Starting from the last meme launched by Julian Wraith on the Future of Content Management, an interesting debate about the latest up to date definition of the “ECM” definition (and collaterals such as on the EWCM) was launched.

Do we need a new ECM definition (again)?

First of all, let’s rapidly review some of the various definitions for the ECM term. I must admit that there are probably as many definitions of the ECM term as analysts or ECM companies around the world.

The situation is even certainly worst than that because the perception of the market is still different. Most of the time you can hear: “Our ECM, do you mean our Document and record Management System?” … and ECM = DMS/RM in the spirit of most customers as if the content of an organization was only consisting of some Word, Excel and PDF files stored in a structured repository.

DAM: “An image catalog for the marketing folks?” WCM: “A content silo maintained by a PHP geek?” The Corporate Portal: “Some fancy webapps bundled with their own DB?” SoCo software: “The new Playmobil enforced by the HR department without the consent of the IT dept?” SharePoint: “As a CTO I condemned the use of such a tool 12 months ago and I now have a dozen of illegal installations in my departments, what should I now do?”; Search: “Let’s move first to Windows 7 so that employees can already start find their files on their local C drive”. Shared SMB drives: “A big mess consisting in 90% of our company knowledge”. Etc...

And clearly speaking all these “content silos” could be classified from one manner or the other as “Enterprise Content”.

Let’s take now the most common ECM definition:

AIIM defines it as: “Enterprise Content Management (ECM) is the strategies, methods and tools used to capture, manage, store, preserve, and deliver content and documents related to organizational processes. ECM tools and strategies allow the management of an organization's unstructured information, wherever that information exists.”

http://www.aiim.org/What-is-ECM-Enterprise-Content-Management.aspx

Does this mean that structured information is not part of ECM?

Even Wikipedia is not clear about what ECM is. The first thing you can read by coming onto the Wikipedia ECM page is: “This article may be confusing or unclear to readers. Please help clarify the article.”

You can also read about WCM:

 “Enterprise content management claims to integrate web content management (WCM). […] It is however worth noting that many in the industry do not consider WCM as an integral component to an ECM system. There are very few examples of successful implementations whereby a shared repository for documents (the core purpose of ECM) and web content are managed together. Indeed very different techniques and philosophies to structure and organize content are utilized for external facing web content than for internal facing document content.”

And here again this is heavily based on the assumption that Enterprise Content = Document Content.

Two years ago Russ Stalter already did a similar blog post: “Disillusioned with the current definition of ECM” which is listing additional ECM definitions (e.g: Gartner one). I am sure you will be able to dig the Web and find the latest Gartner, Forrester, Gilbane or CMSWatch definition of the ECM term, all with different meanings.

AIIM also developed the following model to explain ECM:

I already prefer this one because it tries to better distinct:

-         the global Enterprise Information Architecture (grey background)

-         the various Content Enabled Applications which are running on top of it (green boxes)

-         the processes, strategies and methods put in place to manage Enterprise Information (top of the graph)

In such a vision ECM means what is semantically means: all content managed within an organization whatever the application, if it is siloed or integrated with other or if the content is stored directly on a hard drive, on a web page or in a Corporate IM chat session.

The most recent definition is the one of Laurence Hart – What ECM Needs to be Today:

“Enterprise Content Management is the empowerment of all content within an organization. This is accomplished through the centralized management of content, allowing for people and systems to access and manage content from within any business context using platform agnostic standards.”

This definition is then closer to the one of the Gartner regarding what they call Information-centric Architecture (as of July 2008) and also quite similar to what I described in one of my previous blog post as a CCII: “The information-centric infrastructure is the technology building block within an organization's enterprise information management program. It represents the principles, models and requirements necessary to flexibly share and exchange information assets, as expressed within the enterprise information architecture. This infrastructure's architectural style is modular, distributed and loosely coupled. It features the alignment of metadata, semantics, standards and content formats to achieve the consistent and seamless capture, persistence, transformation and delivery of content across the enterprise, independently of how it is stored in source systems or packages. It thus breaks the tight binding of content to applications — a critical first step in increasingly service-oriented, context-aware and cloud-based computing environments.”

We can then distinct two positions: the one which are tending to say that the future of ECM is to become the information backbone of any organization and to offer all basic “library services” (e.g: versioning, check-in/check-out; metadata; file plans, permissions, observations…) any content item might need whatever the application whichcreated or is using it and others which are saying that ECM is the whole (aka the Information-centric Architecture + the Content-enabled applications leveraging the Information-centric Architecture + the processes and methods around it).

In my humble opinion, I would rather prefer the first solution that’s to say to focus ECM usage on:

ECM = Information-centric Architecture

which empowers all the compliant Content-Enabled Vertical Applications (CEVA).

Finally at the end of the day everything is content, isn’it? So an ERP, a CRM or an HR system could also perfectly leverage or even be built on top of such an Information-Centric Architecture. However they are not part of AIIM graph. Annd considering that all applications of a company are now part of the spectrum of ECM is a bit too vague. This is also perhaps why the “ECM Suite” term does not mean today anything any more.

The bottom-up approach: Content Interoperability Standards

Such a new usage of the ECM term is also strongly related to the recent content interoperability and standardization efforts initiated during the last years (= mainly the JCR and CMIS standards).

Even if it certainly took and is still taking much more time than originally estimated, the (J)CR standard is rapidly standardizing this “Information-centric Infrastructure”. Similar to the EJB 1.0 specification, the JCR 1.0 was clearly not covering all required features required in order to develop some really portable Content-enabled applications. Release after release, this gap is being shorten.

So the JCR tends to not only become a content repository but to act more and more as a “CEVA container”. Who knows perhaps after the Servlet container, the Portlet container we will see a new generation of Content containers.

It will then become possible to develop a content-enabled application from a truly content vendor-agnostic manner and deploy it on any kind of Information Infrastructure already in place be it an open source stack or a proprietary one, a low-cost commodity or a high-end solutions.

Is such a vision a long term utopia?

Yes, certainly. Most content items are still siloed in legacy systems. CMIS or JCR (or other content interoperability standards if any) will take a lot of time. And it will need to face the classical vendor lock-in situation.

But there are also a new generation of Content Management software vendors which are trying to change this blocked situation. Here at Jahia we are for example currently fully refactoring our WCM CEVA application on top of the JCR+CMIS standards. Even if there are still lots of grey undefined areas between the JCR API, the middleware-oriented components and framework running on top of it and the CEVA application itself, we hope to make it independent of any Information-centric Architecture as soon as possible. Day Software is also doing a great job to better distinct its Information-centric Architecture (CRX) from its CEVA WCM (CQ).

Nuxeo, an open source ECM vendor, is now selling their core ECM as an Enterprise Platform which is “the perfect fit for CEVAs” (http://www.nuxeo.com/en/products/ep/). ExoPlatform, another open source CM vendor, recently contributed their core JCR library to the the RedHat Jboss middleware, in order to focus on their CEVA applications.

Strangely, for one time, European open source driven companies looks like precursors in this evolution of the content industry. In all the cases, the fact that CM vendors starts to better distinct what is part of their Information Infrastructure stack and what is part of their CEVA stack definitively signals the end of the tight binding of content and applications.

So time to change the definition of ECM?