A standard with a chance of success

It is no exaggeration to say that the content management world was recently rocked by the announcement of a new standard. The standard (or proposed standard to be correct about it) is called the Content Management Interoperability Services Specification, and has the acronym of CMIS (pronounced CeeMiss).

The intent of the standard is hardly revolutionary, the idea being to promote interoperability between various enterprise content management (ECM) systems. And, in fact, the ECM sector has a long history of "adopting," but then almost immediately failing to actually adopt standards. Take for example DMA, ODMA, DMWare, iECM and JCR 170—all great initiatives, but at best adopted half-heartedly, thereby rendering them useless. So why then the explosion of interest in yet another content interoperability standard? Simple, this one actually has a fighting chance of success.

What CMIS does is to provide an agreed on set of basic services that, if adopted, will allow the interchange of documents across systems. In other words, if I have EMC, IBM and Oracle ECM systems running alongside one another, with the use of CMIS, I can check documents into one system and at the same time let other systems be aware of that action.

Or, as my colleague Kas Thomas recently described it, "CMIS is ... a set of protocols, exposed via REST [representational state transfer] and Web services definitions, for platform-independent interchange of repository content. Using CMIS-defined HTTP calls, you will be able to do standard CRUD [create, read, update, delete] operations against any compliant repository, regardless of the underlying repository architecture. Notably, CMIS leverages the Atom Publishing Protocol [APP] in its REST model (and indeed requires compliant repositories to honor APP, although they can optionally honor additional transfer representations."

So far so good, a solid set of basic services using open standards that everyone agrees on. What CMIS does not do, though, is to agree on more advanced ECM services such as security protocols, rendering, records management or capture. But, in fact, it is the omission of all those other ECM services that means it might well have a chance of being successful. By agreeing on basic essentials and nothing more, there is little contentious territory to argue over.

Another key reason CMIS might succeed where others have failed is the simple fact that it was developed and agreed to by IBM, Microsoft, EMC and Oracle—the four behemoths that dominate the IT software industry. When those four agree on something, things typically happen. The fact that other ECM vendors such as Open Text and Alfresco were also involved bodes well for its adoption.

Hence, I am cautiously optimistic about CMIS’ chances of success. If it turns out well, it will be a huge boon to organizations both large and small that operate siloed repositories—saving them the grief of deciding between one or another, or even worse, the torture of content migration exercises. For vendors, it allows a more harmonious coexistence with an implicit acknowledgement that buyers don’t typically throw one installed ECM system out in favor of another. Rather, they just add to the collection as the data mountains grow, and as the use of ECM across the enterprise expands.

Of course, as a professional cynic, I have to point you back to the miserable history of standards adoption the sector has witnessed. It will only take one of the big vendors to get cold feet or, worse, to figure they can "enhance" the standard, for it to fall apart rapidly. But at the moment things look rosy, and this could well herald in good times for ECM buyers globally.