Technology of Content

This is a second part of my response to Julian Wraith’s future of content management thread; the first part was more about the technical decisions, this one more about the architecture, and responding to some of the other issues. I have a new post which is more of a general view on content enabled vertical applications

Stéphane Croisier in his post in the thread says

There is currently an unclear separation between applications frameworks and content infrastructure. But at the end of the day everything is content and every application has first to deal with content items rather than with processes, states, UI components or other application oriented paradigms.

In my general work in content management I think this is one of the things that has become very clear, and that “unclear separation” is very apparent. First content; it is a very good start, and every project needs to be grounded in content, and in the structure of content, the architecture of content and the user IA. However, the processes, states, UI and other parts of the web application are beginning to dominate projects. In the Gartner terminology we are building CEVAs (Content Enabled Vertical Applications that is), as integration, process, e-commerce, CRM parts of the project start to dominate the requirements over the purely content based parts.

It seems that most of the contributors to Julian Wraith’s future of content management thread who mention it see content management moving to a clear split between repositories (Common Content Information Infrastructure as Stéphane calls them) and applications and content management systems and CEVAs implemented on top of these.

I don’t think we can yet see what the successful content infrastructure stack will be; as I said in my earlier post there are technical decisions that have to be made that there is not yet agreement on (except between me and Peter Monks!) and the existing putative standards (CMIS and JCR) do not extend far enough to take a position on. But we can see that this is the way things are going. Quite clearly the standards for the infrastructure will be open, and most implmentations will be open source. There will be some vendors who do not embrace standards, but they will need to be the few large ones or they will lose out. Infrastructure environments remember (think Linux, Apache) are mainly open source, although there is scope for proprietary layers at the very high end (think Amazon, Google).

At the application layer, as Stéphane says, everything is a mashup, content from different systems, content from other APIs, this is the we application layer. It needs to be content aware, very much so, but it needs to be an application development environment. This is where most people will see the value added in the content management business, although in fact the value here is in implementation, design and integration services, not the technology itself. Application development environments no longer make a lot of money, and again they are dominated by open source (think Java, Eclipse, JBoss, Django).

Once you take out content infrastructure and application development, and the other tools like search, workflow, there is a core of tools for working with content, to support reuse, refactoring, cleaning, import and export, that one might call a Content Workbench. There is a lot of potential value if these types of tools are the value added end of the business, as they can differentiate vendors and add value. Interfaces for merging changes and so on would be part of this type of toolkit. This is the stuff where good UX means timesaving for content workers, but it is difficult to build on a customized per-project basis, so this still offers value from a particular vendor.

Overall then we see a picture where the monolithic CMS starts to break apart into infrastructure, application and toolkit layers, that can perhaps gradually be mixed and matched together to build content applications. We are just seeing the beginnings of this now.

6f82f1d2683dc522545efe863e5d2b73

Tagged: CEVA, CMS