Watch videos with subtitles in your language, upload your videos, create your own subtitles! Click here to learn more on "how to Dotsub"

From World-Weary to World-Ready: Meeting Today’s Content Globalization Challenges

0 (0 Likes / 0 Dislikes)
Allereerst wil ik een applaus voor onze MC die vandaag de dag allerlei moeilijke namen heeft moeten uitspreken en ik denk dat hij een bewonderswaardig werk vervuld heeft in zijn rol. Ik zou ook graag de organisatie bedanken die mij de mogelijkheid gegeven heeft om vandaag te mogen spreken We gaan overschakelen naar een versnelling hoger op - een ogenblik geduld a.u.b - laat me mijn dia's even nemen and up on the screen and hopefully technology will be with me today I am not using the traditional PowerPoint tool so hopefully I will be lucky enough for the technology to work with me. I would like to shift gears a little and share some thoughts, with you about, what it is that we actually do. We heard today discussion about some of the technology around localization. some of the business aspects and when I say shift gears I would like to talk more about the actual content that we are in the business of localizing and producing I would also like you to help me think about the production of the content not just the translation and localization but also how the content is produced and some of the challenges that exist around producing and then globalizing content and try to structure this in the form of three main challenges that we face in doing that On e of the things that I think is important for us to remember is that those who create content and those who localize content are really in the same business We are all in the business of getting that content out to the end customers So as we discuss these challenges today We'll be thinking about how it is that these challenges affect our ability to do, in fact, to accomplish that mission and get that content to the end users Now before we get too far, lets talk for a second about what content is If you think about in a basic form content really is getting something from one head out across a medium into a bunch of other heads so that they will perform some action That's really the root definition of content whether its a marketing statement or an instruction for use or for that matter music or art The idea is to get something out of my head and get it into yours and there is some medium that we use to do that What happens I think, in our business is that we tend to focus, a really lot on the medium that we use to transmit this content so we get lost very often in the pipes the plumbing in the business of getting content to other heads We talk about content management systems, we talk about workflow TM, MT, XML and all kinds of things like that Very important things, don't get me wrong, but, we end up getting really focused on this technique, this system of moving the content from one head to the other. And so that really presents us with our first challenge. And that is efficiency, and we have heard here today and we have heard it in many ways from the industry for years and years. Really it's about how can we get the content produced and localized, faster and cheaper. I am on the localization service provider side of the business and this is a question that I get asked, every hour of every day by every customer. How can we do it faster and how can we do it cheaper? And those of you who are paying attention here will notice that there is something missing in that request. What's missing? What's missing is "better". No one comes to me every hour of every day and asks: Can you do it better? They only want to know can you do faster and can you do it cheaper. We'll get back to better in a moment. So are focus now becomes, how do we efficiently take this process and move it ahead? I'll share a couple of thoughts with you. I am not here to discuss this in any sort of thorough way but just a couple of ideas for us all to think about as we go about answering some of these questions. The first thing that we often reach for in doing things more efficiently is Reuse. How can we recycle content? How can we get it so that we can repeatedly use it? So there are really three main aspects to that. The first thing is that we have to centralize all the content - so how can we store it all in one place? If you are going to reuse something you are going to have to be able to find it. So we come up with technologies and systems to centralize the content. The next thing that we look to do is to find ways of matching content. So this means, How do I find the same thing again? How do I find something that is similar? How can I create all kinds of systems for reusing that content? And in a way that will make me more efficient. And finally, this is a really important point that we tend not to talk about a lot in our industry. How can I figure out how to ignore this reused content? This goes to very fundamental questions like 100% matches, do I review them or not? In the localization business the dream is for 100% matches completely cost free. And for many service providers that means you don't touch them and you ignore them completely. So this is kind of the philosophy that is around the idea of efficiency. If I can ignore those portions of content and reuse them if I can find and match then I can be more efficient. Our colleague from HP mentioned that if I have 85% reuse then I only have to pay for 15% of the content the next time around. That means I have effectively found a way to ignore 85% of that content. The next thing, and this is mixing a little bit the worlds of those who produce content and those who localize content. The next thing to do is to figure out how to grind up the content. We break it down, break it up into smaller pieces. Traditionally, and when I say traditionally I mean 10 years ago. Content was in large pieces, in linear form, in the form of books or documents or these kinds of things Now we look at content in sort of micro granules. And the goal here to be efficient with content production and localization, is to accomplish a few things by doing that. Separate first of all the text from the layout This way we can immediately separated these two problems and the text can now be dealt with separately from the presentation that it is in. So we talk about XML, we talk about all sorts of markup languages, we talk about meta content and so forth. The idea there, once again is that I can be much more efficient, if I have small pieces. And breaking up the content into chunks, that I can then reuse. This is where we talk about segments and we talk about sub-segments even and all kinds of things like this. So the idea here once again is to break down the content into more reusable manageable pieces. And then to churn the content through the system quickly and efficiently as possible, and in an automated way is preferable. So you will notice that I chose for the graphic under grind content, a pile of coffee beans. Because frankly, this kind of conversation reminds me an awful lot of coffee. The high churn, micro dropped kind of environment that we are in, starts to feel an awful lot like a commodity of coffee beans. I suppose some would even argue that it does not amount to a hill of beans if that metaphor makes any sense, to those of you who are not native speakers of English The idea here really is that we are finding ways to break content down so that we can manipulate it much more effectively and much more quickly. The final way to increase efficiency is to put technology in control of the content. So instead of previously the paradigm was having humans controlling content we can now actually use technology to actually run the content. This is where TM comes in, which is a maximum reuse tool that has been implemented across the board in our industry to really find ways to make efficient reuse possible. With MT coming to the fore, again now we have even more power now at our disposal so we can really move this content efficiently through the system. And some would even argue that as we move into the crowdsourcing world and create tools for that, we can find an even more efficient way of doing things because it requires, some would argue, no effort. The crowd can do all the work. So we are letting technology drive our effort into meeting this challenge of improved efficiency. Now what the challenge of efficient raises though is a deeper, underlying fundamental challenge. And that is, we have many more sources of content now. One of the things that we don't talk about much is where does all this content come from? And I think that if we go back to that model that we discussed earlier on, of getting content from one head into other heads, we are now seeing a world where lots of heads want to get information into all the other heads. So we are now talking about content coming and going from all kinds of sources and audiences and these concepts are getting to be quite mixed. So we are having to deal with a lot of new sources of the content coming into our system, in terms of making it more efficient. There are new types of content now, it used to be that companies pushed content out in terms of going with their products or to their customers. But we now have content that is going the other way Content that is being exchanged, content that is being manipulated and lots of types of content that are no longer the static kinds of public h??? that we grew up with in the last 10 years And we now have new places that this content is going to places where it is shown , where it is coming to be used, on the social media networks. All of those kinds of ways that the content is moving, around the world have become quite a big challenge. So fundamentally what we are really facing now is the depper challenge of more and more stuff. That has to be moved around, more and more stuff that has to be created and is being created and has to be localized. It is interesting today, the first two speakers mentioned as their first challenge for their take on the industry as volume. Volume being the first challenge. That an absolutely fundamental issue here and that is the volume question is definitely becoming a deeper challenge for us. This explosion of content that we have to deal with. It is critical that we address this. Not just in terms of the symptomatic challenge above that which is to be more efficient but to really start to look at the question of teh volume itself and the kinds of things it creates, the kinds of choices it makes us make. I think it is probably true and everyone in this room hopefully will agree that volume drives almost everything, we do in terms of content. It is the fundamental driver of so many aspects of our business. Time is directly influenced by how much volume there is, how much content there is, and how much time it takes to process the content and translate and localize and globalize it. All of that is directly related to the volume. The cost, those of you in the localization business know that cost is a direct result of the volume. We charge by the word, I don't know how much more obvious it can be. The more words there are, the higher the cost is and all of the various aspects that go with that, even beyond words, highly driven by volume. The technology choices that we make, the tools that we decide to pull out of our toolboxes, are based on volume. They are based on how in the world do we process all this content? This content that is coming at us in an efficient way. Arguably, MT's re-explosion in the last five years say is a direct result of this explosion of the volume of content. There is suddenly a need a to handle all this growing volume in a very serious way. That was less the case 10-15 years ago. And finally, this is a very important point that we overlook very often, volume is a driver of usability. Usability is a term that the localization industry talks about almost NEVER. Usability is a concept that the content producers, the content creation community talks about a lot. They talk about it. They don't do very much about it. The point is that it is something that is higher on their list of priorities. Usability means: the ability for the end customer, the recipient of the content, to get value out of that content. Volume is a very important driver of the end-users ability to use content. And we will talk about that a little bit more in just a second. So where does all this content come from? For once the localization industry can take a step back, because the content first has to get produced. It first has to get created. It's the content authors, technical authors and everybody else who is creating content that are producing all this, that we then have to globalize and localize. I'm not going to talk too much about this but just want to make a couple of points. About, the sources of all this content. One of the things that is important for those of you who are in any kind of technical writing function, or deal with technical writers, may understand this. I think there is a content imperative at work here. And basically what it means is that, content, almost has a drive of it's own. Similar to the 40's and 50's when there was an explosion of nuclear technology, in the post World War II period, Lord Sully Zuckerman called it a technological imperative. The deal was that whatever can be built must be built. The rationale for building the H-bomb and the neutron bomb We did not need those bombs but the technology forced it's way to go forward. I think in the same way, with content we have an imperative. That whatever can be said, must be said. So technical authors and content authors find a need to create content, based on its own forward motion and forward momentum that is very difficult to stop. Another thing is that content production does not have any real controls, on it's own part, that is generated by its own rationale. So for example really taking a careful look at the audience profiles. Who are the consumers of this content? Why do we need this content? What are they going to do with it? So often, and I deal a lot with technical writers and authors of content that surrounds products, and their description of their audience is anyone with a pulse. So that leaves the door wide open for creating massive amounts of content, once again, no controls based on the content itself. How about usability testing? There was a time when that was a mantra in the content creation community. That's gone. Hardly a company exists anymore that does usability testing anymore on their content. They may still do it on their products, but they don't do it on their content. So they have no idea what the end users are getting from this content. No idea what value there is. And how about integrating the content with the product? Really working at how the content fits with the product, what voids it may fill. What are the real needs for documenting? There are very companies left that do anything like that anymore. The instruction simply seems to be write it down, just write it down. I think you can count on one hand, companies that take this kind of issue very seriously. For example I would mention Apple, the iPhone is the only product I personally have ever, dealt with that came with no product documentation whatsoever. The product came out of the box and there was no paper. And that because the effort was made very forcefully to integrate the product with the content rather than create a product and then document a whole bunch of stuff on top of it. But unfortunately that is the paradigm that we see so much. The controls that there are on content that there on the volume of the content, really are external. They are derivative, they come from somewhere else. So the time that we give content producers to create content is driven by other factors. Not by the content itself. Cost is driven by other factors. The technology that we give them, put at their disposal, is what drives the volume of the content. Not the other way around. So we find that the tools ultimately making decisions on what kind of volume will be generated. Even translation is an external force that provides controls for content. And we see this in the translation community , in the localization community. Customers finally find out how much their content costs them, when they come to us and ask us to localize it. They suddenly realize it: Oh my God, It is going to cost how much? Larry Ellison, Chairman of Oracle, 3/4 years ago, mentioned localization for the first time, in a public forum. Unfortunately it was not very good for localization because his remark was we are spending way too much on localization and the reason for that, of course, was that there was way too much stuff to localize. But that control did not come into play until the translation and localization suddenly became a requirement, and ballooned out of control. One of the things too that drives this volume that we have, is that we worship reuse.We have created systems like content management and translation memory systems, that put reuse on a pedestal. They create essentially untouchable gods out of these assets. Their value to the process has become far disproportionate to the actual value they contribute. We have developed them into these sacred objects, that can't be touched. So that continues to drive us forward on the volume as well. So the response to that, is let's try and put some kind of restraints on this. Let's control content. This is when you hear about content management. You notice it is not content removal and it is not content elimination, it is content management and that is the effort that gets put in place with all this volume. They say we have to find a way to store this content, label it, create meta tags and meta content and by the way, I think it is ironic because we need to manage so much content, more volume, the way we are going to do that is create MORE content! About the content so that we can manage it better but I digress. We have to be able to find things in this vast morass of content, or sort it so that we can work with it more effectively, to make open up to everybody so that everybody can have access. What we are really looking at is a giant walk in closet of content volume that is infinitely expandable. So it's like your dream walk-in closet in your home, that you can just keep on expanding ad-infinitum. That's the response to so much volume. Another response is Let's lower the costs since there's so much of it, we will lower the costs, by adopting a couple of techniques, off-shoring, we've heard that mentioned here a couple of times. But what off-shoring really is, is moving the problem. We'll take this problem and move it away someplace else, it's less visible and further away, and it's cheaper, so hooray for us! Another thing we have see implemented in this process is to say we'll eliminate steps. Since there is so much stuff and time and money and everything is in a crunch, we'll just get rid of things we used to do. This famously happened with content producers who got rid of editors. I recall Bill from IBM, who went around to his folks in the late 1990's and asked his content creation teams what they had done to fix this volume problem, this efficiency problem. And they said they eliminated the editors, and he's not an author and he found that somewhat strange and he asked, That seems a little bit odd, why did you eliminate the editors? And the content author said, well, no one knew what they did. The idea here is that we simply eliminate the step that we think doesn't have any value. And all we are doing is shifting the burden of dealing with this content and volume to the other people in the supply chain, who are going to have to figure out how to deal with it. Yes we have lowered the costs, so chalk up another one for us. What we haven't done is address the fundamental problem. Then there is the ultimate quick solution which is we are just going to remove content. I think we even heard an example of that today from .... uhh I don't remember who it was, where in the Thai version of Office, everything is beautiful and you click help and you get English! The reason that you get English is that the decision was made to cut that content from the localization. If we just take it out, we've lowered our costs and once again, hooray for us! But all that does is treat the symptoms of the disease. The disease of the volume of the content and the fact that we have to deal with it, is still there. We've just covered it up by taking some of the pain of the cost. It is immediate soothing treatment but it does not solve anything. One of the great solutions is to reduce time. The way we can control costs and reduce volume is we'll just spend less time on it. Does it strike anyone as counter-intuitive, that the way we are going to fix the problem is by spending less time on the content? It seems to me that this is counter-intuitive. But once again we are treating symptoms but not the cause. What it boils down to is a much deeper challenge that we are facing, with all of us in the business of producing and localizing content, and that's the quality issue, and someone mentioned it in the audience issue earlier on. I know a lot of you are going to say, oh god he's going to talk about quality. And the translators are saying, hooray, he's going to talk about quality, but I think it is important for us to have the proper perspective on what quality is. Lets not talk about quality models, lets not talk about systems, lets not talk about standards, sorry Michael, we can talk about them another time, not that they aren't important but as service providers in this business we need to get to a much more fundamental definition of quality. Peter Drucker got it right. Quality in a product or service is not what the supplier puts in, it is what the customer gets out and is willing to pay for. Something isn't high quality because we say it is, or because we took a lot of time to work on it, or because we put a lot of parts into it, or because it was hard to do, That's complete nonsense.Customers, our end customers will only pay for what is of use to them, and what gives them value.That's the only measure of quality that we should be talking about. So it means that producers of content and globalizers and localizers of content need to be working together more, to help make the determination on what is the quality of this content, and focus almost radically on that. Otherwise we are doing a disservice to our customers ultimately by not helping them produce a quality content product for their end customers. One of the things that we do, again harking back to our earlier challenges, is that we rely too much on technology. I hate to say it, but technology is really dumb. As much as we want MT to be really smart, and really know what it is doing, I don't think we are at the stage yet where we could call MT intelligent. I think it is making great strides. But ultimately MT, and I don't want to pick on MT, but any of these technologies, TM especially, but all kinds of other CMS, you name it. They can't help us determine what quality for the end user is. Or how to create it. Here is an example of a sentence, from a piece of content, from a very well known customer, who has translated this sentence and many thousands like it into 36 languages. If you read this sentence. You will discover that you have learned nothing. The reason you've learned nothing is because, the sentence does not tell you anything. This is a stupid sentence. This sentence has no reason to exist. And yet it is part of content that gets used, recycled, stored , sorted and continued to be built upon. Onward and forward and forward, and no matter how much we efficiently translate and store it and reuse it, we are still serving up a stupid sentence to our end customers. So that is not creating quality for our clients or for their end customers. You probably cannot see this too well. Another example of what technology has done, technology can't see alternative possibilities, technology can't present you with other options for how to present the content, to your end customers. Here is a case from the help system from that same product from a very well known company and what no technology can do is create from this top example, a helpful, succinct, useful, usable type of content, that is shown in the bottom example. Technology cannot help you do that. We can translate that and process it very efficiently. But we can't change it into something useful for the end user. Some would say, hey, but the crowd is entering the picture. And the crowd can do many many things. The crowd can help us solve this problem. Despite the fact that the crowd is actually helping create the problem. Because it is creating much more content for us to have to deal with. This is something that we are busy sorting through, and I am not sure what I believe yet, in terms of crowdsourcing and many of you are probably in the same boat. But it seems to me that we should be careful about reaching for another kind of solution here. If it isn't fundamentally addressing the problem. Is crowdsourcing a false solution? Are we in fact trading the infallibility of experts, who used to be the ones who created content, for the so called wisdom of the crowd. Are we really just replacing one god with another? I think that is something we have to be very careful about. Crowdsourcing just like technology tends to emphasize quantity over quality. And we have seen this today, and have had discussions with many of you already, sometimes more is just more. More isn't better and more isn't more helpful. It is just plain more. I do think that crowdsourcing fills a void right now because what it means is that as a community we are still in the search for how to produce quality. We are still looking for ways that the quality can be found. The crowd may have a contribution to make there. I think it is important to not put too much faith in that solution as it comes down the road and develops. And as part of that thought it is important to remember that not all content is created equal. We tend to think, especially in the localization side of the business that content is this giant block of ice, whatever it is, whether it is help or instructions or marketing documents, that this block of ice can be pushed into our system and processed through it, and come out. But not all content is equal. We have to be careful in determining what types of content need what type of treatment. And here the MT community is helping us very much with that process, of making that determination. But so much of the customer side of it wants desperately to believe that, all content can be processed the same way. And that all content can be treated the same way. It is important for us to keep a distinction about that as we try to help our customers, and create that quality for their end customers. Finally, I am a great believer in not looking for a grandiose, solutions, that will come down from the sky and, solve all our problems at once. I think there have been few such cases in the history of mankind. Mostly, problem solving is a step by step process. And there are three things that we can do as, as writers, as translators, as globalizers, as project managers, as those in the community of helping our customers create and globalize content, that will help them improve quality and create more quality qucikly. These are reflected by research done by Jacob Nielson, a world reknowned web usability guru. His focus is strictly on how does the end customer get value out of content. He doesn't care about writing, he doesn't care about translation, He does not care about any of that. He just cares about how does an end customer get value. The very first option, by far, is to reduce volume. Every research study he has produced says, if we cut volume of content, we improve usability for the end customer. It is very very clear, teh number one way. The second way is to make the information scannable and have the information found quickly. A lot of use are going to talk about content for the rest of our lives perhaps. About content, how we produce it, how we localize, how we store, how we translate it. In the end, the end customer does not care about any of that. The end customer just cares, can I get the information I need, when I need it and have it be accurate and correct. That is all they care about. One way to do that is present the content in a way, that they can find what they are looking for quickly. And the Third method that Jacob Nielson identified, is to stop selling. We spend so much time selling to end customers with the content, we need to present the information and let them move on with what they are trying to do. Instead of constantly marketing to them, which companies find very hard to stop doing. And so I will leave you with this final thought. As we discussed the nitty-gritty of translation, localization tools and process, and we argue about what is more important, content or design, or format or protocols or all these things, None of that matters. Most important for the end customers, is use. And the more we can help our clients to get their end customers to get good valuable use, the better service we will be able to provide to our customers. And the more value we will be able to create for ourselves in the process. Thank you.

Video Details

Duration: 37 minutes and 55 seconds
Country: Thailand
Language: English
Producer: Asia Online
Views: 160
Posted by: asiaonline on Dec 12, 2009

Localization and Translation Thailand 2009
Day 1 From World-Weary to World-Ready: Meeting Today’s Content Globalization Challenges
-Hans Fenstermacher, GALA and Vice President
Twitter: #LTBKK

Caption and Translate

    Sign In/Register for Dotsub above to caption this video.