I was reading a post by Janus Boye with the provocative premise that customers shouldn’t worry about CMIS. As you can imagine, I was shocked.  When I read the post, I saw that he had some valid points, but that his conclusion was only about half-right. As I started writing my response, I realized I was writing something entirely too long, so I brought it here.

The Argument

Janus states four reasons why WCM purchasers don’t need to worry about CMIS.  Here they are, complete with the embedded link:

• CMIS is still only a proposed standard and will take time to move through the process of becoming an approved standard
• CMIS is not implemented yet by any of the vendors in anything but experimental releases of sample code. Early adopters tend to be disadvantaged in this industry as they usually get the lion’s share of bugs and cost.
• CMIS is a standard for enterprise content management and not web content management
• You will probably change your CMS again in 3 years and by then CMIS may well be a more mature proposition

All true to various degrees.  I agree that CMIS compliance should not a determining factor in any purchase right now.  There are too many things to consider without making a non-final, incompletely and inconsistently implemented standard a requirement.

CMIS is probably a year from being finalized, assuming the current momentum. That is a lot of time for changes to be introduced and for vendors to build a production-ready implementation. Even if a vendor has a production-ready implementation, until the spec is finalized, it will remain a beta/trail release.

That being said…Every vendor should have a plan for support.  Are they going to be a consumer?  Are they going to be a provider?  Will they be both? If you are making a strategic purchase, then any vendor that doesn’t have a set of answers to how and when they will use CMIS may not have the vision to keep their product viable for the 3+ years you plan to use it.

As for those current test implementations? They aren’t required, but they do show that the answers that are being given may be more than words.

This is less of an issue with pure WCM vendors, but if you are looking in the broader ECM world, CMIS needs to be part of the discussion. Just be talking to them about it helps it get back to product management that CMIS matters.

The Argument for WCM and CMIS

WCM vendors will need to hop on board.  If they can use CMIS, they gain many valuable advantages over their competitors.  In the long run, they will need it to survive in the larger world.

• WCM as Federated Consumers: Web content comes from the entire organization.  CMIS is an easy way for a WCM system to consume it from SharePoint, Alfresco, Documentum, or some other system without having to write a custom connector.  Now the WCM system can manage all of that web content while the supporting marketing information can be created where the domain experts spend their day.
• WCM as a Scalable System: One problem that some WCM vendors face is scale.  There is just a lot of content. If they offload that to a ECM provider, then they can focus on what makes them so valuable in the WCM marketplace, building and managing web sites. WCM and ECM are different mindsets. Just imagine how much better some of the WCM systems would be if the low-level management of the content was offloaded to a larger ECM system. This isn’t something I expect to see for a while, but when it does arrive, I expect to see a new market leader emerge.

If a WCM vendor tries to tell you that CMIS doesn’t matter, then they don’t get it and they may not be the right partner to help you reach your long-term goals.  They don’t need an implementation, but they should be able to tell you how CMIS is impacting their product plans and how you, the user, can expect your life to be easier once they do support it.