[Originally posted on IT|Redux]
Edwin Khodabakchian, the brains behind Oracle BPEL Process Manager, posts on a set of possible ?BPEL enhancements? suggested to him by Oracle?s application groups ? eBusiness Suite, PeopleSoft, Siebel? They include:
1. Business Process Outline ? ?Enable business analysts to build the skeleton of a business process, skeleton which can be then implemented by an application composer. The outline view also offers a foundation for self-documenting business process and audit trails.? Imagine that!
2. Business Events ? Allow the process to raise events processed by external apps and to subscribe to events raised by apps, providing ?the glue between business activities and the foundation for realtime analytics.? He proposes a new event language standard, not mentioning the fact that IBM has proposed one already (apparently without unanimous support) and uses it in its own BPM/BAM integration.
3. Variable Entity Binding ? When BPEL is used not in an integration context but within an enterprise app, persist BPEL variables automatically in that application.
To my mind, those innovations sound suspiciously like what BPM calls process modeling, BAM, and embedded application workflow ? assumed ?must-haves? in the BPM world but exotic enhancements to the BPEL mainstream. But unlike IBM (and now BEA), Oracle is not trying to appeal to BPM buyers or industry analysts, just better support Oracle?s own apps. They?ll figure it out eventually.
[comment by Edwin K - from IT|Redux]
Bruce,
It is a little unfair to criticize us for taking a more incremental/layered approach to what you call BPM.
If you step back and look at the real world adoption of BPM, you will notice that although BPM solutions have been available for almost 10 years, the companies behind them have had a hard time breaking out of the $10-15M revenue band (and this is despite the fact that they are targeting the deep pocketed line of business user). Why?
The ?BPEL/BPML? BPM generation has been taking a different approach, one that I believe is richer, more open and more composable.
Richer ? BPEL supports more orchestration and interaction patterns than any of the proprietary workflow definition languages out there (parallel, parallel for each, pick, onEvent, alarms, links, compensation, fault handling, early join patterns, etc.). This is unusual because standards usually represent the lowest common denominator. In a world where interactions and business processes are becoming increasingly sophisticated, this is a must-have.
Open ? BPEL allows you to change the binding systems (WSIF initially and Service Component Architecture going forward), the expression language (XPATH out of the box, E4X or even a rule language when necessary), the type system (XML/XML Schema out of the box, but also more sophisticated entity binding when necessary), and the activity set (annotate activities or add new activities when necessary).
Composable ? BPEL is NOT a monolithic bag ( = NOT including initially organization context, forms/documents, auditing, notation, policies, security, analytics?). Other standards have emerged/will emerge to address those other aspects. The power of BPEL is that it lives in an ecosystem of standards.
Few people understand this last aspect, especially those ?BPEL haters? ? most often people that have a ?lipstick on a pig BPEL implementation? ? have been trying to create FUD around, with erroneous arguments such as a) BPEL only works with Web services, b) BPEL can not orchestrate human activities, or c) BPEL is a low level scripting language and it is incompatible with business notations.
The purpose of my blog is to mainly give insight into our product roadmap, while at the same time encouraging people to challenge some of this FUD. The outline view for example will address c) (note: I am not claiming that the notion of shared model is a discovery!!!)
The business event and entity variable binding aspects are more interesting. May be we can discuss this further when more information is available.
Best,
Edwin
[comment by bruce on IT|Redux]
Edwin,
Thanks for responding, and more than that, thanks for providing the first BPEL designer/runtime on the market and still more than that, offering it for free! Not being a BPEL-hater myself, I think your basic beef is not with me but with even smarter guys like Phil Gilbert. I?d love to have you two square off in a full-throttle slapdown on IT|Redux sometime!
No, the intent of my post, which you didn?t address, was really about how Oracle, like other BPEL providers from the infrastructure/middleware space, is suddenly ?discovering? the BPMS concept. And that?s a good thing! A complete BPMS, or what Ismael now calls BPM 2.0, emphasizes things like business-IT alignment through modeling and simulation directly linked to the executable design, and integrated BAM and process analytics.
If you ever go to a BPM Conference like Brainstorm or DCI, which is aimed at those who self-identify with an interest in ?business process management,? you?ll find that users care much more about these aspects than whether the underlying language is BPEL or not. Among the BPMS pureplays, who toil slavishly for their paltry $10-15 Million, the need to integrate modeling, human workflow, adapter frameworks, business rules, BAM and analytics, etc is taken for granted.
The big middleware vendors, on the other hand, appear to view BPEL first and foremost as about business integration, not BPM, which is viewed as a secondary special market. IBM was the first of them to provide a complete BPMS when they integrated WebSphere Modeler and Monitor with Process Server v6 last fall. BEA is apparently jumping in now with the Fuego acquisition. Your post says to me that Oracle is planning to offer similar functional enhancements, but rather than say that it?s to address the needs of BPM, you frame it in terms of working better with Oracle applications. That?s why I say, speaking as an advocate of the BPMS/BPM 2.0 concept, that you?ll ?get it eventually.?
[comment by Edwin K on IT|Redux]
Bruce,
My view is that BPM is a promise/a vision. A promise that users will take control of their business processes by having better visibility into them and by being able to continously adapt and extend them.
The key question is ?what do we need to deliver on this promess??.
Glancing through Phil?s blog, I actually very much agree on the fact that a) human interactions need to be a first class citizen of the solution and that b) a shared model is a must have.
Here are a few points where I have a different perspective:
1) I believe that business processes are only one aspects of business applications, pervasive user interfaces and pervasive business intelligence (historical, real-time and forcast) being the two other. Pure play BPMs tend to have a very simplistic view when it comes to user interface and monitoring.
2) I believe that business processes will become increasingly sophisticated and dynamic going forward and will therefore require a) much tighter integration with business events and complex event processors – which will provide a more intelligent glue between business activities and b) support more sophisticated orchestration patterns (see previous my previous post). Pure play BPMs tend to focus too much on simple directed flowcharts.
3) I believe that although the goal of BPM is to give control to line of business users, we can only deliver on the BPM vision if BPM is an integral part of the *enterprise* architecture. This is where the need for standards and the relationship with SOA, EDA (event driven architecture) and SCA (Service Component Architecture) is key: End-to-end security, interoperability, compensating business transactions, context, correlation, addressing, eventing, reliability – all these key aspects can only be abstracted out if BPM is part of an architecture. Pure play BPMs were born before the blueprint of this architecture was defined and most often do not have these elements in their DNA (Although their datasheet claim support for some of these, this is usually what I call the ?lipstick on the pig? approach).
Let?s connect back in May. By then, we will have a first iteration of a shared model combining the strength of BPEL, BPMN, Business Events, BAM KPIs and advertising organizational model and human activities as firt class citizens. Phil claims that it can not be done in what he calls the forseable future (less than 3 years). Let?s see if we can prove him wrong!
Please let me know next time you are in the bay area. I would love to meet face-to-face.
Best,
Edwin
[...] What struck me most was the implication that BPM at both companies is ultimately just an embellishment of their application story, just like workflow was in the 1990s. In that context, Edwin K’s re-framing of BPM concepts as Oracle application extenders makes a lot more sense. [...]
Don’t Be A Hater: A BPEL Agnostic Testifies…
There’s been an interesting back-and-forth between Edwin Khodabakchian of Oracle and Bruce Silver of Bruce Silver regarding BPEL, BPMN and the promise of business process management in general. To get to it, read Bruce’s post, Is Oracle Discovering …
[...] You gotta love the press release: “Oracle?s BPM product portfolio, which now includes IDS Scheer?s ARIS Platform, will support…” Before today’s announcement that it was OEMing parts of ARIS, Oracle to my knowledge did not offer a BPM product portfolio. Sure, they had an SOA Suite that included BPEL Process Manager and BPEL Designer – both excellent products – but the company never mustered the courage to say, “We understand what BPM is and we have a BPM suite.” A few months ago I tried to convince Edwin K and other BPEL honchos at Oracle to officially declare themselves in the BPMS battle, but they either felt they weren’t ready or had bigger fish to fry with SOA. The missing piece was on the analytical modeling side. Now that piece is no longer missing. And it looks like they’re in. [...]