As I discussed yesterday, I’ve been waiting a long time for the Content Management Interoperability Services (CMIS) standard. There is still a fair amount of excitement out there as more people join the conversation. I’m still excited, but the excitement is beginning to be tempered by reality.
There are two primary factors to standard adoption:
- Is the standard technically sound? It has to actually solve the stated problem. It is okay if a standard is limited in functionality in initial drafts as long as it evolves to accomplish everything required. At the same time, it must be easy enough to use. These are not small technical challenges.
- Is there vendor support? Let’s face it, if the vendors don’t support it, then it will fail. The JSR-170 and JSR-283 standards are perfect examples. They aren’t supported by a critical mass of vendors. The reasons range from the technical (we work in Java), to the philosophical (it is a bad standard, let’s focus elsewhere), and to the lazy (nobody cares so let’s ignore it).
Customers are important, but it takes a large mass of them to force the vendors to act. I would qualify them as a secondary factor. While I digest the technical aspects, take a look at the Vendor Support factor.
How Many Vendors Can Change a Light Bulb?
If you ask that question at an AIIM seminar, the answer is simple, The one that built the lamp. That answer won’t work for the interoperable SOA world that we are trying to build. The answer needs to be like it is now with lamps, anyone with the correct wattage.
The problem with that answer is that it is too big to be a realistic starting point. If we need every vendor to support CMIS out of the gate in order to be effective, then we are doomed. Luckily for us, we don’t need that level of support. What we need is a critical mass of ECM platform vendors and Content Rich Application including CEVAs providers.
The number of platform vendors required depends more on the customer base of the supporting products than a raw corporation count. Microsoft provides more weight than Alfresco. Open Text gets full credit if they only have one of their products support the standard. While they may get killed in a sales competition, it is the PERCEPTION of adoption and support that is important.
This is where the application vendors step into play. Many develop an application with either their own back-end or built on-top of an existing application. Sometimes they write multiple versions to support multiple ECM platforms. If they perceive support for CMIS among the platform providers, the application leaders will invest the time and money to change their application to connect through a CMIS interface.
This will reward the ECM platform providers by enabling the sales of the Content Rich Applications and CEVAs to drive sales of their platform. The lagging platform vendors will see the money being lost and implement support for the standard, sell-out quick, or fade away.
The leaders for successful adoption of CMIS need to be the ECM Platform vendors.
Will the Leaders Please Step Forward
I figure we need a 75% seat rate, existing and growth, for a locked-in success of CMIS. 50% would be a challenge. There are scenarios that throw those numbers off, but I’ll get to those in a minute.
If we look at the list for CMIS, we are already there. Unfortunately, it isn’t that easy. The JSR standards had that level of “support”. We need to measure support by actual released product and public, ongoing, commitment to CMIS in the Vision of each company. We need to judge the vendors by their actions over the next six months. Will they continue to talk about CMIS and will they implement a supported interface layer? That is the criteria, not press releases and web sites.
The one thing that could speed things up is if some open source vendors implement the standard and the user community responds. The ECM platform vendors will then scramble to draw even. Alfresco has started down that path, but more on that in another post.
Next, I’ll look at the depth of initial vendor support and see if we can guess who the leaders might be for CMIS down the road.