Integration World Day 1: BPMS 7.1

Pete Carlson (Product Development) and Matt Green (Product Marketing) gave some of the high points of the new 7.1 release of webMethods BPMS. Their first question to the audience was to determine who is already involved in a BPM project in their organization: almost no hands went up. As it turns out, most people are here to learn about BPM and what webMethods has to offer. Given that the audience at the conference is mostly IT and mostly work with the webMethods ESB product, this isn’t all that surprising; BPM is a relatively new concept to most of them even if they’ve been involved in the integration-centric end of the BPM spectrum. I think that the biggest challenge for the Software AG webMethods group is, in fact, to gain greater visibility in the business areas of customer organizations, which is where many of the other BPMS vendors already have much more visibility.

A few points about the product:

  • Eclipse-based modelling tool with multiple perspectives for business analysts and developers to share a common model.
  • There’s a process debugger built into the modelling environment, which allows you to step through a process to see how the variables change as the process flows
  • Simulation is new for their BPMS, so they’re pretty excited about this and it’s a big focus at this show
  • BAM is also highlighted as a major part of their suite, and has separate product sessions here at the conference. They do some interesting things using prediction models that I saw briefly in a demo last night, and do automatic baseline metric creation based on running processes.
  • Business calendar, so that you can schedule something for 3 business days versus 3 calendar days (this is definitely behind the BPM curve as a new feature)
  • Integration of Cognos for business intelligence to be able to drill down into data that you might see in a dashboard environment; this will be fully embedded within the webMethods product suite in next year’s version
  • Integration of Blaze Advisor for business rules management, although the rules management user interface won’t be fully integrated into the webMethods environment until next year
  • A new UI design environment that allows for codeless generation of portals and form-type interfaces.

Carlson gave a live demo, showing how they can call services from their ESB, services outside their ESB, business rules and human steps all in the course of a process. Interestingly, one thing that he showed was an executing process instance, including the information that it was in the state "Queued" — does this imply that an instance can only have one state at one time, hence can’t execute parallel paths? Or does that say something about the small number of states that can exist?

He showed their simulation, which has the requisite animated dials to show where work is piling up in the system; although you can take a snapshot of the conditions and the results of any particular simulation scenario in order to manually derive improvements, there’s no tools to suggest ways to improve processes, such as I’ve seen in some other products. Instead, it’s a matter of tweaking the parameters and watching the effect on the simulation, or exporting the results to Excel to perform some other analysis on them.

He showed how individual tasks in a process map are edited using the new version, which is a view that allows you to modify the roles, data values, events KPIs and user interfaces at this point in the process. In particular, the event capabilities seemed pretty powerful, although I’m not sure if it supports the full BPMN set, and it’s certainly not represented in the process map as BPMN events.

The presentation finished up with a very quick look at the design time and runtime architectures, then (for the 3rd time today) a look at Forrester’s two waves that show them as the top vendor in integration-centric BPMS as of Q406, and in the leader category (barely) in human-centric BPMS as of Q307. They’re obviously pretty pleased with being a leader in both categories, although they’re pretty neck-in-neck with TIBCO, with BEA and IBM trailing a bit behind but still having a strong showing in both waves.

Integration World: Peter Schwartz

The futurist Peter Schwartz gave the final keynote this morning; I saw him giving a closing keynote at the Gartner BPM conference in February this year and really enjoyed it; unfortunately, this was pretty much the same talk. Long tail…blah, blah, blah…nanophotonics…blah, blah, blah…flattening world…blah, blah, blah…global warming…blah, blah, blah…ubiquitous broadband.

Conference organizers should definitely take a look at who’s been doing the generic keynotes at recent conferences that their attendees are very likely to have attended.

Integration World Day 1: Peter Kurpick

Peter Kurpick, CPO (Chief Product Officer) of webMethods Business Division, gave an overview of the technology direction. He talked about the paradigm for SOA governance, with the layers of technical services, business services and policies being consumed by business processes: the addition of the policy layer (which is the SOA governance part) sets this apart from many of the visions of SOA that you see.

He brought along Susan Ganeshan, the SVP of Product Management and Product Marketing, to give a (canned) demo similar to one that we saw yesterday at the end of the analyst sessions. She showed the process map as modelled in their BPM layer, where the appropriate services were called and other points of integration using webMethods, then we saw the custom portal-type interfaces for customers, suppliers and internal workers. They have Fair Isaac’s Blaze Advisor integrated with the BPMS that allows them to change rules for in-flight processes, and their own monitoring and analytics as well as some new Cognos analytics integration. She also showed us the CentraSite integration, where information about services and their policies are stored; CentraSite can be used to dynamically select from multiple equivalent services based on policies, such as selecting from one of several suppliers. The idea of the demo is to show how all of the pieces can come together — people, web services, B2B services, legacy services, and policy governance — all using the webMethods suite.

The original core functionality provided by webMethods is the ESB (originally from the EAI space), but now that’s surrounded by BPM, composite applications, B2B integration and legacy modernization tools (from the Software AG side). Around that is BAM, which is being raised in importance from being just an adjunct to BPM to being an event-related technology in its own right. Around all of this is SOA governance, which is what CentraSite brings to this.

The next release, due sometime in 2008, will be a fully-integrated suite of the Software AG and webMethods products, although Kurpick didn’t provide a lot of information.

Integration World Day 1: David Mitchell keynote

David Mitchell, COO of webMethods Business Division, was up next with a much greater focus on BPM. webMethods had been positioning BPM — including both system-to-system and human-facing — as their growth area before the acquisition, and that’s still the message. Companies are competing on process, not just products and services, since process represents the ability to meet customer requirements in the face of changing conditions. As I’ve written many times before, the value is in process agility and visibility in a heterogeneous environment.

He had a number of case studies of how customers have improved their processes and operations based on what they’ve done with webMethods to integrate their systems for the goals of agility and visibility. He said "If you have unlimited time and unlimited money, you can pay IBM to do this for you", followed by the clear message that the rest of us should be using webMethods instead.

He also discussed the impact of the merger: long-time Software AG customers engaging with the webMethods product line since they trust Software AG to help them with their modernization efforts. There were some interesting comments about mergers in this marketplace, and the issue of aligning the products and customers of the acquired and the acquirer. As Mitchell put it, that question has now been answered for webMethods, but is still up in the air for companies such as BEA and TIBCO.

Integration World Day 1: Karl-Heinz Steibich keynote

Well, not really day 1: there were classes yesterday, but this is the first day of the Integration World conference per se. Someone in webMethods management must have a serious love for electric guitars, since the prize at last night’s reception was a Gibson Les Paul guitar, and a loud (for 8:30am) rock band was on stage to open the conference. My ears are still ringing.

Mark Jeffries of webMethods led off this marathon 2-hour keynote session, then Karl-Heinz Streibich, CEO of Software AG, was up to talk about the combined corporate structure and their general future plans. He showed the evolution of the software industry, from build to buy to compose, and how Software AG sits directly on that infrastructure space focused on composite applications. It’s not like companies are throwing away all the applications and infrastructure that were built (or bought), but the growth is in the composition area. This growth, plus the webMethods acquisition, will push Software AG into the $1B revenue range next year (although those are, unfortunately, rapidly declining $US 🙂 ).

I attended the BPM Advisory Council breakfast prior to this, and had an interesting discussion with a couple of large (and long-time) webMethods customers about how webMethods stacks up in the BPM space. Clearly (to me, anyway), they’re behind the curve in terms of the pure BPM technology, but they play to their strengths in the integration end of things. The customers pooh-poohed the BPM vendors that don’t provide the whole integration stack as "just calling individual services", which is a bit of a simplistic view, but considering that these customers are coming from the traditional EAI-type usage of webMethods, it’s understandable. When you’re modernizing your legacy systems, there needs to be something in there to allow you to easily plug into those old systems using a web services interface, and that’s provided by webMethods. The Software AG acquisition will allow them to strengthen those linkages for Adabas/Natural applications, which are technologies still well-represented within large organizations.

Software AG analyst briefing at Integration World

Thanks to Air Canada’s ever-shifting schedule (why does the same flight number that I flew to Orlando two weeks ago now leave an hour later?), I arrived too late to the analyst briefing to hear the Peter Kurpick and David Mitchell of the webMethods division, but did sit in on the latter half to hear about the enterprise transaction systems division, which includes their Adabas and Natural products. There’s a press release in the packet about a new Software AG release due for tomorrow: Natural for Ajax, reminding me that there’s a ton of companies out there with Adabas and Natural applications on their mainframes that are still supported by Software AG, and need to find some alternatives to keep the life in those systems. I have a few financial services customers that have mission-critical Adabas/Natural applications, and they have no intention of rewriting them for some other software platform.

Simply replacing the mainframe with some other technology platform isn’t always the best solution, and is a total non-starter in many situations where there’s a significant amount of legacy business logic and data in those systems. I’ve been referring to mainframes as "just big servers" for quite some time now, and there’s no reason not to think of them that way: big database and back-end application servers that serve a useful purpose within large organizations. You might not start building a new system from scratch on a mainframe, but there’s often good reasons to keep them around if they’re already in place. Software AG, by providing tools such as Natural for Ajax, is providing some of the tools that organizations need to make the transition into more modern technology, without requiring a full rip-and-replace. I’m not sure that I completely agree with their vision of a mainframe Renaissance, however.

We had 20 minutes of corporate financial review from the CFO, which reminded me firmly why I went into engineering instead of accounting; I’m sure that there’s many capable financial analysts who have written about the numbers so I can slack off here. In looking at the overall financials, I’m reminded of the presentation that I saw at the New Software Industry conference earlier this year about how the proportion of product to services shifts as a software company ages: if you ignore software maintenance revenue — which is about 2/3 the size of the licence revenue — then the product-to-service ratio is about 1:1, and if you include maintenance in services, then the ratio goes to about 3:5. The theory is that as software companies age, they get better at providing services, and therefore it’s more profitable, hence tend to encourage the growth of that part of their business. This is completely off on a tangent, but I’m trying to keep my mind on the presentation as the CFO discusses the "Brazilian effect", and I have no idea what he means (since I assume that he’s not talking about the type of Brazilian effect that you get at a salon).

I had a chance to review the materials from the presentations that I missed, and there’s some interesting things there. This is a condensed version of what will be presented to customers here at Integration World, and I have a one-on-one meeting with Peter Kurpick on Wednesday so will undoubtedly see this again over the next two days. And hopefully have some more insightful comments at that time.

Integration World next week

This week I’m not travelling — a chance to catch up on the work that I’m actually paid for 🙂 — but next week, I’m off to the Software AG/webMethods user conference, Integration World. Watch for my blogging coverage from there.

Disclosure: as with all vendor user conferences that I attend, Software AG is covering my expenses, although I am not paid to attend.

Meeting the bloggers at BRF

Last week at the Business Rules Forum gave me a chance to meet many people who I’ve never met face-to-face, but feel that I know from our exchanges of blog comments and emails: at one point, I was standing around talking to James Taylor, Rolando Hernandez and Scott Sehlhorst.

James was certainly the most prolific in blogging about the conference: he live-blogged the sessions that he attended (even mine), so you can compare with the posts on those sessions that I wrote. He has a wrap-up post with pointers to all of the blogs that he found with coverage of the event.

BRF Day 3: Good Business Rules in Process — Eliminate 65% of the Activities

I couldn’t quite drag myself out of bed for the 8am sessions, but I did want to hear Kathy Long of the Process Renewal Group talking about process and rules. She talked about how to derive rules from processes, and use them as guides to the process. There’s a number of process-related problems that can occur when the rules are not explicit: assumed policies, activities with experience as the only guide, and inconsistent (and therefore likely non-compliant) processes. The key things to consider when analyzing the guides for a process can be focussed around what happens at a given activity (are what knowledge is required, what decisions are required, what reports have to be generated) as well as a number of other factors; she presents a number of different questions to ask in order to drive out the rules and make them explicit.

She also made a distinction between policies and rules, where the key differentiator is that rules are actionable, whereas policies must be interpreted into more concreted business rules in order to take action. Within rules, there’s both structural rules (can’t be broken) and operative rules (which have a bit more wiggle room); this sounds a bit like the distinction between a fact and a rule that I heard in a session yesterday, which makes me unsure that there’s a really common vocabulary for some of these things.

Looking at some of their process analysis techniques, she presented categories of activities as real value-added (impact the customer’s requirements), business value-added (required to run business, such as regulations), and non value-added (that 65-85% of work that doesn’t contribute to either RVA or BVA). There’s a whole list of verbs — adjust, approve, expedite, inspect, verify and many others — that tend to indicate that activities are NVA and should be considered for elimination. Many of these are because something wasn’t done right the first time; a lot of the NVA activities can be cut if there’s ways to reduce the error rates in the RVA and BVA activities. This isn’t, of course, really about rules: it’s about process improvement. Sure, the appropriate addition of business rules can certainly lead to process improvement, but it’s also about the myriad other ways that we improve processes, such as establishing accountability and eliminating unneeded steps that are only there for historical reasons. Some thoughts that Long gave us to take away:

  • The greatest opportunity to improve a process is by changing the rules
  • Challenge all policies
  • Validate all compliance interpretations
  • Eliminate the use of assumed policies
  • Ensure that all rules are documented including the use of experience/knowledge
  • Create consistent rules across the enterprise
  • Structure rules so that they can be easily changed
  • Allow the business to design its own processes

I was surprised that she didn’t talk at all about some of the technology issues such as how BPM and BR can be used together to improve processes, but her focus was not at all on technology: her only case study was about improving a process based on a manual procedural change.

I have to head off for a mid-day flight home, so that was the end of my Business Rules Forum experience. I’ve actually learned a lot here, which has made my time here definitely worthwhile. However, I’m still left with the feeling that I mentioned on my first post back on Tuesday: we need to start having much more crossover between different technology areas such as BPM and BR. I’ve been writing since mid-2005 about the importance of looking at BPM and BR together, but in spite of the technology advances that have occurred since then to facilitate this, I’m not seeing much happening in the real world.

BRF Day 2: How Business Rules Re(Define) Business Processes: A Service Oriented View

For the last session today, I attended Jan Venthienen’s session; he’s a professor at Katholieke Universiteit Leuven. He talked about different representations of rules, particularly decision tables (at length, although in an interesting way). He talked about the problems with maintaining decision trees, then as he moved on to business processes, he showed how a business process with the rules encoded in the process as routing logic was really just a form of decision tree, and therefore difficult to maintain from a rules integrity standpoint. As rules are distilled out of and separated from the processes, the processes become thinner and thinner, until you have a single branch straight-through flow. I have the feeling that he’d like to reduce the process to a single activity, where everything else is done in a complex rule called from that step. I’m not sure that I agree with that level of stripping of logic out of the process and into the rules; there’s value in having a business process that’s understandable by business users, and the more that the logic is encapsulated in rules, the harder it is to understand how the process flow works by looking at the process map. The critical thing is knowing which rules to strip out of the business process, and which to leave in.

He’s doing research now to determine if it’s possible to specify business rules, then automatically derive the business process from the rules; an interesting concept. In order to do this, there must be rules that constrain the permission and obligations of the actors in the process, e.g., an order must be accepted before the product is shipped. This presents two possible architectural styles: process first, or rules first. In either case, what is developed is an architecture of rules, events and services, with a top layer of business rules and processes, a middle layer of services and components, and a bottom layer of enterprise applications.