Tue, 18 Sep 2007 11:36:48 +0000
There has been lots of chat on the OMG BMI [*1] mailing list recently on the future of BPMN [*2] (the popular process orchestration notation), BPDM (the new and currently rarely-seen “process” metamodel), and their relationship to SOA. Meanwhile, here in Florida, the EPTS is meeting today while the Gartner BPM Summit takes place, and later in the week we swap over and the Gartner Event Processing Summit takes place. And next month, another variant on a process, rules, gets an outing at BRForum. And the question arose: as these are just different types of process (and potentially covered under a single process metamodel), would it make sense to combine all 3 of these events into a single Process Automation and Management Summit [*3]? Business Process Management is about documenting and managing (i.e. controlling) your business processes. Typically (and pretty much universally) this translates to a process orchestration editor and long-running process engine. In tools like the highly respected TIBCO iProcess, you get things like user interface forms, process simulation, hooks into services such as those provided by TIBCO BusinessWorks, and components like BAM and Business Rules (the latter via iProcess Decisions, a decision table tool). And a standard, popular, “process definition language” (XPDL). All good stuff. But of course, there is more than 1 way of defining a process, and while BPMN-based tools are (rightly) popular, there are always some processes (and services) which make more sense to define in some other way. Like a bunch of production rules representing some underwriting or pricing business rules. Or an event-driven process capturing and processing customer interaction events. And of course you can “manage” business rules and event processes just like regular “business processes”…
Possibly Gartner (and other BPM show organisers) should consider including multiple process types at a single event: cover business process from all orchestration/workflow, event-processing and rule-driven perspectives under one roof.
[*1] The Business and Modeling Integration “domain task force” (that always makes me smile!) covers the business-IT interaction mechanisms in OMG: areas like BPM, business rules, and (difficult to find) the Business / Computation Independent Model (”CIM”) stack of OMG MDA. It is also the new home of BPMI.
[*2] The BPMN 2.0 RFP has recently been published, and it seems BPMN is to be extended to include some of the other process organization types (like choreography [*4]) currently defined via BPDM. Controversially, BPMN 2.0 is proposed to be renamed to Business Process Model and Notation, and meant to include a persistence mechanism (good news, except there is already a well-adopted persistence mechanism in XPDL, and bad news if this (a) bloats BPMN, (b) causes the BPMN standard to be forked, and (c) dilutes the concept of BPDM by tying it to just processes that BPMN covers). This is causing lots of discussion, naturally, which hopefully will translate to more assistance/guidance being given to the OMG team(s) working on these specs/proposals.
[*3] Ignoring the fact that BRForum is not a Gartner event, and that Gartner have no Business Rules Summit at the time of writing…
[*4] Notes on notes? Just a thought: is choreography just a particular pattern of event processing that is conducting services?
I would like to create an application with communication between processes,
application tightly coupled, have you please an idea about an API or
a tool that allows me to generate such application?
Thank you so much (11 Replies)
I need a help.
i have an mapped drive application. I want to start this application using cron. I have downloaded cron setup from the below link:
And it crontab file works fine.
But if i want to start an application. Then process... (2 Replies)
I'm trying to use application-managed stacks, i.e. allocating the stack myself and passing it to the pthread creation attributes through pthread_setstack -function. My problem is that the memory allocated for the stack seems to leak after the thread terminates.
Only information I have been able... (2 Replies)