In his post a couple of days ago, Bruce Silver despaired at the state of BPDM - the OMG specification that allows for real portability of BPMN. I agree.

You know desperate men,
they'll do desperate things

- Jimmy LaFave

However, standards take time... indeed if you don't take that time then you end up with something that won't be used. And given the importance of business process, and the new process platform that will replace ERP as the center of the business world, I think a few months is not too high a price to pay to get it right.

BPDM stands for Business Process Definition Metamodel. It's a hifalutin way to say "the file format," essentially. One key reason Microsoft Office has become so dominant is that its files are transportable across versions of Office (for the last few years) and across platforms (Mac and Windows). Without that portability, the desktop productivity tools would be a bit more fragmented around the world.

An implementation-independent process semantic is the lynchpin for this new platform. It will provide this common file format for the visualization of business processes (the "BPMN" standard, used in tools from 60+ vendors). It will provide the implementation-independent format that execution-oriented vendors who use BPEL or XPDL can use.

Without an implementation-independent standard, the business process world will splinter into dozens, or hundreds, of proprietary silos. Remember, before Java, what the development world was like? BPDM is for process what Java is for development.

There is real business value here because with this capability, business users can use tools they can use, and then transfer the processes to the implementation folks who can use the same file (essentially) for their work. It's exactly like being able to draft a high level PowerPoint presentation, hand it off to your graphics artist for cleaning up those drawings, and then getting back the revised PowerPoint and being able to use it.

Different tools were used by different people, based on their skills, but it was all able to be transformed from and into a common format - up and down the line.

Those who oppose BPDM are taking a short view, and most likely a proprietary view so that you're locked in to their world or because they don't have the money to invest in standards for their product.

A few months ago, the first semi-public version of the spec was published (it's available to any OMG member), and it has been updated 2 or 3 times since then. In December, after dozens, if not hundreds, of person-years work, the committee working on BPDM is expected to formally submit it for adoption. If everything goes well, the OMG will adopt BPDM and it will be available to the general public in January. There is a chance all this is shifted to March, but for now the BPDM committee tells me they will be ready to make the submission.

But regardless of whether it's December or March, the day is soon approaching when the new business process platform has its equivalent of a common file format. This will unleash innovation among vendors unseen in this space, because then, everyone's tools can work with everyone else.

So keep up the pressure, Bruce... we need to get this wrapped up. But also don't sweat a couple of months and start telling your clients to go to a silo'd implementation language like XPDL... because the dawn of implementation-independent portability is fast approaching, even if it's pretty dark out there right now.

Technorati Tags: BPDM, BPMN, Bruce Silver