Column 2 : SOA

{ Category Archives } SOA

Software AG Technology Innovation Fair

For once, I don’t need to travel to see a (mini) vendor conference: Software AG has taken it on the road and is here in Toronto this morning. I wanted to get an update of what’s happening with webMethods since I attended their user conference in Miami last November, and this seemed like a good way to do it. Plus, they served breakfast.

Susan Ganeshan, SVP of Product Management, started the general keynote with a mention of Adabas and Natural, the mainframe (and other platforms) database and programming language that drive so many existing business applications, and was likely the primary concern of many of the people in the room. However, webMethods and CentraSite are critical parts of their future strategy and formed the core of the rest of the keynote; both of these have version 8 in first-customer-ship state, with general availability before the end of the year.

First, however, she talked about Software AG’s acquisition of IDS Scheer, and how ARIS fits into their overall plan, following on today’s press release about how Software AG has now acquired 90% of IDS Scheer’s stock, which should lead to a delisting and effective takeover. She discussed their concept of enterprise BPM, which is really just the usual continuous improvement cycle of strategize/discover and analyze/model/implement/execute/monitor and control that we see from other BPMS vendors, but pointed out that whereas Software AG has traditionally focused on the implement and execute parts of the cycle, IDS Scheer handles the other parts in a complementary fashion. The trick, of course, will be to integrate those seamlessly, and hopefully create a shared model environment (my hope, not her words). They are also bringing a process intelligence suite to market, but no details on that at this time.

Interesting message about the changing IT landscape: I’m not sure of the audience mix between Adabas/Natural and webMethods, but I have to guess based on her “intro to BPM” slides that it is heavily weighted towards the former, and that the webMethods types are more focused on web services than BPM. She also invokes the current mantra of every vendor presenter these days about how the new workforce has radically different expectations about what their computing environment should look like (“why can’t I google for internal documents?”); I completely agree with this message, although I’m sure that most companies don’t yet have that as a high priority since much of the new workforce is just happy to have a job in this economy.

She discussed the value of CentraSite – or at least of SOA governance – as being a way to not just discover services and other assets, but to understand dependencies and impacts, and to manage provisioning and lifecycle of assets.

A few of the BPM improvements:

Also a common message from BPMS vendors this week, she talked about their composite application environment, a portal-like dynamic workspace that can be created by a user or analyst by dragging portlets around, then saved and shared for reuse. This lessens the need for IT resources for UI development, and also allows a user to rearrange their workspace the way it best works for them.
They’ve also added ad hoc collaboration, which allows a process participant to route work to people who are not part of the original process; it’s not clear if they can add steps or subprocesses to the structured process, or whether this is a matter of just routing the task at its current step to a previously unidentified participant.
They integrate with Adobe Forms and Microsoft Infopath, using them for forms-driven processes that use the form data directly.
They’ve integrated Cognos for reporting and analytics; it sounds like there are some out of the box capabilities that run without additional licensing, but if you want to make changes, you’ll need a Cognos license.

Since the original focus of webMethods was in B2B and lower-level messaging, she also discussed the ESB product, particularly how they can provide high-speed, highly-available messaging services across widespread geographies. They can provide a single operational console across a diverse trading network of messaging servers. There’s a whole host of other improvements to their trading networks, EDI module and managed file transfer functionality; one interesting enhancement is the addition of a BPEL engine to allow these flows to be modeled (and presumably executed) as BPEL.

They have an increased focus on standards, and new in version 8 are updates to XPDL and BPEL support, although they’re still only showing BPMN 1.1 support. They also have some new tooling in their Eclipse-based development suite.

She laid out their future vision as follows:

Today: IT-driven business, with IT designing business processes and business dictating requirements
2009 (um…isn’t that today?): collaborative process discovery and design; unified tooling
2010: business rules management and event processing; schema conformance
2012: personalized, smart-healing processes; centralized command and control for deployment and provisioning
2014: business user self-service and broad collaboration without organizational boundaries; elastic and dynamic infrastructure

She finished up with a brief look at AlignSpace for collaborative process discovery; I’m sure that someday, they will approve my request for a beta account so that I can take a closer look at this. Not only process discovery and modeling, however, AlignSpace will also provide a marketplaces of resources (primarily services) related to processes in particular vertical industries.

They have a complete fail on both wifi and power here, but I no longer care: my HP Mini has almost six hours of battery life, and my iPhone plan allows me to tether the netbook and iPhone to provide internet access (at least in Canada).

Tagged CentraSite, Software AG, webMethods

2009 10 23 Sandy Kemsley BPM Comments (1)

Process Design Slam 2009 – The Final Judgement #SAPTechEd09 #BPXslam09

To wrap up the proceedings from last night, I was asked to critique the efforts of the groups and pick a winner: as it turned out, I was the only judge. Each of the groups did great work, and I want to call out some of the specific efforts:

The Business Use Case group had a great written story, including a lot of cultural and social background for our fictional city in order to provide context for the implementation.
The BPM Methodologies group had excellent documentation on the wiki, including graphics and charts to make it clear how the methodologies fit with the other groups.
The Business Rules group were stars at collaboration with the other groups, in part because everyone quickly realized the importance of business rules to data, UI and process, and solicited their input.
The UI and Dashboards group created mockups of monitoring dashboards that provide a starting point for future design slam work.
The Collaborative Modeling group led at international collaboration, using Gravity (process modeling within Google Wave) interactively with team members in Europe during the session, and produced a business process model.
The Service Implementation group also kicked off implementation, creating a service orchestration process model as a starting point.

In general, everyone seemed to have a good understanding of the importance of data, rules and process, but there could have been better cross-pollination between the groups; in future design slams, that could be helped by requiring some group members to move partway through the evening in order to ensure that there is a better understanding on both sides, something that is fairly common in real-life businesses where people are seconded from one department to another for part of a project. Although a certain amount of collaboration did occur, that was one area that requires more work. I saw one tweet that referred to the design slam as crowdsourced rather than collaborative, although I’m not sure that I would say that: crowdsourcing usually has more of a flavor of individuals contributing in order to achieve their own goals, whereas this was a collaboration with common goals. However, those goals were a bit fragmented by group.

Another issue that I had was the lack of an architectural view of process design: although all of the groups are contributing to a common process (or set of processes), there is little thought around the transformations required to move the process list developed by the Business Use Case group to the process model developed by the Collaborative Modeling group to the process design developed by the Service Implementation group. In enterprise architecture terms, this is a case of transforming models from one layer to another within the process column of the architecture (column 2 if you’re a Zachman fan); understanding these transformations is key so that you don’t reinvent the process at each layer. One of the goals of model-driven design is that you don’t do a business-level process model, then redraw it in another tool; instead, the business-level process model can be augmented with service-level information to become an executable process without recreating the model in another tool. In reality, that often doesn’t happen, and the business analysts draws a process in one tool (such as Visio, or in the case of the design slam, Gravity), then IT redraws it in a tool that will create an executable process (NetWeaver in this case). I have a couple of suggestions here:

Combine the Business Use Case and Collaborative Modeling groups into a single group, since they are both doing high-level business analysis. This would allow the process list to be directly modeled in the same group without hand-off of information.
Reconsider the use of tools. Although I have a great deal of appreciation for Gravity (I am, after all, a geek), the fact that it does not share a model with the execution environment is problematic since the two groups creating process models were really off doing their own thing using different tools. Consider using NetWeaver 7.2, which has a business analyst perspective in the process composer, and having the business use case/collaborative modeling group create their initial non-technical models in that environment, then allow the service implementation team to add the technical underpinnings. The cool Wave collaboration won’t be there, or maybe only as an initial sketching tool, but the link will be made between the business process models and the executable models.

When it came down to a decision, my choice of the winner was more a product of the early state of the design slam rather than the efforts or skills of the group: I suspect that my view would change if I were judging in Vienna or Bangalore when the process is further along. I selected the Business Use Case group as the winner at this point based on the four judging criteria: although they failed to include alternative media, their story was clear and well-written, it fit well with the other groups’ efforts, and they used good social and collaborative methods within their group for driving out the initial solutions.

The winning team was made up of Greg Chase, Ulrich Scholl and Claus von Riegen, all of SAP, with input from a few others as subject-matter experts on public utilities and electricity production, and started the discussions on pricing plans that ended up driving much of the Business Rules group’s work. Ulrich also has solar cells on his house that connect to the grid, so he has in-depth knowledge of the issues involved with micro-generation, and was very helpful at determining the roles involved and how people could take on multiple roles. They leveraged a lot of the content that was already on the wiki, especially references to communities with experience in micro-generation and virtual power plants. Besides this initial leg up on their work, they were forced to work fast to produce the initial use cases and processes, since that provided necessary input to the other groups to get started with their work, which left them with more of the evening to write a great story around the use case (but, apparently, not enough time to add any graphics or multimedia).

There was a huge amount of effort put into the design slam, both in the preceding weeks through conference calls and content added to the wiki, and at the session last night in Phoenix. I believe that a huge amount of groundwork has been laid for the design slams upcoming in Vienna and Bangalore, including process model, service orchestration diagrams, business rules decision tables, and monitoring dashboard mockups.

I had a great time last night, and would happily participate in a future process design slam.

2009 10 15 Sandy Kemsley BPA
BPM
BRM
SAP Comments (0)

Process Design Slam 2009 #SAPTechEd09 #BPXslam09

8pm

We’re just getting started with the Process Design Slam: one of the face-to-face sessions that make up the collaborative design process that started a couple of months ago on the Design Slam wiki. Marilyn Pratt has identified the six groups that will each work on their part of the design, collaborating between groups (a.k.a. poaching talent) as required, and even bringing in people from the Hacker Night and Business Objects events going on in the same area.

Business Use Case, led by Greg Chase
Collaborative Modeling, led by David Herrema
Business Rules, led by James Taylor
Service Implementation, led by John Harrikey
BPM Methodologies, led by Ann Rosenberg
UI and Dashboards, led by Michelle Crapo

Right now, everyone has formed into initial groups based on their interests, and is having some initial discussions before the food and beer arrives at 8:30. Since there was an initial story and process model developed by the online community, everyone is starting at something close to a common point. Participants within a group (and even the leaders) could change throughout the evening.

By the end of the night, each team will have created a story about their work, and give a 5-minute presentation on it. The story must include additional media such as video and images, and in addition to the presentation, it must be documented on the wiki. Each story must also be related to the output of the other teams – requiring some amount of collaboration throughout the evening – and include pointers on what worked and didn’t work about their process, and what they would do differently in the future.

At that point, the judging panel, which includes me plus Marc Rosson, Uli Scholl, Ann Rosenberg and Dick Hirsch, will render our judgment on the creations of the groups based on the following criteria:

Clarity and completeness of the story on the wiki, particularly if it could be understood without the presentation.
Creative use of media.
How well this story ties into the overall storyline of the night.
The social process that was used to create the story.

I’m floating around between groups to listen in on what they’re doing and some of their initial thoughts.

8:30pm

Beer o’clock. The Business Rules team is still deep in conversation, however, and Business Use Case comes over to them to ask for help in bringing the business rules and business use case together. Business Use Case outlines the actors that they have identified, and the high-level business processes that they have identified in addition to the initial business process of bringing new consumer-producers online.

9pm

BPM Methodologies has a much wider view than just this project: developing methodologies that can be used across (SAP) BPM projects, including assessing the business process maturity of an organization in order to determine where they need to start, and identifying the design roles. In the context of the design slam, they will be helping to coordinate movement of people between the teams in order to achieve the overall goals.

9:30pm

Service Implementation – viewed by groups such as Business Use Case as “the implementers” – have revised the original process map from a service standpoint; looking at the services that were required led to a process redesign. They are using the Composite Designer to model the service orchestration, including the interfaces to the services that they need and external services such as FirstLook, an wind assessment service based on location data. In their service orchestration process, they assume that the process is initiated with the data gathered from a user interface form, and they focus primarily on the automated process steps. Ginger Gatling doesn’t let me leave the table until I tell them what they have to do to win; I advise them to update the wiki with their story.

9:50pm

The Collaborative Modeling group is modeling the business process using Gravity, online with a couple of participants in Europe. This is a process model from a business standpoint, not an executable model; there is no concept of the linkage between this and what is being done by the Service Implementation team. I suggest that they should head over there to compare processes, since these should (at some level) just be different perspectives on the same process.

10pm

Business Use Case is identifying the necessary processes based on their earlier collaboration with Business Rules: this has given them a good understanding of business case, goals and incentives. They’re considering both human and automated usages, and have fed their results to the UI, Business Rules and Collaborative Modeling teams.

10:10pm

Business Rules states that they’ve had to gather information from numerous sources, and the challenge is to sequence it properly: data is captured by the UI, but is driven by the Business Use Case. They didn’t work with the Collaborative Modeling group directly, but there are links between what they do and what’s happening in the process. They’re also interested in using historical usage data to determine when to switch consumers between usage plans.

10:20pm

UI and Dashboards managed to recruit a developer who is actually coding some of their interfaces; they were visited by many of the other groups to discuss the UI aspects, since the data gathered by the UI drives the rest of the process and rules, and the data generated by the process drives the dashboard interfaces. They feel that they had the best job since they could just be consumers and visualize the solutions that they would like to have.

10:35pm

Presentations start. Marilyn Pratt is being the MC, and Greg Chase is wrangling the wiki to show what has been documented by each of the groups. Half of the Service Implementation team just bailed out. I have to start paying attention now. Checking out the wiki pages and summarizing the presentations:

Business Use Case worked with the UI, Collaborative Modeling and Business Rules teams, since those teams required the business use cases in order to start their work. They developed a good written story including cultural/social background about the fictional city where the power generation plan would go into effect. They defined the roles that would be involved (where one person could take on more than one role, such as a consumer that is also a producer), and the processes that are required in order to handle all of the use cases. They did not use any presentation/documentation media besides plain text.
BPM Methodologies had excellent documentation with the use of graphics and tables to illustrate their points, but this was a quite general methodology, not just specific to this evening’s activities. They worked briefly with the other groups and created a chart of the activities that each of these groups would do relative to the different phases in the methodology. I found the methodology a bit too waterfall-like, and not necessarily a good fit with the more agile collaborative methods needed in today’s BPM.
Business Rules focused on the rules related to signing up a new user with the correct pricing plan, documenting the data that must be collected and an initial decision table used to select a plan, although no graphics or other non-text media. They worked with the Business Use Case team and the UI team to drive the underlying business use cases and data collection.
UI and dashboards created the initial mockups that can be used as a starting point for the design slam in Vienna in a couple of weeks. They worked with Business Rules and Business Use Case in order to nail down the required user data inputs, and what is required for monitoring purposes, and included some great graphics of the monitoring dashboards (although not the data collection form).
Collaborative Modeling used Gravity (process modeling in Google Wave) not just for modeling with the group around the table, but also with participants in Germany and the Netherlands. They included photos of the team as well as screen snaps of the Gravity Wave that they created, although the text of the story documented on the wiki isn’t really understandable on its own. I’m not sure that they spent enough time with other groups, especially the Service Implementation group.
Service Implementation talked to the Business Rules and UI teams to discuss rules and data, but felt that they were running blind since there wasn’t enough of the up-front work done for them to do any substantial work. They used placeholders for a lot of the things that they didn’t know yet, and modeled the service orchestration. The documentation in the wiki is very rudimentary, although includes the process map that they developed; it’s not clear, however, how the process model developed in Collaborative Modeling relates to their map.

11:30pm

And now, on to the judging – I’ll write up the critique and results in a later post.

2009 10 15 Sandy Kemsley BPA
BPM
BRM
SAP Comments (1)

Can packaged applications ever be Lean? #BTF09

Chip Gliedman, George Lawrie and John Rymer participated in a panel on packaged applications and Lean.

Rymer argued that packaged apps can never be Lean, since most are locked down, closed engines where the vendor controls the architecture, they’re expensive and difficult to upgrade, they use more functions than customers use, they provide a single general UI for all user personas, and each upgrade includes more crap that you don’t need. I tend to be on his side in this argument about some types of appls (as you might guess about someone who used to write code for a living), although I’m also a fan of buy over build because of that elusive promise of a lower TCO.

Gliedman argued the opposite side, pointing out that you just can’t build the level of functionality that a packaged application provides, and there can be data and integration issues once you abandon the wisdom of a single monolithic system that holds all your data and rules. I tend to agree with respect to functionality, such as process modeling: you really don’t want to build your own graphical process modeler, and the alternative is hacking your own process together using naked BPEL or some table-driven kludge. Custom coding also does not guarantee any sort of flexibility, since many changes may require significant development projects (if you write bad code, that is), rather than a package app that may be more configurable.

It’s never a 100% choice between packaged apps and custom development, however: you will always have some of each, and the key is finding the optimal mix. Lean packaged apps tend to be very fit-to-purpose, but that means that they become more like components or services than apps: I think that the key may be to look at composing apps from these Lean components rather than building Lean from scratch. Of course, that’s just service-oriented architecture, albeit with REST interfaces to SaaS services rather than SOAP interfaces to internal systems.

There are cases where Lean apps are completely sufficient for purpose, and we’re seeing a lot of that in the consumer Web 2.0 space. Consider Gmail as an alternative to an Exchange server (regardless of whether you use Outlook as a desktop client, which you can do with either): less functionality, but for most of us, it’s completely sufficient, and no footprint within an organization. SaaS, however, doesn’t not necessarily mean Lean. Also, there are a lot of Lean principles that can be applied to packaged application deployment, even if the app itself isn’t all that Lean: favoring modular applications; using open source; and using standards-based apps that fit into your architecture. Don’t build everything, just the things that provide your competitive differentiation where you can’t really do what you need in a packaged apps; for those things where you are doing the same all every other company, suck it up and consider a packaged app, even if it’s bulky.

Clearly, Gliedman is either insane or a secret plant from [insert large enterprise vendor name here], and Rymer is an incurable coder who probably has a ponytail tucked into his shirt collar. Nonetheless, an entertaining discussion.

2009 10 08 Sandy Kemsley Forrester
Software design Comments (5)

Patterns for Business Process Implementations #GartnerBPM

Benoit Lheureux from Gartner’s Infrastructure and Architecture group gave a presentation on process implementation patterns. I think that he sees BPM as just part of SOA, and presents as such, but I’m willing to give him a pass on that.

He discussed five styles of flow management in SOA:

Microflows: fine-grained services implemented via flows amongst software components. This is a process from a software development standpoint, not a business-level process: probably 3GL code snippets assembled into what we old-timers might refer to as a “subroutine”.
Service composition: coarse-grained services implemented by assembling fine-grained flows (microflows). This may be done with a BPMS tool, but is low-level service composition rather than business processes.
Straight-through process: automating business processes involving multiple services across systems, but without human intervention.
Workflow: pretty much the same as STP, but with human intervention at points in the process.
Semi-structured processes: a combination of structured processes with unstructured activities or collaboration.

He has some good strategic planning assumptions based on these four patterns, such as 75% of companies will use at least three different products to implement at least three different styles of flows. His primary focus, however, is on B2B, and how internal process connect to multi-enterprise processes, and the ultimate goal of shared process execution across enterprises. This led to the four B2B flow management styles:

Blind document/transaction exchange: loosely-coupled, with each partner managing their own internal processes, and no visibility outside their own processes.
Intelligent document/transaction exchange: visibility across the shared process to provide a shared version of the truth, such as a BAM dashboard that provides an end-to-end view of an order-to-cash process across enterprises. Although this isn’t that popular yet, it is providing significant benefits for companies that are implementing it, and Lheureux estimates that 50% of B2B relationships will include this by 2013.
Multi-enterprise applications: shared execution of a process that spans the enterprises, such as vendor-managed inventory. This may be hosted by one of the partners, or may be hosted by a third-party service provider.
Multi-enterprise BPMS and rules: centralized processes and rules, such as shared compliance management on a shared process. By 2013, he predicts that at least 40% of new multi-enterprise integration projects will leverage BPMS technology.

He showed a chart that I’ve seen at earlier conferences on identifying process characteristics, classifying your processes as case management, form-driven workflow, content collaboration, multiparty transactional workflow, participant-driven workflow, and optimization of network relationships based on the unit of work, process duration, degree of expertise required, exception rate, and critical milestones that progress work. Then, consider when to use BPMS technology rather than code when there are specific process characteristics such as complexity and changeability.

The final recommendations: don’t try to use the same tool to handle every type of process implementation, but be aware of which ones can be best handled by a BPMS (and by different types of BPMS) and which are best handled in code.

2009 10 05 Sandy Kemsley BPM
Gartner Comments (3)

Tutorial: enabling flexibility in process-aware information systems #BPM2009

Manfred Reichert of Ulm University and Barbara Weber of University of Innsbruck presented a tutorial on the challenges, paradigms and technologies involved in enabling flexibility in process-aware information systems (PAIS). Process flexibility is important, but you have to consider both build time flexibility (how to quickly implement and configure new processes) and run time flexibility (how to deal with uncertainty and exceptional cases during execution), as well as their impact on process optimization.

We started by looking at the flexibility issues inherent in the imperative approach to BPM, where pre-defined process models are deployed and executed, and the execution logs monitored (in other words, the way that almost all BPMS work today). As John Hoogland discussed this morning, there are a number of flexibility issues at build time due to regional process variations or the lack of a sufficient information about decisions to build them into the process model. There’s also flexibility issues in the run time, mostly around exception handling and the need for ad hoc changes to the process. As all this rolls back in to the process analyst through the execution monitoring, it can be used to optimize the process model, which requires flexibility in evolving the process model and impacting work in progress. The key problem is that there are way too many variants in most real-life processes to realistically model all of them: there needs to be a way to model a standard process, then allow user-driven configuration (either explicitly or based on the instance parameters) at run time. The Provop approach presented in the tutorial allows for selective enabling and disabling of process steps in a master model based on the instance conditions, with a lot of the research based on the interaction between the parameters and the soundness of the resultant models.

Late binding and late modeling approaches use a pre-specified business process with one or more placeholder activities, then the placeholder activities are replaced with a process fragment at run time either from a pre-determined set of process fragments or a process fragment assembled by the user from existing activity templates (the latter is called the “pockets of flexibility” approach, a name that I find particularly descriptive).

Up to this point, the focus has been on changes to the process model to handle variability that are part of normal business, but possibly not considered exceptions. Next, we looked at runtime exception handling, such as the failure or unavailability of a web service that causes the normal process to halt. Exceptions that are expected (anticipated) can be handled with compensation, with the compensation events and handler built into the process model; unexpected exceptions may be managed with ad hoc process changes to that executing instance. Ad hoc process changes can be a bit tricky: they need to be done at a high level of abstraction in order to make it understandable to the user making the change, yet the correctness of the changes must be validated before continuing. This ability needs to be constrained to a subset of the users, and the users who can make the changes may require some assistance to do this correctly.

This was a good tutorial, but I wanted to catch the process mining session so skipped out at the break and missed the last half.

2009 09 09 Sandy Kemsley BPA
BPM
BPM2009 Comments (2)

Community participation in a hosted BPM system #BPM2009 #BPMS2’09

Rania Khalaf of IBM’s T.J. Watson Research Center presented a paper on enabling community participation for workflows through extensibility and sharing, specifically within a hosted BPM system.

She is focused on three areas of collaboration: extension activities (services), collaborative workflow modeling, and collaboration on executing workflow instances. There are two key aspects to this: method and technical enablement, and the business and security aspects.

This is really about the community and how they participate: developers who create extension activities, process designers who create process models and include the extension activities, and participants in the executing workflows. For extension activities, they’re leveraging the Bite language and runtime, which uses REST-based interaction, and allows developers to create extensions in their language of choice and publish them directly in a catalog that is available to process designers. Workflow designers can provide feedback on the extensions via comments. Essentially, this is a sort of collaborative SOA using REST instead of WS-*: developers create extensions and publish them, and designers can pull from a marketplace of extensions available in the hosted environment. Much lighter weight than most corporate SOA efforts, although undeniably more nimble.

Process models can be shared, either for read-only or edit access, with others both within and outside your organization in order to gather feedback and improve the process. Once created, the URL for instantiating a process can be sent directly to end users to kick off their own processes as designed.

This is part of several inter-related IBM efforts, including the newly-released BPM BlueWorks and the still-internal Virtuoso Business Mashups, and seems to fall primarily under the LotusLive family. This is likely an indication of what we’ll see in BlueWorks in the future; they’ll be adding more social capabilities such as process improvement and an extensions marketplace, and addressing the business and security issues.

2009 09 07 Sandy Kemsley BPM
BPM2009
cloud Comments (0)

The Open Group’s Service Integration Maturity Model and SOA Governance Framework

I had a chance last week for a pre-release briefing from The Open Group’s Chris Harding, Forum Director for SOA and Semantic Interoperability, on two new standards that they are releasing today: the Service Integration Maturity Model (OSIMM) and the SOA Governance Framework. These are both vendor-neutral (although several large vendors were involved in their creation), and are available for free on The Open Group’s site. In their words:

OSIMM will provide an industry recognized maturity model for advancing the continuing adoption of SOA and Cloud Computing within and across businesses. The SOA Governance Framework is a free guide for organizations to apply proven governance standards that will accelerate service-oriented architecture success rates.

OSIMM is a strategic planning tool: it is used to assess where you are in your SOA initiatives relative to a standard, vendor-neutral maturity model, and help create a roadmap for how to move on to the higher levels of maturity. At the heart of it is the OSIMM matrix, with maturity levels as columns progressing from left to right, and the different organizational dimensions being measured as rows: business view, governance and organization, methods, applications, architecture, information, and infrastructure and management.

Within each cell of the matrix are the indicators for that dimension and maturity level: for example, if you’re using object oriented modeling methods, that indicates that your methods are at level 2, whereas using service oriented modeling would move you up to level 4 or 5 in the methods dimension. Behind this matrix, OSIMM includes a full set of maturity indicators and attributes, plus assessment questions that organizations can use to determine where they are in terms of maturity: each dimension can be (and likely will be) at a different level of maturity.

This has the potential to be an incredibly useful self-assessment tool for organizations: rather than the very product-specific measurements that you see from vendors (“Not using our product? Oh, you’re not at all advanced in your SOA efforts…”), this is independent of whatever products that you’re using: it’s more about the type of products, and the methods and governance that you’re using to apply them. You’ll be able to use it to understand services and SOA, assess the maturity of your organization, and develop a roadmap to reach your goals.

The first version of the OSIMM Technical Standard will be available here for free download, although that link was still not working at the time that I wrote this. Other industry-specific standards organizations are free to use OSIMM directly, or extend it with their own dimensions and indicators as required.

The other major announcement today is about the SOA governance framework, which helps an organization to define their governance processes and methods. This is more of a practical framework for defining policies aligned between business and IT, aiding communication and capturing vendor-neutral best practices. This includes best practices around both lifecycle management and portfolio management, for both services and service-based solutions.

Lifecycle and portfolio management are quite different: for example, a service lifecycle would include the idea or motivator for the service, the service definition, service creation, putting the service into operation, modifying and maintaining the service, and eventually retiring the service from operation. Service portfolio management is more concerned with reusability, and the practice of looking in the portfolio in the early stages of service lifecycle to see if there is an existing service that suits the requirements. The same applies to solution lifecycle and portfolio management; this differs from any other type of solution governance since there may be service-specific issues such as composition to be considered.

This generic reference model for SOA governance is provided as a standard, to be used by companies to create (and constantly monitor and update) their own specific governance model and best practices. The SOA governance framework may be used in the context of another governance framework, such as COBIT or ITIL; the SOA working group did a mapping of COBIT to this framework as part of the framework development process, and plan to do more in the future in order to help organizations preserve their investment in COBIT/ITIL training and implementation.

The SOA Governance Framework will be available here for free download.

2009 08 24 Sandy Kemsley EA Comments (2)

SAP NetWeaver BPM

This post is both long, and long overdue. It’s based on several online sessions with Donka Dimitrova and Jie Deng of the SAP NetWeaver BPM product management team, then an update with Wolfgang Hilpert and Thomas Volmering at SAPPHIRE in May when the product entered unrestricted release. In the past few weeks, there’s been a series of “Introduction to SAP NetWeaver BPM” posts by Arafat Farooqui of Wipro on the SAP SDN site (part 1, part 2, part 3 and part 4, which are really about how to hook up a Web Dynpro UI to a human task in BPM, then invoke a process instance using web services from a portal), and I’m inspired to finally gather up all my notes and impressions.

The driver for BPM with SAP is pretty obvious: Business Workflow within the SAP ERP suite just isn’t agile or functional enough to compete with what’s been happening in BPM now, and SAP customers have been bringing in other BPM suites for years to complement their SAP systems. I had to laugh at one of Dimitrova’s comments on the justification for BPM during our discussion – "process changes in an ERP are difficult and require many hours from developers" – oh, the irony of this coming from an SAP employee!

The Eclipse-based Process Composer is part of the NetWeaver Developers’ Studio, and is used to create processes in the context of other development tools, such as the Yasu rules engine (which they bought) and user interfaces. Like most modern BPMS’, what you draw in the Process Composer (in BPMN) is directly executed, although user interfaces must be created in other development tools such as Web Dynpro or Adobe Interactive forms, then linked to the process steps. There are future plans to generate a UI from the process context data or provide some sort of graphics forms designer in place, but that’s not there yet.

As with most Eclipse-based process modelers that I’ve seen, Process Composer has multiple perspectives for different types of process design participants, with a shared process model. Initially, there is only a process architect (technical) perspective in the modeler, and the business analyst view will be released this year. Future releases will include a line-of-business manager view to see task sequences and parallelism, but no details of gateways; and an executive view of major phases with analytics and KPI dashboards.

There is no link between ARIS-based modeling (SAP Enterprise Modeling Applications by IDS Scheer) and the NetWeaver BPM in this version; integration is planned for later version, although it will be interesting to see how that plays out now that IDS Scheer has been purchased by Software AG, which competes with SAP in (at least) the BPM arena.

Although all you can do now is create your BPM processes in this environment, in the future, there’s plans to have a common modeler and composition environment provide visibility into ERP processes, too, which will be a huge selling point for existing SAP customers who need more agility in their ERP processes. This common process layer will provide not just a unified design experience, but common runtime services, such as monitoring and performance management.

One huge issue from an orchestration standpoint is the lack of support for asynchronous web services calls, meaning that you have to use the existing NetWeaver Process Integrator (PI) environment to create system-centric processes, then invoke those (as synchronous web services) from NetWeaver BPM as required. I didn’t get a clear answer on future plans to merge the two process management platforms; keeping them separate will definitely cause some customer pushback, since most organizations don’t want to buy two different products to manage system-centric and human-centric processes, as they are encouraged to do by stack vendors such as IBM and Oracle.

Taking a look at the Process Composer environment, this is a fairly standard Eclipse-based BPMN process modeling environment: you create a process, add pools, add steps and link them together. For human-facing tasks, you use the properties of the step to connect it to a UI for that step, which must already be built by a developer using something like Web Dynpro. As I mentioned previously, the first version only has the “process architect” perspective, and is targeted at creating human-centric processes without full orchestration capabilities, since that’s what SAP’s customers who were involved in the product development cycle said that they most wanted. The environment is fairly technical, and I wouldn’t put it in front of any but the most technical of business analysts.

Roles can be set by lanes and overridden by task role assignment, which allows using the lanes for a department (for example) and overriding manager-specific tasks without moving them to another lane. Also, expressions can be used to assign roles, such as manager of the user that started the process. User IDs, roles and groups are pulled from the NetWeaver user management engine (UME).

Each step can have other properties, including deadlines (and the events that occur when they are exceeded) and user texts that appear for this step in the user worklist, which can include parameters from the process instance. These are all maintained (I think) in a task object, which is then associated with a step on the process map; that allows the same task to be easily reused within the same process or across processes.

There are a number of things that I like about Process Composer:

Some nice UI pop-ups on the process map to make specifying the next step easier.
An explicit process data model, called the process context, driven by master data management concepts; this is used for expressions and conditions in gateways, and to map to the inputs and outputs of the UI of human steps or the service interface of automated steps. It can be imported as an XSD file if you already have the schema elsewhere.
The visuals used to map and transform from the process context to a human or web service step make it obvious what’s getting mapped where, while allowing for sophisticated transformations as part of the mapping. Furthermore, a mapping – including transformation functions – can be saved and reused in other processes that have the same process context parameters.
Lots of fairly obvious drag-and-drop functionality: drag a task to create a step on a process map, drag a role to assign to a pool, or drag a WSDL service definition to create a system task.
Nice integration of the Yasu rules engine, which can be purely within the context of the process with rules appearing as functions available when selecting gateway conditions, or as a more loosely-coupled full rules engine.

Process Composer is just one tab within the whole NetWeaver Project Explorer environment: you can open other tabs for UI design, rules and other types of components. This allows the process to be visible while rules are being modeled, for example: handy for those of us with a short attention span. Rules are created using decision tables, or by writing in a Java-based rules language; Dimitrova referred to the latter as being “a bit complicated for business people”, which is a bit of an understatement, although decision tables are readily usable by business analysts. Future releases will have a business perspective in the rules modeler.

The Rules Composer is a full rules modeling environment, including debugging for incomplete or over-constrained rules in a decision table, and rules versioning. Parameters from a process context can be passed in to rules. Rules can be exposed as web services and called just like any other web service; in fact, although there is tight integration between the rules and process environment allowing for easy creation of a rule directly from within the Process Composer perspective, the rules management system is a separate entity and can be used independent of BPM: really the best of both worlds.

Having spent about 3 sessions going through the design environments, we moved on to process execution. Processes can be initiated using a web services call, from an Adobe form, or manually by an administrator. Since process models are versioned, all of the versions available on the server can be seen and instantiated.

Human tasks can be seen in the SAP Universal Worklist (UWL) through a connector that I heard about at SAPPHIRE, appearing along with any other tasks that are surfaced there including SAP ERP tasks or other systems that have developed a connector into the UWL. I like the unified inbox approach that they’re presenting: other BPM systems could, in fact, add their own human tasks in here, and it provides a common inbox that is focused on human workflow. Although an email inbox could be used for the same purpose, it doesn’t provide adequate management of tasks from a BPMS. The UWL is fairly independent from NetWeaver BPM; this is just one way to provide a worklist of BPM tasks that is provided by SAP in a portal environment, but it doesn’t have to be done that way.

Once a task is selected and opened, there is a frame across the top with standard task information that will be common across all tasks: information such as start date, deadline and status; common task functions of Close, Delegate and Revoke; and notes and attachments to the task. Below that is the Web Dynpro UI form that was connected to that task in the Process Composer, which contains the instance data that is specific to the process context for this process. The user can interact with that form in whatever manner specified by the Web Dynpro developer, which might involve accessing data from databases or ERP systems; that part is completely independent of NetWeaver BPM.

The user can also click through to a process view showing where they are in the context of the entire process map, plus runtime task parameters such as priority and start date.

Considering the all-important areas of monitoring and management of work in progress, that’s a bit weak in the first version. In the next version, there will be a dashboard showing process status and cycle time, with drill-down to process instances, combining exported BI data and realtime work in progress statistics. There is no way to update the process design of work in progress; there are actually only a few BPMS that do this very well, and most either don’t do it at all or require manual modification of each instance. Wherever possible, things that might change should be put into business rules, so that the correct rule is invoked at the point in time that it is required, not when the process instance was created.

At the end of all the demos, I was impressed with what SAP has released for a version 1.0, especially some of the nice handling of data and rules, yet aware of the many things that are still missing:

task UI generation
simulation
KPI measurement
asynchronous web services calls
links to/from ARIS
common process composition environment across BPM and ERP processes
BPEL translation
business analyst perspective in process and rules modelers
BPMN 2.0 support
strategy for merging or coexisting with NetWeaver process orchestration platform

In the briefing at SAPPHIRE, I did see a bit of the roadmap for what’s coming in the next year or two. In 2009, the focus will be on releasing the common process layer to allow for discovery, design and management of processes that include core (ERP) processes, human tasks in BPM, and service orchestration. This, in my opinion, is the make-or-break feature for NetWeaver BPM: if they can’t show much deeper integration with their ERP suite than any other BPMS vendor can offer, then they’re just another behind-the-curve BPMS struggling for market share. If they do this right, they will be positioned to win deals against other BPMS vendors that target SAP customers, as well as having a pre-existing relationship with SAP customers who may not yet have considered BPM.

Also in 2009, expect to see convergence of their BPM and BI, which is badly needed in order to provide dashboard/monitoring capabilities for BPM.

Further out, they’re planning to introduce a UI generator that will create a simple forms-based UI for tasks based on the process context (data model), as well as reports generated from the process definition and point-and-click integration of analytics at process steps. There will be more robust event provisioning tied to the existing event structure in the ERP layer, allowing events to be propagated to external applications such as BPM, and intermediate message events integrated with Business Suite. As mentioned previously, there will be new perspectives in the Process Composer, initially a business analyst perspective with a different focus than the existing technical perspective, not just a dumbed-down version as I’ve seen in other tools, and eventually they’ll use the Eclipse rich client platform (RCP) for an even lighter weight (and less geeky) Eclipse interface. There are plans for allowing ad hoc collaboration at a process step – necessary for case management functionality – as well as allowing operation managers to have control over interactive rule thresholds, providing greater business control over processes once they are in operation.

There’s a lot still missing in this first version; : simulation, KPIs, asynchronous web services calls just to name a few. That doesn’t mean, however, that it’s not usable – I know many customers using BPMS’ that do support those functions, but the customers never use them: great demo and sales tools, but not always used in reality.

NetWeaver BPM is not the best BPMS on the market. However, they don’t need to be the best BPMS on the market: they need to be the best BPMS for SAP customers. They’re not quite there yet, but it’s an achievable goal.

2009 08 20 Sandy Kemsley BPM
BRM Comments (7)

Heather Kreger, IBM, on SOA standards

It’s impossible for me to pass up a standards discussion (how sad is that?), so I switched from the business analysis stream to the SOA stream for Heather Kreger’s discussion of SOA standards at an architectural level. OASIS, the Open Group and OMG got together to talk about some of the overlapping standards impacting this: they branded the process as “SOA harmonization” and even wrote a paper about it, Navigating the SOA Open Standards Landscape Around Architecture (direct PDF link).

As Kreger points out, there are differences between the different groups’ standards, but they’re not fatal. For example, both the Open Group and OASIS have SOA reference architectures; the Open Group one is more about implementation, but there’s nothing that’s completely contradictory about them. Similarly, there are SOA governance standards from both the Open Group and OASIS

They created a continuum of reference architectures, from the most abstract conceptual SOA reference architectures through generic reference architectures to SOA solution architectures.

The biggest difference in the standards is that of viewpoint: the standards are written based on what the author organizations are trying to do with them, but contain a lot of common concepts. For example, the Open Group tends to focus on how you build something within your own organization, whereas OASIS looks more at cross-organization orchestration. In some cases, specifications can be complementary (not complimentary as stated in the presentation ), as we see with SoaML being used with any of the reference architectures.

Good summary, and I’ll take time to review the paper later.

2009 07 21 Sandy Kemsley BPM standards
EA
Open Group Comments (1)

The Open Group Conference

I was already planning to attend the Open Group Conference in Toronto next week to catch up on what’s happening in the enterprise architecture space, and now I’ve been invited to join Dana Gardner’s panel on Monday morning, which will also be recorded as a podcast. The panel is on architecture’s scope extending beyond the enterprise, bringing together elements of EA, SOA and cloud computing. Also on the panel will be John Gotze, president of the Associate of Enterprise Architects, and Tim Westbrock from EAdirections. There are big aspects of business process to this, specifically when you start orchestrating processes that span the enterprise’s boundaries, and I hope that we’ll have time to explore that.

I’ll probably be at the conference each day to check out some of the other sessions, and I may stick around for some of the evening events, such as CloudCamp on Wednesday. If you’re there, drop by my session and say hi.

2009 07 17 Sandy Kemsley EA
cloud Comments (0)

Pragmatic BPM and SOA webinar

The webinar that I recorded a couple of weeks back for SearchSOA has finally made it online, although I can only find it on the SearchCIO site. You have to register for the site in order to listen to it, although it’s free.

2009 03 23 Sandy Kemsley BPM Comments (0)

Webinars and podcasts

This seems to be my month for webinars and podcasts. Here’s the line-up:

I recorded a webinar for SearchSOA a few weeks ago on a pragmatic approach to using SOA and BPM together, particularly in the area of service discovery and specification. Unfortunately, I can’t find it on their site, so not sure if it’s been published yet. Keep looking.
On March 18th, I’ll be doing a live webinar on BPM centers of excellence that will become part of the Appian-sponsored BPM Basics informational site. You can sign up for the webinar here if you want to listen to it live, which will include Q&A from the audience; the version without Q&A will be available for replay on the BPM Basics site.
That same week, I’ll be recording a podcast with Savvion’s Dr. Ketabchi on BPM in a down economy. There have been a few other webinars on this topic lately, but right now it’s a very popular message and there’s lots to talk about. This will be published on ITO America, which provides broad coverage of technology issues for higher-level technical management.

The fun part of these three is that not only are they three completely different topics, they’re targeted at three different audiences: the first for developers and other technical people, the second for business and mid-level project team members, and the third at CIOs. Although doing webinars and white papers is a small part of my business, the research, analysis and writing that goes into them really helps to hone my ideas for applicability with my enterprise clients who are implementing BPM.

2009 03 06 Sandy Kemsley BPM Comments (0)

Innovation World: Day 1 keynote

Karl-Heinz Streibich gave the opening keynote; some of the same messages as his address at the media and analyst forum yesterday, plus some messaging about how SOA is a paradigm shift, and the Net generation entering the workforce is a strong driver for modernization and integration. He made the point that process innovation outranks product innovation, and how BPM is the future: companies who apply process management are more agile (hence more competitive) and have more optimized processes (hence are more efficient).

He finished up by referring to their new process frameworks — vertical templates including process models, user interfaces, business rules and KPI definitions for building solutions quickly — but didn’t elaborate; however, they issued a press release today discussing them in more detail. Software AG is providing the following process frameworks:

Order-to-Cash
Procure-to-Pay
Payments
Underwriting
Product Recall
New Product Introduction
Monthly/Quarterly Closing
Employee On-boarding

In addition, their partners are providing frameworks for automotive claims management, electronic check processing, claims management, field services jeopardy management, invoice dispute management, and telecommunications service provisioning.

We also heard from Dr. Peter Kürpick, with much the same message as he delivered yesterday at the analyst forum, although I think that the animated graphics in his slides were nicer today, especially the one showing a process orchestrated across several organizations participating in the end-to-end supply chain. He talked about some of their customers and how they’re improving their business processes using SOA and BPM.

I believe that he had an error in one of his slides, however: in showing how Software AG is a leader in several categories of the analyst reports, and their competitors are not, he showed that they’re a leader in Forrester’s human-centric wave but TIBCO is not. I’m looking at the the wave report right now, and TIBCO actually places higher in the leader category than Software AG. I may have mis-read the slide, it went by quickly and I didn’t have a chance to snap a photo.

He finished up highlighting some of the things coming out of their research lab that will be seen at the demo jam competition today, including BAM capabilities that can be viewed on the iPhone.

2008 11 04 Sandy Kemsley InnovationWorld Comments (5)

Innovation World: Media and Analyst Forum

I’m spending the morning at the media and analyst forum at Software AG’s user conference, Innovation World, in Miami. The first half of the morning covered mainframe modernization, plus a presentation by Miko Matsumura (who I met last week at the Business Rules Forum), Deputy CTO, on the state of SOA adoption. He’s just published a book — more of a booklet at 86 pages — on SOA Adoption for Dummies, continuing Software AG’s trend of using the Dummies brand to push out lengthy white papers in a lighthearted format. For example, chapter 10 is SOA Rocket Science, which covers three principles of SOA:

Keep the pointy end up (instrumentation)
Maintain upward momentum (organization)
Don’t stop until you’re in orbit (automation)

He finished up with a discourse on SOA as IT postmodernism, casting postmodernism as an architectural pattern language: given a breakdown in the dominant metanarrative and the push towards deconstructionism, a paradigm of composition emerges…

After the break, we heard from Ian Walsh from webMethods product marketing to give us an overview of the webMethods suite:

webMethods BPM, including process management, rules management and simulation
CAF (composite application framework), for codeless application design and web-based composite applications
BAM, including process monitoring and alerting, and predictive management

He stated that the “pure-play” BPMS vendors (mentioning Lombardi, Savvion and Pega) are having problems because they sold on the ability to allow the business to create their own processes quickly, but that doesn’t work in reality when you have to integrate complex systems. He also said that the platform vendors (Microsoft, IBM, Oracle) have confusing offerings that are not well integrated, hence take too long to implement. He mentioned TIBCO as a special case, neither pure-play nor platform, but sees their weakness as being very focused on events: good for their CEP strategy, but not good for their broader BPM/SOA strategy.

Walsh sees their strengths in both BPM and SOA as their differentiator: customers are buying both their BPM and SOA products together, not individually.

Bruce Williams was up next speaking on the BPM as the killer application for SOA. He’s a Six Sigma guy, so spent some time talking about BPM in the context of quality management initiatives: if we can manage processes well, we can achieve our business goals; in order to manage processes, we need some systems and infrastructure. He defines the killer app as being flexible and dynamic, not a fixed state or a system with unchangeable functionality. He sees BPM as being the language that can be spoken and understood by both business and IT: not the Tower of Babel created by technology-speak, but how process ties to business strategy.

Logistics are not great: they’ve billeted me in the down-market Marriott Courtyard next door rather than at the Hyatt where the conference is being held (I had to change rooms due to no hot water, can’t run the a/c at night because of the noise, and I have a view — complete with sound effects — of the I95 onramp), and there’s no wifi or power in the lecture hall. There’s supposed to be wifi, but it’s a hidden, protected network that only some people seem to be able to connect to (yes, I added it manually to my wireless network settings). They’ve promised us power at the desks and some assistance with the wifi after lunch.

In case my policy about vendor conferences isn’t crystal clear from previous posts, Software AG is paying my travel expenses to be here, although they are not compensating me for my time nor do they have any editorial control over what I write.

2008 11 03 Sandy Kemsley BPM
InnovationWorld Comments (2)

Business Rules Forum: Pedram Abrari on MDA, SOA and rules

Pedram Abrari, founder and CTO of Corticon, did a breakout session on model-driven architecture, SOA, and the role that rules play in all of this. I’m also in the only room in conference center that’s close enough to the lobby to pick up the hotel wifi, and I found an electrical outlet, so I’m in blogger heaven.

It’s a day for analogies, and Abrari uses the analogy of car for a business application: the driver representing business, and the mechanic representing IT. A driver needs to have control over where he’s going and how he gets there, but doesn’t need to understand the details of how the car works. The mechanic, on the other hand, doesn’t need to understand where the driver is going, but keeps the car and its controls in good working order. Think of the shift from procedural to declarative development concepts, where we’ve moved from stating how to do something, to what needs to be done. A simple example: the difference between writing code to sum a series of numbers, and just selecting a range of cells in Excel and selecting the SUM function.

The utopia of model-driven architecture (MDA) is that business applications are modeled, not programmed; they’re abstract yet comprehensive, directly executable (or at least deployable to an execution environment without programming), the monitoring and analytics are tied directly to the model, and optimization is done directly on the model. The lack of programming required for creating an executable model is critical for keeping the development in the model, and not having it get sucked down into the morass of coding that often happens in environments that are round-trippable in theory, but end up with too much IT tweaking in the execution environment to ever return to the modeling environment.

He then moved on to define SOA: the concept of reusable software components that can be loosely coupled, and use a standard interface to allow for platform neutrality and design by contract. Compound/complex services can be built by assembling lower-level services in an orchestration, usually with BPM.

The key message here is that MDA and SOA fit together perfectly, as most of us are aware: those services that you create as part of your SOA initiative can be assembled directly by your modeling environment, since there is a standard interface for doing so, and services provide functionality without having to know how (or even where) that function is executed. When your MDA environment is a BPMS, this is a crystal-clear connection: every BPMS provides easy ways to interrogate and integrate web services directly into a process as a process step.

From all of this, it’s a simple step to see that a BRMS can provide rules/decisioning services directly to a process; essentially the same message that I discussed yesterday in my presentation, where decision services are no different than any other type of web services that you would call from a BPMS. Abrari stated, however, that the focus should not be on the rules themselves, but on the decision service that’s provided, where a decision is made up of a complete and consistent set of rules that addresses a specific business decision, within a reasonable timeframe, and with a full audit log of the rules fired to reach a specific decision in order to show the decision justification. The underlying rule set must be declarative to make it accessible to business people.

He ended up with a discussion of the necessity to extract rules out of your legacy systems and put them into a central rules repository, and a summary of the model-driven service-oriented world:

Applications are modeled rather than coded
Legacy applications are also available as web services
Business systems are agile and transparent
Enterprise knowledge assets (data, decisions, processes) are stored in a central repository
Management has full visibility into the past, present and future of the business
Enterprises are no longer held hostage by the inability of their systems to keep up with the business

Although the bits on MDA and SOA might have been new to some of the attendees, some of the rules content may have been a bit too basic for this audience, and/or already covered in the general keynotes. However, Abrari is trying to make that strong connection between MDA and rules for model-driven rules development, which is the approach that Corticon takes with their product.

2008 10 29 Sandy Kemsley BRM
BusinessRulesForum
Software design Comments (2)

Ultimus: V8 technical demo

I ended up wrapped up in a discussion at the break that had me arrive late to the last session of the day; Steve Jones of Ultimus is going through many of the technical underpinnings of V8 for designers and developers, particularly those that are relevant to the people in the audience who will be upgrading from those old V7 systems soon.

A nice way to integrate with web services, where the WSDL can be interrogated and a data structure matching the interface parameters created directly from that; most other systems that I’ve seen require that you define the process parameters explicitly then map from one to the other. Of course, there’s lots of cases when you don’t want a full representation of the web services interface, or you want to filter or combine parameters during interface, but this gives you the option for setting up a lot of web services really quickly.

The integrated rules editor allows you to drag and drop process variables — including recipients — onto a graphical decision tree; you don’t have the full power of a business rules system, but this may be enough for a lot of human-centric processes where most of the complex decisions in the process are made by people rather than the system.

For interfacing with any of the external components, such as the email connector or a form, it’s possible to drag and drop data fields from the process instance schema or org chart/ActiveDirectory directly to assign variables for that component, which is a pretty intuitive way to make the link between the data sources and the external calls. They’ve also eliminated some of the coding required for things like getting the current user’s supervisor’s email address, which used to require a bit of code in V7.

Ultimus provides a virtual machine with the software pre-installed as part of their training offerings, which is a great way to learn how to work with all of this; I don’t understand why more vendors don’t provide this to their customers.

I looked back to some old notes from early 2007 when I had a demo of Ultimus V7; my impression at that time is that it was very code-like, with very little functionality that was appropriate for business analysts; V8 looks like a significant improvement over this. They’re still behind the curve relative to many of their competitors, but that’s not completely surprising considering their management upheavals over the past year. If you’re a pure Microsoft shop, however, you’ll likely be willing to overlook some of those issues; Forrester placed Ultimus in the leaders sector (in an admittedly small field) in their report on human-centric BPM on Microsoft platforms. In the broader market of all BPM vendors, Gartner placed them in the visionaries quadrant: good completeness of vision, but not quite enough ability to execute to make it into the leaders quadrant, although this latter assessment seemed to be based on the performance of the previous management team.

Steve spent a bit of time showing the V8 end-user interface: reconfigurable columns in task lists, including queries and filters; shared views to allow a personal view to be shared with another user (and allow that other user to complete work on your behalf); and the ability to run reports directly out of the standard user environment, not a separate interface.

They’ve also done some performance improvements, such as moving completed process instances to a separate set of tables (or even archived out to another database) for historical reporting without impacting the performance of work in progress.

That’s it for me for the conference (and the week); tonight, we’ll be down by the Riverwalk drinking margaritas while listening to a Mariachi band. Tomorrow is an Ultimus partner day and I’ll be on an early morning flight home. Next week, I’ll be at the Business Rules Forum in Orlando, where I’m giving a presentation on mixing rules and process. The following week, I’m headed to Miami for the Software AG analyst/blogger roundtable and a day at their user conference, a late addition to my schedule.

2008 10 23 Sandy Kemsley BPM
BRM
Ultimus Comments (1)

Ultimus: V8 Technical Deep Dive

Chris Adams is back for a somewhat longer session — I think that he zipped through the previous overview session in about 5 minutes to make up time on the schedule — to give us a lot more detail on the V8 product features. Some of this will only be of interest to Ultimus customers, but I find that it gives some good insight into how the product works and the directions that they’re taking.

First, he discussed what’s already in the released 8.x product:

Flobot connectors are now reusable. “Flobots” are the Ultimus connectors to other systems, with about 10 types available out of the box including web services calls (and I now have a very cool Flobot USB key); previously, you had to reconfigure each connector for every use. For example, for the email connector, you had to set up all parameters for the email connector (ports, authentication, etc.) each place it was used in the process, and change it whenever there was a change to, for example, the recipient. Now, they’ve allows for a reusable connector that has some or all of the parameters predefined to allow that to be more easily used in the process.
XML data storage replaces the V7 spreadsheet data structure that was previously used (which previously limited each data element to 255 characters, a limit that I sense from the audience was a sore point). My first reaction was “you used to keep your process instance data in a spreadsheet?”; sometimes you only find out about weirdnesses in a product when you hear about their upgrade out of that state.
A new Ultimus rules engine replaces event conditions, with a graphical representation of the rules. Rules actions can be related to steps in the process, or call .Net code or web services. Previously, the event conditions were kept in the spreadsheet data structure, and you had to reference the spreadsheet cell address rather than a schema variable name within rules. Now, you can add rules to processes directly in-line using the process parameters in the rule definitions.
Native ActiveDirectory support, so that you can (for example) assign a step to a group that exists in AD. You can still use their org chart functionality to create groups directly in Ultimus.
Attachments to process instances have been moved off the BPM server, and into SharePoint. You can use another content repository, but they do SharePoint out of the box and feel that it’s the best integrated solution.

Coming up in 8.2 in December:

BPMN support, although you can still convert back and forth to the Ultimus shapes if you’re more familiar with them. He showed a screenshot that looked pretty rudimentary, but it’s not released yet so I’ll reserve judgement until I see the final version.
Increased visibility into process incident history, to be able to step through exactly what happened in any particular process instance, including which rules that fired. You can actually playback
Enhanced development environment by adding Ultimus awareness to Microsoft Visual Studio for a single environment.
Fully exposed APIs, that is, access to the same APIs that the out of the box system is built on to allow you to build the same functionality into your own custom applications, with any function that you see in a pop-up menu also available through an API.

He showed us some architecture diagrams showing their new open architecture, including the client services for building custom client applications, BI services for custom reporting applications, and Flobots for external connectors.

2008 10 23 Sandy Kemsley BPM
BRM
Ultimus Comments (0)

Forrester Integration-Centric BPM report available

Forrester has released the 2008 version of their Wave report on integration-centric BPM suites; you can find it on Vitria’s site here (registration required).

I won’t reproduce the chart here since that always seems to get me in trouble, but suffice it say that Software AG (the former webMethods product), IBM (various WebSphere bits), Vitria, TIBCO (ActiveMatrix), Oracle (SOA Suite; the BEA products were not evaluated due to timing relative to the acquisition), SAP (NetWeaver) and Cordys should all be very happy.

2008 10 16 Sandy Kemsley BPM Comments (2)

Business Process Driven SOA using BPMN and BPEL

I just received a review copy of Matjaz Juric and Kapil Pant’s new book, Business Process Driven SOA using BPMN and BPEL. It’s on my list of recent books that I’ve received to review, and I hope to get to it soon.

According to the authors’ description, you’ll learn the following from this book:

Modeling business processes in an SOA-compliant way
A detailed understanding of BPMN standard for business process modeling and analysis
Automatically translating BPMN into BPEL Executing business processes on SOA platforms
Overcome the semantic gap between process models and their execution, and follow the closed-loop business process management life cycle
Understand technologies complementary to BPM and SOA such as Business Rules Management and Business Activity monitoring Approach

I’ll let you know if I learned all of that once I’ve had a chance to read it.

2008 09 20 Sandy Kemsley BPM
BPM standards Comments (1)

Column 2