Today, as always, globalization remains an afterthought in the majority of cases. No matter if this is immensely more expensive, or totally unreasonable, it's the way things are.

But content is now a hot topic, as are content management systems (CMS) and the whole area of enterprise content management (ECM). The ECM concept means developing an enterprise-wide methodology to handle all enterprise content. If the enterprise is global, it follows that ECM must deal with global content in a variety of languages.

This is where globalization management systems (GMS) come into play, helping manage the cost of translating a constantly growing body of content, and helping reduce the cost of maintaining all these translated documents. This article examines what happens when CMS meets GMS.

In the best scenario, the merged GCMS will spread globalization awareness throughout the enterprise, from the translation boiler room up to the corporate boardroom, so that for the first in time in history, globalization will cease to be an afterthought and become one of the primary requirements for any significant development. Enter the global information age.

What is content?

Inasmuch as "content" is a form of data, I believe that to truly understand what content is, we need to position content on the well known data pyramid: data-information-knowledge-wisdom.

This pyramid was developed by Harlan Cleveland in his 1985 book The Knowledge Executive: Leadership in an Information Society, and represents for me the past, present and future of computer intelligence—from raw-data, to human-useable information, to self-descriptive machine processible knowledge, to non-deterministic context-dependent wisdom. The pyramid itself was inspired by the following poetic passage:

Where is the life
    we have lost in living?
Where is the wisdom
    we have lost in knowledge?
Where is the knowledge
    we have lost in information?
 
       from The Rock
       by T.S. Eliot (1934)

If "content" is so important, why did it not make it onto the pyramid in the first place? The reason, of course, is that "content" is a fairly empty word, and so politically correct it does not actually say anything! (And "knowledge" is still tainted with the failures of artificial intelligence).

The figure below illustrates where I believe content fits on the pyramid:

Content is a cautious first step (for enterprise computing) in the area of knowledge management. Content management takes "information" (documents) and adds a few XML tags to help manage the information. This allows people to re-use the knowledge stored in documents. When machines can re-use the knowledge on their own (possibly via sophisticated semantic tagging), we will then have genuine knowledge management.

This is not to say that content is not important. Quite the contrary, content is very important because it is a realistic first step that builds on information management and brings us closer to true knowledge management.

Where is content in the enterprise?

How much content awareness is there in the enterprise? Has it reached executive levels?

In point of fact, a few large enterprises, seeing the trend, have appointed Chief Knowledge Officers; some have also appointed Chief Content Officers (CCO). The following table shows the search engine hit counts (obtained at www.google.com) for various corporate titles:

We can see that CKO and CCO are still emerging concepts. The impact of content in the enterprise has not yet reached its climax. The time to prepare for the globalization revolution is now! Translators of the world unite!

How solid is the content trend?

Is content just a fad? or is it a fundamental trend? How did we get where we are today? What other trends co-exist and possibly interact with the content trend?

The following whiteboard extract is an attempt to better visualize several such trends evolving over time to gain some insight:

The DATA row is just another way of representing the data pyramid as a trend. It shows content again sitting between information and knowledge.

The LOGIC row shows the trend from traditional software, which contains a lot of logic and a bit content, to Web sites, which contain a lot of content and a bit of logic. This has fueled content hypergrowth in the following two ways:

1. as companies discovered the Internet, intranets and extranets, the advantages and simplicity of browser-based access to information became obvious, and more content was needed,
2. with the increased connectivity of the Internet, the world became smaller and the need to do business with customers and suppliers in their own language also required more (translated) content.

Finally, while software development had always been under R&D's jurisdiction, Web sites were often initially handled by the marketing department. With both R&D and marketing now facing global content creation and management issues, whether they agree or not, the "content" buzzword is hitting the corporate agenda.

The FORM row shows the trend towards processing unstructured data. A long time ago, when ORACLE started working on text summarizers and similar tools, it estimated that 90% of the world's content was not structured. The advent of the Web eventually proved their vision quite insightful.

The TECHNOLOGY row shows the trend in enabling technologies. While relational databases were the dominant method for handling structured data, it wasn't until the advent of XML that unstructured data found its niche. The immense success of XML as a pervasive enabling format has fostered the development of content management tools.

The MANAGEMENT row shows the trend from personal productivity tools to the workflow tools of today. The complexity of managing Web content and, a fortiori, the greater complexity of ECM, demonstrate a clear need for workflow technology.

Content is not a fad. Several major technologies are converging towards ECM: workflow, XML, Web, etc. Content is a fundamental, inevitable trend because information is the most precious resource of the information age. Content management signals the beginning of our ability to handle that unstructured 90% of the world's information.

The merging of CMS and GMS

The ultimate merging of CMS and GMS into a single integrated system for managing global content (GCMS) seems inevitable. This is not new; many analysts, myself included, have been predicting this for some time. While it may not happen tomorrow, there is increasing pressure from the market for multilingual management capability. After all, 66% of the Ws in WWW stand for World-Wide.

More and more alliances between CMS and GMS are being announced; most GMS try to work with all major CMS, although some combinations are better integrated than others. IDIOM for example have developed a version of WorldServer for Vignette entirely compliant with the Vignette Application Framework API. Similarly, GlobalSight have always been very tightly integrated with Interwoven.

In fact, a major design imperative for GMS systems is to integrate as seamlessly as possible into CMS systems. Both have workflow systems, both manage content, both have workflow definition and management interfaces, both need to understand roles and responsibilities, both must support testing, etc. Will all this be duplicated? Will there be two different interfaces for the CMS part and for the GMS part? No there will not! And this is why seamless integration is required.

In short, rather than simulate a unique integrated GCMS with a very tightly integrated CMS+GMS pair, why not merge them once and for all into a real GCMS? After all, does the customer really need two workflow systems? Does he want to pay for them?

The strategic disadvantage of a CMS+GMS pair is that the GMS comes second—globalization is still an afterthought once content has been authored. The advantage of a truly integrated GCMS will be to present globalization issues up-front for maximum leverage.

The 50% effect

It was a great day for the globalization industry when Microsoft sales revenues for the rest-of-the-world exceeded the 50% mark, thus beating domestic sales. It was around that time that resource files first appeared, and localization started to become more rational and less masochistic. The "50% mark" is simple, effective and sometimes the only way to wake people up to the reality of globalization!

Then the Web came along and things got messy again.

But as CMS and GMS merge, as enterprises start managing their content globally—as they certainly must in order to achieve economies of scale, consistent branding and quality—many will soon realize that more than 50% of their content is in "other languages". The "50% effect" may again trigger a new "globalization gold rush", where there will be much demand for content globalization services. This is a unique opportunity and our industry should position itself relative to these upcoming needs.

Our key advantage: Our industry has probably managed more content in more languages than any other industry.

The big question: Beyond the usual set of translation and localization services, will we be able to parlay our unique experience into a more targeted and valuable service offering, ultimately aimed at managing global content in and for the enterprise?

Pierre Cadieux (pcadieux@i18n.ca) is president of i18N Inc. (www.i18n.ca), a firm specializing in internationalization training and consulting for embedded systems, shrink-wrap software and Web sites. Pierre is the Technology Editor of the LISA Newsletter. Formerly VP Technology at Alis, he pioneered the transparent handling of Arabic and Hebrew languages and created the core bidirectional technology licensed by Microsoft. As Director of Localization Technology at Bowne Global Solutions, he published the first generic model of Web globalization technologies. Pierre also regularly presents workshops at LISA events.