properly determine this after the project is done, though Edit Distance, BLEU, Average Words/Hour and other effort measurement approaches were also discussed. It is interesting to me that my own blog on this issue written in March last year is the most popular post on my blog even today. I find that using “trusted translators” to establish a priori rates, are a very reasonable and fair way to establish fair compensation rates. However, this does require some skill with proper sampling technique. For some very specific guidelines on how this could be done a priori check out this article from Asia Online.
• A survey of ELIA members suggested that the average post-editor throughput was 5,189 words per day and that the range seen was from 1,500 to 10,000 words/day per post-editor. 
• The presenters felt that there was an urgent need for a good PEMT tool that facilitates error detection and error correction, since it was felt that MT had very different error patterns than TM typically does.
• The presenters also felt that dealing with low quality MT output was worse than TM 0% matches and should perhaps be penalized and charged at much higher rates, since the translator had to spend more time making this determination. Asia Online provides a solution for this problem by providing segment level confidence indicators and thus low quality segments could be pre-identified and processed differently to minimize the bad segment detection effort.

What was missing from this discussion was a focus on the HUGE impact that the MT system being used has on the post-editing experience.  While I admit that all the suggestions and findings presented at the session would be useful for almost any PEMT exercise, some MT systems are more adjustable and configurable and thus ensure a better and more productive PEMT experience.  I know that within the Asia Online experience, MT systems go through several rounds of tests on small representative data subsets AND corrective actions BEFORE being put into production. During this MT system refinement process, high frequency problematic error patterns are identified and addressed to both minimize post-editor frustration and maximize throughput and productivity. This molding of the MT system can only be done with some very selective MT systems  but I think this is a critical step if you wish to avoid tedious, repetitive errors like many shown in the sample slides and maximize your ROI. The slide to the side shows how a typical Asia Online system evolves and shows which error types are the easiest to correct. In general spelling, punctuation, capitalization and basic terminology errors are the easiest to permanently correct and the grammar and syntax errors are the hardest to completely fix.

I found the session by Diego and Guillem Vidal – NOVA  interesting, as here we have an LSP who has reached a level of competence with MT (with an expert partner) and are seeing that they can provide better productivity, better terminology control, faster turnaround and lower error rates even with medical domain content. Their actual experience resulted in a 6X increase in MT word volume over two years to 10 million MT processed words in 2012. It is refreshing to see this type of competence when we still see examples of half-truths based on very shallow assessments being presented as conclusive fact in articles published in Multiingual.

I also had a fireside chat session with Renato where we discussed industry trends and much of the material we covered is summarized in a previous post where we talked about how volume is growing, continuing flows of micro translation tasks are increasing and how MT and automation are gaining by the day. One point we disagreed on was about the impact of  “new” translation focused ventures like Smartling,  Cloudwords and Lingotek. I feel these initiaves are all very interesting and relatively innovative, and make the whole translation services purchase and management  process much easier and simpler. Renato felt that while they had succeeded in raising money and had a “technology story”, they had yet to prove that they could provide the same level of “service”. Given that nobody can really define “service” with anything resembling clarity, I think it is quite possible that some these new ventures could displace some LSPs (Multi-Language Vendors – MLV) and become the new aggregators of translation purchasing activity  because they do the following things well:

1. Simplify the translation purchasing process (without the slow and laborious and often customer hostile TEP mindset where the customer is always wrong),
2. Eliminate the need for buyers and agencies and translators to keep multiple suites of incompatible translation CAT tools on hand, by simply ingesting translation related content into their technology infrastructure straight from the content creation systems (CMS) and return the translated content straight back to the customer CMS via straightforward web-based interfaces,
3. Handle small projects as well as large projects with equal ease and efficiency,
4. Provide collaboration focused software infrastructure for translators, buyers and project managers in the cloud, so real work related conversations can happen without hundreds of emails with receipt notifications being used,
5. Enable translators to spend most of their time focusing on linguistic work rather than dealing with file format conversions, tag management and data transformations before they actually get to the translate step,
6. Easily handle multimedia, video and mobile content which will continue to grow in importance,
7. Greater facility to handle, and mix and match different customer content types to different production methodologies which include TEP, customized MT, crowdsourcing and productive and efficient PEMT.

Elina’s slide on what she would like to see in the future are clear indications of what lies beyond the SDL (software and documentation localization) world for a modern buyer: more competence with multimedia, new business models for microtranslation,  more innovation from tools vendors and better standards (so that data can flow more quickly and easily in and out of translation processes).

We live in a world where faster and cheaper production at “reasonable quality” is beginning to be linked to business survival. Companies that don’t get it done in time or don’t get enough done in time lose market share.   As the volume of smaller (SME) companies going global increases, they will likely find these new portals much easier to work with, rather than have to go to the arcane and archaic client-server software world of SDL et al.  Innovation matters more and more and while I cannot say with any real assurance that these new portals are THE winners of the future, I would bet on them over those with the SDL mindset. Innovation usually means making it simpler and more efficient. You can see this lack of enthusiasm from investors reflected  in the stock market performance of  both LIOX and SDL as they trade at market capitalizations way below their annual sales. Even Google is working as an aggregator for video subtitling projects in addition to their widely used MT which I assure you many translators use on a regular basis. In Stefan’s session it was mentioned that the greatest trigger for organizational change is reaction to competitive action, but in this industry it seems that change is sneaking up in a way that many don’t even realize it is happening.

I learnt that the people of München like their beers and potato balls large, in fact very large, as you can see from this photo of Irina Voronova’s hand versus the potato ball which was about the size of an American softball.

I asked Stefan Gentz, who is second to none in terms of conference attendance what he thought the best conferences were from all those that he had attended , and his almost instant reply was that GALA Miami was the best in terms of balancing both quality content with great networking opportunities.

I also got to walk around a bit and wandered into the English Gardens which inspired Hans Cousto to develop his theories about The Cosmic Octave which later led to the creation of the planetary series of gongs made by Paiste. For those in the know, Germany is a leader in the science of sound healing, something which is gaining acceptance as a way to deal with a variety of illnesses.  As somebody who is interested in really good sound and an amateur musician who plays the sitar I find this quite interesting, and even fascinating, and I really appreciate a culture and the people who would place a well tuned piano in a public park so that anybody could walk up and play it. In the brief time I was there I saw some very accomplished pianists walk up and play Bach and Beethoven, and also some who played that old favorite “ Chopsticks”.

For a completely different view of the conference from the lovely ladies of WTH (who some lucky attendees got to meet au naturel in the sauna) check out their blog post on the event.

Emerging Language Industry & Language Technology Trends

As the year comes to a close, it is sometimes useful to review and look ahead on where things may be going, and even though many of these type of ruminations can be self-indulgent and self-serving, I have decided to throw in my two cents anyway. These are personal opinions on other opinions, and like much of what I do in this blog, this is also a collection of information that I consider most worthwhile to share on this subject of trends.

The translation industry remains a highly fragmented industry with relatively inefficient production and business models. In 2012 we still have over 25,000 language service providers (agencies) of varying quality and professionalism doing the work of business translation across the globe. Efforts to define the final product or service produced by these firms are unsuccessful despite valiant efforts from industry associations.   However, many have been talking about change and disintermediation and many of us are aware that something is afoot. My intent here is to collect and organize different opinions rather than only promote my own and hopefully I succeed in creating a broader clarity on these emerging trends and possibly starting some discussion on this.

A trigger for this post was a conversation with Bob Donaldson who presented on this theme at Translation Forum Russia. I have also added some material gathered at other conferences I attended this year that extends these initial opinions. Bob has simplified my task by gathering and sharing the opinions on key trends of several different viewpoints as summarized below. (I have kept the text exactly as presented in his slides at TFR but you could get clarifications and detail beyond this slide verbiage by directly contacting him). 

Multi-Language LSP Vendor (MLV) Perspective by Renato Beninatto

•  
• Rise of Micro translations (interesting response to this point by Luigi Muzii)
• Outsourcing to translator teams
• Demand for “long tail” languages

CAT Tools Training Perspective by Angelika Zerfass

•  
• Terminology Management gaining traction (finally)
• New content types (twitter) don’t fit old processes
• File management becoming more complex

Translator Perspective by Jost Zetsche

• Deep integration of MT into translation workflows
• Limited lifespan of LSP as (mere) middleman

End Buyer Perspective by Anonymous

• Demand for continuous translation with very little context (Micro translation)
• Declining Quality Expectations
• MT will fill the gaps created by the first two at an ever-increasing price

Single & Regional Language Vendor (SLV/RLV) Perspectives in aggregate

• Greater usage of MT
• Multi-faceted approach to quality
• “Price compression” will drive small/inefficient players out of business
• “Disintermediation” will show up in various forms
• Greater demand for self-service portals

Bob Donaldson Top 4 Trends Summary

•  
• Transition from “Project Orientation” to “Content Stream” orientation
• Increasing integration of MT at all levels
• Increasing emphasis on velocity rather than price or quality
• Increasing reliance on global SLV partners rather than freelancers

All resulting in changing and needed innovation in business models



As I consider all these views, my own sense is that the following trends are increasingly understood to be clear and continue to gain momentum:

• Business translation is shifting focus from intermittent project work of relatively static content to continuously flowing streams of information that might enhance international business. The old “software and documentation localization” (SDL?) view of the world is becoming a smaller part of the core translation challenges that global enterprises face to be successful in international markets.  There is also a growing awareness that translation should be able to flow from document/video to PC/web to mobile/tablet easily, quickly and efficiently.
• An expanded view of critical and translation-worthy content that includes more informal corporate content as well as customer generated content and social media conversations about products. Social media has dramatically changed the traditional top-down views of marketing, and this impacts the decisions on what is important to translate as enterprises realize that purchase decisions are being made in social online conversations and information sharing.
• The importance of automation and collaboration increases. This is more than just MT, it includes greater integration of content flows from the information creation process all the way to information consumption. Successful use of comprehensive automation and collaborative processes will help create meaningful differentiation and competitive advantage amongst LSPs and help identify superior players.
• The increasing importance of cloud based services and infrastructure to facilitate collaboration and standardization of translation-related informational flows. This will also mean that desktop tools (TM, MT) will become less important over time as usage shifts to the cloud.

On the MT front I expect the following trends, much of this is already in place and also gaining momentum:

• Increasing awareness amongst translation professionals that domain focused MT produces the best results in terms of production efficiency and productivity gains. We will hear of many more successes of these kinds of focused systems.
• Increasing understanding of post-editing based translation production and processes. While there will be some or many “premium” translators who refuse to work on PEMT projects, more and more translators and LSPs will learn to work effectively with MT.
• Continued momentum in the understanding of MT system quality which will result in better PEMT experiences and trusted, fair and equitable compensation practices. This is essential for broader long-term adoption.
• A shift away from free and instant MT solutions to expert collaboration and expert-built MT systems.  (Some will say this is self serving and to some extent it is.) It has become increasingly easy to get some sort of MT system into place by throwing some data into a hopper, but very few of these systems provide long-term productivity gains and strategic advantage out-of-the-box. MT in 2012 is still very complex and getting some kind of basic system together quickly should not be equated to building long-term production efficiency. Experience and knowledge about MT system development matter, and the best, i.e. the highest productivity and best overall ROI systems will still come from experts. As Malcolm Gladwell says, “Practice isn't the thing you do once you're good. It's the thing you do that makes you good.” Experts are people who have built hundreds or thousands of MT systems. Many who experiment with Moses and other instant MT solutions will learn that deep expertise is required to move the system quality beyond the initial engine capabilities and that long-term business advantage only come from continuously improving MT systems. In 2013 MT system development is still an evolutionary process and a skill based technology, not the instant iPhone-like gadget that some want it to be. There is a difference between using MT well and just blindly using MT because it  is in vogue. If you don’t know what you are doing and what you will do after your initial system is in place, being able to do it quickly initially is not going to really add much to your business leverage. 
• Better understanding of what MT can and cannot do, and more pro-active use of MT to build long-term competitive advantage rather than just be a means to react to cost pressure or client demands. This means that some LSPs will build MT systems BEFORE they actually have a customer to ensure that they have an advantage in particular domains that they feel have strategic promise and potential.

I have discussed the importance of automation (process integration which includes MT and much more than traditional project management) and collaboration (which also means that you respect your workers and customers) as important elements of new business models that can effectively respond to and take advantage of these trends. I would like to add agility as a critical third element. What is agility or agile? I think this is increasingly becoming more important as a critical element for success in the future. 

: Characterized by quickness, lightness, and ease of movement; nimble.
: Mentally quick or alert
: marked by ready ability to move with quick easy grace
: having a quick resourceful and adaptable character

So are there any examples of where all these elements come together? Not really, and definitely not at large LSPs like Lionbridge, SDL et al.  Largeness (over $50M for the translation industry) generally tends to undermine agility and often collaboration (in the sense I use the word) too. I think there are smaller companies where all these elements are more visible and look like they have the potential and promise to bloom. A nice and succinct description of “agile” is presented by Jack Welde, CEO Smartling in the first 8 minutes of this video.

An 8 minute overview of Agile Business Translation

I suspect that many new business translation customers will opt for this type of lean, quick and more cost-effective approach over the traditional LSP sales and TEP process hype, where the customer is often treated like an idiot that needs to be slapped into shape. Lingotek is another company with an approach that has many key elements in place and I think is well positioned to challenge the old model. In both cases outsiders are creating tools to change a cumbersome old business model and facilitate rapid collaborative production. DotSUB and Amara are two that are focusing on facilitating translation of the huge volumes of video content that are increasingly useful to help sell products and services, and are increasingly recognized as more important than a lot of traditional localization content. In all cases these new approaches can steer easily to professional, MT or community based production or any combination of the above at significantly lower prices with “quality” intact. Try and have this discussion about flexibility, speed and various production modes with a large traditional LSP and you will likely see that it may be possible at a significantly higher price, and I suspect the conversation will also be labored and difficult.


All of this for points to examination of changing business models and innovation and the most interesting discussions I have seen on this subject for this industry are at the The Big Wave. Listing all these trends has some value but it is useful also to understand how all these trends mix together and what implications it might have. I can’t say I have the answers but I think these are good things to ponder.  I have seen several interesting posts about this at the Big Wave site. For example here are some selections from this post:

TEP is the unique answer of most translation vendors, an old-fashioned and somewhat obsolete answer too, but it is the only model they know.
At a closer look, none of the big players in the industry, however, has produced substantial product, technological, or process innovations.
When customers don’t get any new value from traditional vendors to meet new or implicit needs, they abandon these vendors and do something different themselves to better accomplish their goals.
This is why innovation in translation has always come from outsiders.

There are also interesting posts On Information Asymmetry and The Disintermediation Myth: Bogy or Opportunity? which are provocative and worth reading even though some might feel they are slightly opaque.

The future is likely to see multiple production models co-exist e.g. TEP, PEMT, Customized MT, Free MT, Crowdsourcing or Community Collaboration as well increasing examples of social translation, as these will all be necessary to solve different types of translation challenges we face. I have often thought that it is too complicated to buy translation from traditional LSPs and I hope that we as an industry make this much more clear and simple for the customer who has never heard the word localization used in relation to translation. There are a lot more of these customers out there than there are customers in localization departments. People usually find it easier to buy, when they know exactly what they will get for a given price. People like predictable outcomes (really hard to do with MT) and would like to be able to easily compare alternatives.

I would welcome any readers who would be interested to share their own perceptions and views on these trends as a guest post. I assure you that I will print it without modification (hopefully no personal attacks or rants).

Translation Crowdsourcing

The phenomena of a crowd (or community) stepping forward and doing real translation work, often for no direct financial compensation is something that troubles many in the professional translation world. Mostly because they see this activity, as work being taken away from legitimate professionals or they see it as a ploy to reduce prices. While in some cases their fears may actually be justified, in the most successful uses of this approach I think it is clear that this is not true.

As I have said before, the growing momentum in the volume of content demands new production models. This momentum which exists both in the corporate world and also the general world out there simply cannot be addressed by ONLY using traditional professional translation production models. New needs require new approaches. For those who insist that the data deluge is a fiction, the rest of this post is probably irrelevant.

Another key driving force behind new crowdsourcing initiatives is the need to engage and interact with users in new markets. In new markets having active conversations with locals is key to building brand awareness and really learning about local needs and behavior. This is frequently a more important driver than cost containment as many in the industry think.

In some cases were it not for crowd-based localization efforts, it would simply not be done as it is not economically feasible to undertake the same efforts (and expenses) as are made for FIGS/CJK for “lesser” languages. Thus crowdsourcing is emerging sometimes as a means to get “lesser” languages done.

If we look at some of the most successful examples of crowd-sourced translation in practice, we can see that they have many if not all of the following elements in common.

A Crowd/Community That Is Invested:

·         TED Open Translation Project – Volunteer translators are often inspired by the content and wish to share it with their friends and countrymen. June Cohen has said that the volunteer translators in general do better quality work than the many of the paid professionals, who initially did a few translations to seed the project because of their passion for the subject and often their subject matter expertise. This effort has now enabled over 20,000 translations into 80+ languages of really challenging material. Many professionals also volunteer because they believe in the high general value of the content.

·         Facebook – Users who wish to build and expand the friend community in their particular language group. This effort has enabled Facebook to grow rapidly in international markets and accomplish very rapid coverage across 60+ languages. Had they used traditional means to do this it may have taken them years to get to the same point. Critics also often miss the point that engaging real users in the translation task also encourages rapid growth of the user base as “user translators” engage friends into their network.

·         Microsoft - MVPs (top accredited reseller partners) who wish to make technical support knowledge about Microsoft products more easily and widely available in their markets. Their efforts are rewarded by lower support costs and also an increase in product sales as more and more users look for self-service knowledge base information. Microsoft has been a trail blazer in making large amounts of knowledge base content available via MT, they are now adding crowd based editing to raise the quality of the translated information. Thus the most used and vital information tends to get the most attention and benefits all users.

·         Asia Online – Student users provide corrective feedback to continue to improve the translation quality of the Wikipedia and other knowledge content that is initially done by highly customized MT engines and paid translators. The students themselves will be the primary beneficiaries of this content, and their efforts will enable them to access high quality educational information. The volume of this information will likely increase a thousand fold.

 ·         Yeeyan:  has 150,000 registered users, who collectively translate 50 to 100 news articles every day from English to Chinese. Since its inception in 2006, the site has grown into a key gateway for Chinese speakers who want to follow international news. It has been so successful that it has attracted the attention of major news sources like The Guardian and ReadWriteWeb. Yeeyan is focused on addressing the problem of ghettoization of information by language through a community collaboration, where members both identify interesting content and also help to translate this content.

·          Adobe: This is a much more carefully managed effort designed to engage influential users, partners and customers to help provide relevant information for the broad Adobe User community in China.

T   Twitter: The translation center asks Twitter users -- all volunteers -- to help translate Twitter's interface into various languages. Once the basic support pages are translated, a select group of the "most active" translators are invited to work with Twitter to "maintain localized versions of the service." Twitter boasts that its translation center has 200,000 translators, and that the localization process for Dutch and Indonesian took just one month from the first call for involvement to its announcement. The availability of its interface in multiple foreign languages is bound to increase its popularity and effectiveness not only for online marketing but also for social and political activism.

Software Infrastructure That Facilitates Contribution & Participation:

In all of the cases above the companies involved crowdsourced translation initiatives need to invest in software that enables tasks to be parceled out, evolve as tasks change, enable efficient administration, maintain quality, gather feedback, and build self-sustaining eco-systems. The tools developed by dotSUB, Lingotek, Yeeyan and Asia Online are all unique collaboration and translation workflow management tools that enable these kinds of initiatives, They make little or no use of industry standard tools like Trados and TMS because of the highly proprietary, rigidity and archaic nature of these tools. These new-generation tools are much more open and are designed to evolve with technical and process advances on the internet today. It is quite possible that these community efforts could produce tools that supercede many of the tools in use today as these new tools focus on collaboration and sharing assets to enhance the efficiency of the collaborative translation process.

The Importance of Engagement and Higher Purpose:

It is interesting to note that translation is not the primary business of any of the companies listed in the examples above. In every case the goal and intent is to make more information available faster. Even for many of the corporations that are exploring crowdsourcing, the rationale is often more about customer engagement than cost savings. It is also important to note that none of these initiatives could even be attempted without the use of automation and large-scale community support and they are enabling initiatives that would not be possible otherwise. This is also true for Facebook who still had to use professionals to translate legalese that their community was not interested in translating.  The role of communities is likely to increase in future as more of the world comes online.

As we move forward we will see much more video and other rich content come online and already it is clear that the old approaches will not enable us to make this new content multilingual in an effective time frame. Crowdsourcing and automated translation will be necessary tools for an organization that seeks to communicate across the globe. As Clay Shirky has pointed out, the ‘cognitive surplus’ of the online population is a force that can be harnessed under the right circumstances and for the right purposes. It is likely that the professional translation world is going to see significant disruption in the coming years, as innovators figure out how to build sustainable models around community engagement, technology and organizational mission. However, as we have already seen, there is much that the crowd has no interest in doing and we should expect that this is not likely to change.

Crowdsourcing is here to stay and is a new mode of production that enables high–volume projects to be undertaken, engage with users and partners more deeply and participate in multilingual social networks where so many branding impressions are being formed. Managing crowdsourcing is also a major opportunity for savvy LSPs who have processes in place to recruit and manage the collaboration of dispersed volunteers and contributors.