My colleague Chip Gettinger and I have a long-running online conversation about various intersections between his world of content strategy and mine of software and content globalization. Recently, we brought SDL’s Alberto Andrade into the mix.
Chip, as SDL’s VP of Solutions Consulting, helps customers with content systems and strategies for documentation, while Alberto, in a parallel role as VP Solutions Consulting for web, specializes in helping customers with web content management strategy. Chip works mainly with structured content, while Alberto has up until recently focused on mostly unstructured. I’ve found that both topics are increasingly coming up together in customer conversations, and so I took the opportunity recently to talk with Chip and Alberto about the intersections and dividing lines between their two domains, and how both are influencing, and influenced by, customers’ need to go global.
As a preamble to the conversation, some definitions for this blog post:
- Structured content refers to content that can be stored and managed in a very organized, hierarchical and searchable fashion
- Unstructured content can be content in a variety of formats such as documents, video, blog posts and social media, which that do not lend themselves to easy organization and searching
The SDL systems we refer to in this article, SDL Web (web content management systems, WCMS) and SDL Knowledge Center (component content management system, CCMS, for structured documentation) together successfully handle both structured and unstructured content. So I’ve wondered how to best differentiate the two in my own conversations. As you’ll see below, Chip and Alberto are expert “teachers” in these domains and the content convergence happening for our customers.
CG: Not at all! Both types of systems manage content in hierarchies, have workflow, language support and provide strong user management. Both use XML, although it’s two different flavors of XML: “well-formed” XML for SDL Web, and DITA-based XML for Knowledge Center. Both types of systems have a lot in common, but they evolved differently based on the source content needs.
AA: That’s right. Back in 2001, companies started seeing the need for content reuse, such as in online law libraries where the same pieces of content can be relevant across many cases. This led to chunking and hierarchical separation, first in a web-centric view of the world, and then in structured documentation.
As you know, Chip and I are constantly traveling for our work. So a “near and dear” example of reuse is airline websites. There, every time they add a new destination, the airline needs to describe the plane that will take you there, such as an Airbus or Dreamliner. That description already exists. It would be expensive to continually recreate it. Destination descriptions are also mostly the same. So reuse is common to both web CMS and CCMS.
JR: Let’s talk a bit about content globalization. I’ve seen in past conversations with Chip how reuse in source content is really critical for translation.
AA: I agree, absolutely! There are huge globalization implications. If you don’t have reuse in your source content, then when you go to translate, you may find yourself translating similar content over and over. Whereas with the right tools and processes, you can streamline at the source and then use the same translation as well. Why should we pay again in time and money for the same work?
Just think about how many times your translation work is duplicated if you have to do everything for web, and then video subtitles and captions, and then translate again for documentation! And then, multiply by the amount of target languages! This is a daunting and expensive task. SDL Web CMS and Knowledge Center CCMS both have built-in the notion that you get your source right, then globalize that, and then you publish out to relevant formats.
JR: You’ve talked about similarities between Web CMS and Component CMS. What do you see as the main differences?
CG: It’s also useful at this point to review some of the history. Content management was in its “Wild, Wild West” stage back in the late ’90s. Traditional websites were flat, with hierarchy created through taxonomy, and a focus on navigating easily around the pages. In the document world, writers had to create hierarchy. They made it like books, with chapters, headings and sections. With documents, writers have the control; with websites, the website architect is key for navigation. So that’s a big difference in the evolution of the two areas.
If you look at product documentation, the content tends to have a long life cycle, with many updates and revisions to reflect new product features. So you have to maintain version history. In regulated industries, it’s also critical to have a rich audit trail. Some industries, like manufacturing software suppliers, also have to maintain older versions of the documentation and still have new versions. Because product support goes on for years, their customers stay on the same software version for a long time. So patches and software updates are needed for older as well as the recent product releases. In contrast, websites get updated often with only the latest version visible.
JR: What about website liability though? Is the content history important there?
AA: That’s an important point to mention. Liability and who did what (and when), can be very important for websites…for example in cases of alleged false advertising and the need to prove it. Or for emergency updates where you don’t want to permanently erase what was previously there, if there is a mistake that needs instant correcting, like using a logo when you don’t have permissions. The ability to track content object history provides a degree of operational certainty.
CG: That’s right – there are different business reasons why you track and version in websites versus documents. Another difference between the two is how the content itself is structured. In the document world, chunking content up into small pieces evolved in the early 2000s into topic-based organization and components.
The other unique thing about documents versus web is how multi-channel output evolved. In the document world, customers often still need to see things in print (i.e. PDFs), as well as online and mobile devices. Knowledge Center evolved as a single source for all these output types. SDL Web also has the omni-channel delivery concept, but with major emphasis on the need for consistent user experience across channels, and by definition, all online.
JR: What about multimedia content, like video and sound files? Is that still different between the two areas?
AA: Well actually, the document world has to deal with multimedia as well now. When was the last time you reached out for a tech manual on a mobile phone? We use videos now instead of, or in conjunction with, documentation. It’s not just millennials – people of all ages are using videos more and more for instructions.
Also the documents themselves, as they move to the web, are becoming more dynamic. And both written and media content have to perform well on mobile devices. For example, now you really need to take into account whether your content is able to switch bit rates in case of bandwidth limitations. Because when you need instructions on the factory floor or CAD drawings on the building site and you only have a 4G connection instead of wi-fi….well, that can be frustrating.
CG: But the reality is that web is eons ahead of documentation, with respect to multimedia. Words are still king in docs. There is a lot to be learned from the web world, including social media aspects and multiple ways to reach the customer. For the most part, documentation is still static, but I see some of our customers moving into media in more aggressive ways. Alberto, I definitely agree with you that integrating video is potentially a huge growth area.
JR: We’ve talked quite a bit about the differences that remain between the web CMS world and that of structured component-based CMS, so that, for now, people really need the two different approaches for different content. But do you see any evidence of convergence between the two?
AA: Definitely on the delivery side. Here’s a great example. I have a favorite watch – it’s very complex, with lots of buttons. On the manufacturer’s support website, manuals are synced with video, chapter by chapter. The videos have narration and subtitles. This is convergence! It was all in pdf format before, and you could not combine that with voice and subtitles.
However, you can see that, as combinations like this become possible, now we have to find the same hierarchy, “chapterization” and reuse for videos that we have already in the document world. This is still cutting edge, and is driven by the need to re-use content economically.
This example I’m giving you is still light years ahead of industry, but it’s coming. And it can be combined with translation, for global content. However, we need some change in the video industry skill set to make this happen. Video people are not technical writers. Web and documentation departments are still “siloed.” But this is going to change, and we see this with some SDL customers already.
JR: So, in future, is it just going to be “the content department”?
CG: I think there will still be a blend of traditional and new approaches based on customer requirements. For example, we have a customer who, a few years ago, trained their tech writers on how to create videos. That is pointing toward that blended future.
On the other hand, a lot of documentation is procedural information. Video has been useful, but there are some tasks where you have to start and stop. Many people still find content is still easier to use for that purpose. I love your watch website example, Alberto, because it gives the user the choice of how they want to consume the information.
We can see a very interesting trend here toward doing useful customer content mashups. If we do it right, we still can maintain a variety of content source types, but customers can consume in different ways based on their preferences.
JR: We are hearing a lot about this concept of mashups becoming more popular. What are some of the challenges?
CG: That’s a great question. I work with a customer who is going through this right now, seeing a convergence of content types in their documentation website. With content mashups, you can go to all these alternate paths for user consumption, including video. But like our customer, if you are managing more than three released product versions, it gets very complex.
So you need a CCMS because you’re still maintaining and making changes to the older product release content. It gets even harder with multimedia content. Most companies make videos that are generic so that they don’t have to re-record for product release updates. Video is still relatively expensive to produce, and it is easy for video content to become obsolete without planning.
With a CCMS, you can just update one topic, and it propagates from there into other “containers,” guided by taxonomy, breadcrumbs and table of contents. Content consistency is easy to achieve with CCMS. There are all these content types, but customers don’t have a unified interface yet. At SDL, we see this as a great opportunity for unified delivery, taking full advantage of content mashups.
AA: Another challenge with mashups is that there still a big difference between CCMS content and web content in terms of standards. I don’t mean there are necessarily different techniques for pieces of content – but there is a history of documentation consortiums working on standardization of formats, like titles and chapters. On the website, like for articles, blogs, PR, no one has set standards that are similar to what exists on the documentation side. This makes it harder to fit the mashup “puzzle pieces” together.
CG: That’s very true. What happened with documentation is that DITA came along. It enabled writers to easily write content as task references, concepts, etc… That evolved over 20-25 years of practice and standardization has really only just happened over the last ten years.
The DITA standard has definitely made content – and writers – more portable between companies. It’s a very precise way to minimize the documentation effort. It supports print and makes it more modular for display on phones and other formats. Some customers embed the help inside their product, like medical devices with help in multiple languages. DITA supports that, and makes it relatively easy to adapt to different delivery methods.
And have you noticed that now, when you want to go out to dinner, you go to OpenTable and Yelp, not to the restaurant website. Company websites are becoming obsolete. We use topics now. We search it up and use a variety of resources, like blogs and social platforms to decide where we want to go. We are used to consuming information in mashup form now.
AA: Exactly! It’s really a mashup revolution in the web content world – just look at travel sites like Trip Advisor, Expedia, Travelocity. But mashup sites do require standards to work.
JR: We have talked a lot about mashup examples in the B2C world. Are mashups relevant for B2B sites too?
CG: Well, a good B2B (and B2C) example is a recent large company merger, where three organizations are blending their documentation content. Luckily all the business units were using DITA, so their content is portable for their product documentation. Company acquisitions and rebranding are very typical examples of B2B mashups. We are working on very exciting innovations now to make it easier to blend content this way. Taking advantage of all the similarities we’ve talked about on the web and CCMS content sides, we are going to be seeing some important evolution coming.
AA: It’s definitely coming! Going back to B2B vs B2C for mashups, it’s pretty standard that nobody does a lot of actual buying off B2B sites, the way they do off of B2C. But in B2B, people want to help themselves solve product questions and issues. They find the answer in structured documentation. They might find they have a missing part. And just think, if that content is dynamic, you can then click on the missing part, get the SKU, and send to a distributor.
Or look at car manufacturers and their dealerships, like VW for example. They have standard content – descriptions and pictures. Dealerships are independent, and they want to have VW’s latest graphics and content so they can use them in their own materials. And VW can charge them a fee to for that service. And then the content goes to consumers, so really B2B2C. The standards not there yet, but this is the goal.
JR: Will it all end up being one system that handles both types of content, structured and web?
CG: We need to think about the source authoring and creation. I believe content will still be authored in separate systems. There are so many specialized authoring tools, for documents, learning management, video, web content, knowledge, etc. The authoring is separate based on governance needs. The customer/user delivery is what converges.
Everyone will author differently and have that flexibility, but the delivery platform will be unified. Taxonomies and facets are used to make that delivered content nice to navigate or search. Think how Amazon has thousands of suppliers, all selling their products independently, but the delivery platform is the same as a consumer to browse or search. We’ll see that very soon with online product content for whatever system it may come from – a video, knowledge article, task, social posting, etc.
AA: An analogy I love is the use of internet-connected cars. We will no longer have dashboards. You’ll bring your phone into the car, and it will control all the separate apps on the car, but the car is the common delivery platform. It reduces the cost for everyone when you can just plug in the apps.
JR: Wow! It sounds like, in spite of the remaining differences between the two types of content, there is a foundation of similarity and a growing convergence….and things are going to be changing a lot over the coming years. You have covered a lot of territory here, Chip and Alberto! Thank you so much for sharing your thoughts with us, and we all look forward to hearing more from you as the word spreads on content mashups and unified delivery.
Thanks to Chip and Alberto for their comments and insights! You can find them, and the SDL crew, at:
Acrolinx Content Connections May 1-3 in San Jose and
Information Developer World (IDW) Menlo Park May 16-17.