Method and Style Validation Update

A few updates on my Method and Style validation tools...

  1. The one that works on the BPMN 2.0 export from itp commerce requires Service Release 5 of their tool. My apologies to those who tried to get it to work with SR4. The good news is that SR5 beta is available from the training area of The GA SR5 should be out soon.
  2. The one that works from Visio 2010 continues to be an interesting ride. My first beta user was from Norway, and naturally I had code that looked for the English text values for things like "Task" and "Event" instead of the language-independent way. This led me to find out how Visio really works, and it's amazing I could get my tool going at all without knowing that. With a little help from David Parkier's book Microsoft Visio 2010 - Business Process Diagramming and Validation - yes, someone has actually written that! - I not only was able to figure it out, but I see a path to some other things I can do with Visio directly. (David's book is mostly about how Visio's built-in Check Diagram function works, and how to extend it with your own rules... pretty cool.) Anyway, today I posted a new version of the tool that uses the index to Visio's enumerated property values rather than their names, and it works in Norwegian as well as English... at least in the one diagram that I have. There are probably more bugs and I'm anxious for people to try it. So if you have Visio Premium (or can get the free trial), let me know that you are interested and I will provide access.
  3. I got a note from Steve White a few days ago about my list of BPMN rules. He took issue with a few of them. BPMN 0101 and 0102 reflect the long-time rule that implicit start and end nodes (i.e., no incoming and outgoing sequence flows, respectively) are not allowed if the process level contains any actual start or end events. Apparently OMG caved at the end of FTF on this and the rule has gone away. Too bad, as it leads to diagram errors -- see Connie's comment here - so I am going to keep those rules but change to "style rules".Also, I had made a comment - not in the validation tool - that the inline expanded subprocess could not contain lanes. That's another longtime BPMN rule that got bounced at the end of FTF. That's fine... BPMN 2.0 doesn't really know what to do with lanes anyway.And re my rule BPMN 01122, which lists the allowed triggers on boundary events, he suggests adding Multiple and Parallel Multiple to the list. Mmmm,.... It's mostly a wording issue. Those event types have their own icons but they don't really exist in the XML; they are simply implied by the fact of more than one eventDefinition element. Since the validation just looks at the XML, it doesn't change the implementation of the rule. Maybe I'll change the wording. Anyway, I'm glad people like Steve are keeping me honest!