We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Anne Stuart’s BPM in Action

Dennis Byron

BPM Viewpoint: The 2008 BPM Standards Debate--Looking Back

user-pic
Vote 0 Votes

Let me start by airing a few of my biases about standards... any kind of standards, not just business process management (BPM) standards and not even just information technology (IT) standards.

  • First, defacto standards established by market forces are better than dejure standards; dejure standards are OK in principle but it typically takes years for everyone to agree on them (see BPM examples below).
  • Two, standards are particularly irrelevant in market research--which is what I do most of the time--as contrasted with product research; that's because standards do not act as market drivers since every supplier typically meets any required standards sooner or later (and if you are a large supplier to a particular IT market you can be last to support a standard as long as you promise to meet it some day).
  • Third and most important, standards are for vendors, not users; standards save IT suppliers development expense and simplify qualifying for certain procurement processes. Interoperability is important for users but standards are not required to make it happen.

My biases come from involvement in the 88open, the beginnings of the Object Management Group (OMG), and various users groups. If you are a user with good experiences with standards and/or standards organizations, please comment or send me an email if you feel the need to counteract my negativity. But basically my advice to users is to key on the functionality no matter what the standards status of a particular piece of software and--except in the rarest circumstances--only acquire software that is going to return your investment in 12 months no matter how many standards it meets.

I mention the above in order to set the stage for this two-part BPM Viewpoint on the BPM Standards Debate of 2008 because four years ago, in November 2004, while at IDC, I wrote (with some grammatical corrections to make it readable today and a few interspersed notes):

"Near term, I believe vanilla business process orchestration will be done with web services and therefore the Business Process Execution Language (BPEL) makes sense. (Note: officially, BPEL is an OASIS-approved dejure standard, WS-BPEL 2.0, approved in April 2007, about five years after Microsoft and IBM created it. The WS stands for Web Services.)
"However longer term, performance needs and attempts to develop market differentiation will lead to outright competitors to or non-standard implementations of BPEL. Given the support of IBM, Microsoft, Oracle (Collaxa) and SAP (Note: the four largest enterprise software suppliers by a wide margin), BPEL compliance by BPM software suppliers will most likely be a must have for RFP checklist reasons even if competitors or non-standard implementations become more popular.
"The open-source BPEL engines now emerging in the market and others on the near horizon will make it easy for BPM vendors to include a BPEL engine as an RFP check off item even if they do not want to devote a lot of development dollars to BPEL and/or do not believe in BPEL's long-term success.
"...until June 2004, there was another competing business-process language standard called the Business Process Markup Language (BPML), promoted by an industry consortium called the Business Process Management Initiative (BPMI). BPML had more of an ecommerce-workflow-specific view of the world than the industry movers and shakers behind BPEL seemed to want turned into a software commodity, which is what standards do to technologies. With its definition of enterprise business processes, the definition of multi-party collaborations, as well as the definition of how complex Web services interact in BPM, BPML might have been a better choice long term (Note: I believe the big win for BPM is in inter-legal-entity workflows).
"The battle between BPEL and BPML ended in June 2004 with BPMI officially endorsing BPEL (Note: Basically they withdrew BPML because of the challenge of competing with a competing standard endorsed by four companies that account for half of the software market). BPMI's next project was promoting the Business Process Modeling Notation (BPMN) standard for BPM modeling tools that had been included with BPML. The group subsequently hoped to provide a semantic-model standard and a query language standard, to operate "on top of" and "beside" BPEL respectively in a stack metaphor.
"These efforts of course could put BPMI/BPMN back in competition with OASIS. But it appears as if BPMI wants to become more like The Open Group, in terms of promoting standards adherence, rather than being a standards group itself (Note: BPMI subsequently became part of the OMG, not really the best thing it could have done if it wanted to be like The Open Group)."

Back in 2004 I was happy to get the issue of BPM standards off my plate. All the major suppliers--my clients at IDC--agreed on one standard, which in my opinion, didn't matter anyways from a market research perspective.

I also had the strong impression based on the questions that my clients were asking me, that they didn't care much about BPM standards either. I don't recall the question coming up once in 2005. I took a sabbatical in 2006. I concentrated on market research in another technology area in 2007 and only returned to closely researching the BPM market in 2008.

So I have to say, I was surprised to find a debate raging over BPMN and BPEL. It was like I had never left. The current iteration of the debate dated at least from September 2008 (if not earlier--see Lombardi's Phil Gilbert in 2006). As I pointed out in ebizQ's "More from the Web" section over the last two months, two bloggers in particular seemed to have joined the fray: Bruce Silver and Ismael Ghalimi (also see our unrelated blog post on Ismael's company, Intalio, on December 4). But it has raged elsewhere in the BPM blogosphere as well (see posts by José D. De la Cruz, David French, Michael Rowley, and Pierre Vigneras).

But when I read all of the above mentioned posts, I decided I was missing something. It didn't appear that anything had changed since 2004. So why the debate? When I asked on the BPM in Action blog on December 4 "What Do You Think about BPMN and BPEL?", one of you commented it's the difference between "UML vs. Java." Another reader emailed me to say, it "is sort of like asking engine vs. transmission - they are complementary and both have a place in the lifecycle from process design to execution (respectively)."  That was my opinion as well.

So part II of this Viewpoint will dig down deeper, looking forward where the above looks back.


  • Are those of us on the fringe of the debate missing something important?

  • How should this affect your planning for upcoming BPM projects?

  • How come BPEL compliance is still an issue? Or is it BPMN compliance that is the issue?

  • Do they not co-exist?

  • Did the open-source alternatives emerge?

  • And otherwise how accurate was my 2004 prediction that there would be some BPEL extensions or outright competitors?

---- Dennis Byron

Update: This post originally said "two good bloggers" referring to Bruce Silver and Ismael Ghalimi. I apologize if anyone thought I meant that the other bloggers mentioned were not "good bloggers." What I meant to say was "two bloggers in particular" (that is, good in the sense that they had written about the subject frequently)

Leave a comment

Business process management and optimization -- philosophies, policies, practices, and punditry.

Anne Stuart

I am the editor of ebizQ.

Recently Commented On

Monthly Archives

Blogs

ADVERTISEMENT