I’m off now to do a keynote for Unisys on “The Future of Content and Process Management” at their conference center at St-Paul-de-Vence, outside of Nice. They run their own 5-star hotel in the grand French style for their best customers. I’ve been there before, and this is really the best gig going for an industry analyst/consultant. After that a couple weeks’ vacation in Provence and Sicily. I may not be blogging much until May.
Add comment April 21st, 2006
One cool thing I saw at Brainstorm BPM was a demo by Cordys of their BPMN-based process designer. I hadn’t heard of Cordys, which is based in Amsterdam , but they sent me the latest Gartner MQ of the “ISE” market (Gartner’s term for SOA management/orchestration platforms — why do they do this?) where Cordys came out highest in the “completeness of vision” axis.
Anyway, they have a really nice BPMN designer, supporting intermediate events and other “hard” parts of BPMN. You make the shapes and lines executable by dragging services and data mappings onto them, and it generates executable BPML (not BPEL today) under the covers. They also have business rules, workflow, BAM, lots of good stuff. Some of the Cordys people I met there said this is Composite Application Framework, not BPM, but the BPMN project leader said, no, this is BPM. I think so too. When they get their messaging straightened out, I think we’ll be hearing more about Cordys in the BPMS space.
4 comments April 21st, 2006
I’m back from the Brainstorm BPM and SOA Conference in Chicago this week, where I spoke on Selecting a BPMS to the BPM crowd and tried to explain BPEL to the SOA crowd. In an event like that my presentations stick out like a sore thumb, as the typical conference attendee is really trying to learn “how to do BPM,” which in that context means documenting the as-is process and modeling an better way to do it against the backdrop of a traditionally stovepiped organization. At that point they’re not usually thinking about BPMS, and that’s fine.
But the annoying thing about conferences like this, and I saw it a DCI too when I was allowed to go there, is a subtle bias on the part of confeence keynoters against BPMS or any other form of “vendor technology” used to automate and monitor process execution. They don’t come out and say “that stuff is useless, stay away!” but instead give the impression that BPM is simply cross-functional “process thinking” and modeling as a “requirements-gathering” technique. Definitely NOT BPM 2.0! I don’t understand why the BPMS vendors, who basically fund events like this through their mini-trade show tables and meal sponsorships, put up with it.
In any event, there was a definite uptick in energy this year, compared to the past two, and enthusiasm for the BPM message, and greater interest in BPMS. I was very encouraged.
Add comment April 21st, 2006
I am expanding my 2006 BPMS Report series, which is available for free from BPM Institute. Over 1500 copies have been downloaded to date. Each report contains a 20-25 page walkthrough of a leading BPMS, all using the same analytical framework and report outline, and products are rated as to their strengths in several distinct process types or use cases. The first round, published last November, covered Adobe, Fuego (now BEA), Global 360, IBM, Pegasystems, Savvion, and Vitria. In the next round beginning in May-June we’re hoping to add Appian, Lombardi, webMethods, updated BEA, and hopefully more. Vendors, if you’d like to be included, please contact me.
1 comment April 17th, 2006
Sandy Kemsley wrote a note recently commending the upcoming BPM Think Tank in DC May 23-25, and I want to second that emotion. It’s put on by OMG, who absorbed BPMI.org in 2005. If last year’s version is any indication, this is the one event where those who really “get” BPM can mingle and argue and generally move the ball forward. Probably as close as we get to “BPM camp.” Early bird discount is available until May 1.
1 comment April 13th, 2006
James Taylor’s blog on ebizq points to another piece on that site which asserts not only, as James paraphrases, that BPM and SOA are no “silver bullet” for the business-IT alignment problem, but that they are at their core no different from all previous attempts to bridge the business-IT gap. Zygmunt Jackowski, PhD, who describes himself as a BPM Specialist with the Australian Department of Immigration and Multicultural Affairs, poses the question this way:
1. Apart from all the hype, is the BPM/SOA approach at its core different from what have been tried before to bridge the business-IT gap?
2. Is there any breakthrough feature (technological, methodological or perhaps architectural) that poses BPM/SOA approach as a likely winner over the previous attempts?
He goes on to say:
My answer to both of the questions above is ‘No’ and I will address them in later commentaries. This means that not denying all the advantages that BPM/SOA approach brings with it (a list of which one can find on BPM/SOA Websites of their choice) we are still not any closer to resolving the issue of bridging the business-IT divide.
The title of the piece is “Bridging the IT-Business Gap With BPM and SOA (Part I).” Oy.
Add comment April 13th, 2006
I received an interesting email yesterday re my BPM 2.0 manifesto from a professed “process analyst” I know:
Another good one, Bruce… I’d quibble only about the role defs for “business analyst” (a common misnomer in vogue in IT today that should be titled “requirements analyst” since they don’t really analyze the business or assist the business in developing strategies, workplace design, etc.) and “process analyst” (being a technically oriented position - when there are already a bunch of process analysts out here and our primary role is doing all that stuff that the business analyst doesn’t do… more involved with business consulting/planning/operations design, etc.).
And, lest you shoot back something along the lines of “well, that’s just your opinion”, I submit to you that the IIBA (International Institute of Business Analysts) defines their role pretty much exactly as we do here at my workplace and everywhere else in North America that I have visited or otherwise know about. And the BPMG study done last year combined with ABPMP’s membership both indicate that the role using the title “process analyst” is very much as I have said, more of an internal business consultant oriented to process measurement, management and change.
I can see where an outside analyst or a software vendor who doesn’t have to actually do this kind of work within a large organization might get confused and think that we’d use the titles rationally to mean pretty much what they intuitively would mean to a casual observer, but much the same way many other roles have developed (manager, leader, architect for example… pulleeeeze) they aren’t really what they sound like.
This notion of “process analyst” is consistent with Ismael’s formulation, which I agree is the key role in moving BPM 2.0 forward. But lest you think this resolves traditional issues of business-IT collaboration and alignment, my friend elaborated in a follow-up note:
What I meant was most “IT managers” don’t actually “manage” they caretake. Most “IT leaders” don’t lead, they exert power. Most “IT architects” don’t design and engineer anything, they set “standards” (usually by copying something else - like vendor handouts) and act as technology cops. IT has a way of using terms, especially roles and titles in ways that imply one thing but actually do another. So “business analysts” don’t really analyze the business, they document requirements for support systems.
2 comments April 11th, 2006
Last week Lombardi Software announced that Cognos, a leader in business intelligence software, had OEM’ed Lombardi TeamWorks for use in a new line of “analytic applications.” This action adds a new twist to the already-blurring boundaries between BPM and performance management/analytics that has been going on for a year or two.
Performance management is one of the 3 legs of the BPMS stool (the others being analytical modeling and process execution), so it’s no accident that BPMS vendors have been poaching on BI turf with features like process analytics, management dashboards, and BAM. The BAM situation has been particularly nettlesome, since it requires both real-time event-triggered KPI update and rule evaluation, and a platform for executing rule-triggered actions. Today’s BPMSs are actually more likely to have this than BI products, the assumed logical “home” of BAM technology. This is the problem that the Cognos-Lombardi deal addresses (at least in part).
On the other hand, for a long time BPM vendors and BI vendors have happily partnered for “deep analytics,” in which BPMS data is passed to BI where it can be merged with other enterprise data streams, mined and analyzed for subtle trends. Cognos, for example, is anxious to point out that the Lombardi agreement in no way affects its partnership program with BPM vendors (including FileNet, Savvion, IBM, EMC Documentum, and Pegasystems, plus Lombardi) for the Cognos 8 platform. The purpose of that program is to jointly develop solutions for:
To date, according to Cognos, “we have mapped a framework manager model of all of the BPM vendors in our program’s process repositories. We have also done much deeper integration with FileNet for number 2 and 3.” But let’s face it, these are basically co-marketing programs, not truly deep integration.
The Lombardi deal is different, representing BPM embedded in Cognos’s analytic applications, which are specific industry solutions layered on top of the platform. Although Cognos has a few such apps today, I believe new ones will be developed to take advantage of the embedded process capability. There the process model will be essentially prebuilt in the analytical app; the process design tool will not be exposed to users. (For the users who feel constrained by this, there is undoubtedly the full BPM upsell opportunity.)
This could turn out to be a big deal for the BPM market. Once BI vendors understand what BPMS can do (I don’t think they get it yet), we should look for more deals along these lines.
Add comment April 10th, 2006
The current issue of Business Integration Journal has an interesting piece from Oracle about my favorite topic, how to keep process models (e.g. BPMN) and their BPMS implementations (e.g. BPEL) in sync, what we call the round-tripping problem. I’ve repeatedly expressed my view that if BPM 2.0 is going to deliver real benefit over what we have today, this capability is essential, but others believe just as strongly that - especially when BPEL is the implementation technology - round-tripping is a mirage, fool’s errand, or worse.
Oracle’s solution, which I don’t believe is available in the current version of Oracle BPEL Process Manager, is what the authors call a “business flow outline” with additional “metadata” that can be populated by a BPMN model using “well-defined guidelines” and fleshed out in a real BPEL design tool like Oracle BPEL Designer. Hey, “outline” — isn’t that what Edwin K was talking about a while back? I thought that was his term for modeling, but apparently it’s more in the nature of a skeleton process design created automatically from BPMN.
Unfortunately the screenshots aren’t in the bijonline version, but the print version makes it look like more than an artist’s conception. (I don’t believe this is in the current version of the product.) The outline, representing a “logical view” of the process, appears to run in Oracle BPEL Designer using a BPMN-ish notation called the Process Analysis palette. A developer can then map those shapes to BPEL activities that represent the actual implementation. To ensure the round-trip, the BPEL shapes must have guaranteed bi-directional mappings to the BPMN-ish shapes in the outline.
It’s not apparent whether Oracle plans to offer the outline as a modeling tool for business analysts or simply a way to capture BPMN models created in third party tools. If you’ve been following this thread on BPMS Watch, you’ll remember that BPMN lets you draw things that don’t map quite so easily to BPEL - or at least the kind of BPEL you’d want to edit and maintain, but there are subsets of BPMN diagrams that can be mapped automatically. If you control the BPMN tool, you can solve the problem by not letting the user draw something that can’t map easily to BPEL.
In the BIJ screenshots, the outline’s BPMN-ish shapes happen to correspond one-to-one with BPEL activities. If that’s how the thing works, it may be just a gimmick, since you’d expect a many-to-one ration of BPEL activities to BPMN shapes in a real process. But I suspect that’s just an artifact of the screenshot. Hoping to find out more…
2 comments April 10th, 2006
SearchOracle.com discusses a recent Forrester report that claims SAP NetWeaver is winning the middleware battle against Oracle Fusion. The logic isn’t fully revealed in the story, but the key seems to be more about the strength of those companies’ enterprise applications and solution partners than the technical merits of the middleware itself. Forrester’s Ray Wang, co-author of the report, explains SAP’s advantage this way:
“A head start, a partner ecosystem and a customer base are pretty much key ingredients for success.”
In addition to the core apps themselves, BPM is one of three fronts where SAP and Oracle will be duking it out over middleware, the others being business intelligence and master data management. But the report also says, as reported by SearchOracle, that
Oracle’s strong middleware platform and better support of open standards make it the right choice for customers who rely heavily on custom development in conjunction with packaged applications…
Hey, isn’t that what BPM is all about? Go figure.
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.
Add comment April 7th, 2006
| M | T | W | T | F | S | S |
|---|---|---|---|---|---|---|
| « Mar | May » | |||||
| 1 | 2 | |||||
| 3 | 4 | 5 | 6 | 7 | 8 | 9 |
| 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| 17 | 18 | 19 | 20 | 21 | 22 | 23 |
| 24 | 25 | 26 | 27 | 28 | 29 | 30 |
