Information Flow

You often hear people talking about “end-to-end” solutions. In the information development business it might be end-to-end translation management or end-to-end content management and localization “all-in-one”. But just how end-to-end are these solutions? Are the ends really there?

It seems to me that the assumption behind this claim is that there is a set of information in a company which can be regarded as independent of all the other information produced. An “end-to-end” solution for technical publications (including perhaps its translation) is actually a myth. The information created in technical publication departments is inextricably linked with, in fact partly based on, information which is created elsewhere – namely in the marketing and engineering departments. These departments start thinking about the product usually some time before it gets to tech pubs. Even in a really slick agile information development environment marketing should be ahead of the game.

So the front “end” of a typical content management system is not really an end at all, but actually a pretty porous interface in terms of the knowledge it needs to share with other applications – engineering applications or web content management systems, ERP systems, parts databases, and so on…

And then there is all the information which comes at the other “end”: after the customer has bought the product information continues to get created to meet after-sales, support and maintenance needs. Of course this isn’t really an end either, but should probably look more like a structured hand-over of responsibility for customer communication. In terms of the infrastructure, this most likely means that information needs to flow from a content management system (or even several of them) to a support database or issue-tracking system, or both. By the way, translation isn’t really an “end” either. Studies have shown that typically less than half of all content ever gets translated at all, and an even smaller proportion will be translated into more than a handful of languages.

The fact that there is no such thing as an “end-to-end solution” for product content has certain implications. For instance it should remind us of the importance of content standards in coordinating customer communication, and the importance of making those standards available across different technical platforms. The only way to ensure that is to demand open standards and interfaces everywhere, allowing information to flow wherever it is needed. There’s a great discussion of the role of standards in localization of on Kirti Vashee’s blog, but in fact information needs to flow between all the content repositories in the organization as well as in and out of localization. The beauty of web services is that the integration between system doesn’t have to complicated, as long as customers demand it upfront and don’t try to bolt it on as an afterthought.

I am just recently back from our user conference, the aQuatic user days – you may have seen Kirti Vashee’s blog on in already. We had around a hundred delegates from mainly West Coast customers, but there were lots of really interesting ideas floating around. The most interesting for me related to a thread that has been cropping up at all sorts of events I have attended recently: the importance of the community in customer engagement.

Francis Tsang, Senior Director of Globalization at Adobe started the ball rolling a three weeks ago, in a great presentation he gave at the CNGL meeting we were attending together as members of the CNGL Advisory Board. Instead of talking about technology he devoted the entire talk to the customer and how Adobe is working on customer engagement.

The big issue here is that the community is becoming the most important resource for customers interacting with, wanting to find out, or complain about, your products. And the catch is that right now there is almost nothing you can do about it; the content is created by people who don’t work for you. It is no longer one-way traffic – with companies creating content about their products for consumers – but rather it’s a relationship. The community creates valuable content, which companies want to make available to as many of their customers as possible. Also at the CNGL meeting was Greg Oxton from the Consortium for Service Innovation – and he was talking about pretty much the same thing.

What does this mean for Information Quality? Well, some people say that IQ doesn’t matter to the community. But they are wrong; majorly wrong. Let’s just think about product forums for a moment: you want your forum content to be available to the community, not just to the person or people following a particular thread. So you need to make threads findable and comprehensible:

• Findable means that they can be indexed effectively for search (aka SEO), which ultimately comes down to tying your corporate terminology and taxonomy to the community jargon and folksonomy. Findable also means that you can find information in a language which is not your own.
• Comprehensible means that once you have found some information you can understand it, even if you are reading it in a language which is not your own. You will also, as a company, want to know what your Chinese user community knows, and what they think of you. You might even be thinking, like most high-tech companies are, of translating forum content on demand with MT. IQ is the critical enabler for MT.

For all of these reasons, IQ is critical to the success of a strategy of “embracing your user community”. But obviously the idea of controlling the content is not going to work. The best we can hope for is to nurture good content. In much the same way as marketing departments are learning to get beyond the idea of controlling their company’s message; the after-sales experience also needs to embrace the uncontrollability of the community.

This post is too long, I will write more about how I think you can nurture community IQ another day…

The history of machine translation is a fairly turbulent story of boom and bust, broken promises and shattered dreams (and some spectacular successes). With the advent of Statistical Machine Translation (SMT), history has repeated itself with uncanny precision. I would say we are now between Phases 2 and 3 of the famous Gartner Hype Cycle (which, annoyingly, is not a cycle at all):

For the last 10 years or so, researchers have been telling us that SMT represented a revolution in machine translation; that there would be no more linguistics, no more rules, just data. “Give me enough data,” went their motto, “and we can work miracles!’. Well Google has enough data, but they still have their limitations. I tried some things out:

1. Ich will diesen Satz übersetzen
   Google says:  I want to translate this sentence

Perfect! If only we could always write like this…

1. Ich will nur dass diser Satz richtig übersetzt wird
   Google says:  I just want that images this sentence is translated correctly.
   Should be:     I just want this sentence to be translated correctly

This is not so great. It just doesn’t make sense – where did “images” come from? So what went wrong?

Well I made two simple mistakes. There should be a comma after “nur” and “diser” should be “dieser”. Let’s fix those issues and try again:

1. Ich will nur, dass dieser Satz richtig übersetzt wird.
   Google says: I only want that this sentence is translated correctly.

Much better, but sounds a bit funny to my (British) English ears. It’s not wrong, just a bit stilted.

Now let’s try something a bit more difficult (although by no means unusual):

1. Ich möchte bitte den Satz übersetzen lassen
   Google says: I would like to translate the sentence

Now this translation sounds good (if you don’t know German), but it’s actually *wrong*. It should be “I would like to have this sentence translated”. A subtle, but quite possibly technically significant, difference.

Now I am not Google-Translate-bashing here; SMT is a great technology; but it’s not magic. Errors in the input to these systems will always lead to unreliable results – yes, you still have to care about the quality of your source content.

You also, by the way, also still need to care about branding, compliance, and liability in your source content – these issues won’t not look after themselves by magic either.