Attached to this post is Upside Research’s implementation study of ActiveVOS at a North American government security agency. While the agency is not named for security reasons, the agency’s staff participated in the gathering of data for this report. The report documents the power of BPM to automate and enhance core processes — in this case, classification of information coming into and going out of the agency.
BPEL is complex and BPMN is simple, right? After all, BPMN has a nice graphical notation. The BPEL standard only specifies what the language looks like in XML. That alone ought to be enough claim the prize for BPMN.
However, what if you use BPMN’s notation for a process but use BPEL for the executable representation? This removes the graphical vs. XML distinction and can “hide” the non-graphical BPEL as represented in XML. You end up with a BPMN model everyone can understand and a BPEL model your computers can execute. It’s like the two sides of a coin: there are different pictures on each side, but the coin itself is always both sides at once.
However the question of which is simpler gets more complicated when you also consider that the new BPMN 2.0 specification includes hundreds of constructs in its meta-model that have no graphical representation. Now, which is simpler, BPMN with BPEL or BPMN with the new BPMN 2.0 execution language? What may seem obvious (BPMN with BPMN 2.0 execution) isn’t the slam-dunk choice many people might expect it to be.
BPMN 2.0 has two different — but equal — compliance points for execution: BPEL Process Execution Conformance and Process Execution Conformance. This means that BPMN 2.0 standardizes the use of BPEL as the execution language for BPMN, but it also offers the option of making BPMN executable by using new constructs that have been added to the BPMN notation specifically to support execution. These new constructs depend on the execution semantics that have been defined for almost everything in BPMN.
So, which is simpler? Believe it or not, using BPMN with BPEL execution is dramatically simpler than trying to execute processes using the new BPMN 2.0 execution language. I know this sounds counter-intuitive, so I will justify it in this post and a series of follow-up posts on the same subject.
Before I get into the details of why I believe BPMN with BPEL is better, a little history might help clarify the question. There are some factors that caused the BPMN 2.0 standard to eventually become more complex than BPEL. (I know, I know, BPEL has the reputation of being far too complex…but hear me out.)
BPMN was designed to be a language for communicating from one person to another, not from a person to a machine. Languages used for human communication have a natural, and appropriate, tendancy to grow. Whenever people find that they frequently need to convey something that is awkward to express with their current vocabulary, they invent a new word. English, which is especially amenable to such growth, surpassed one million words last year. Just consider “unfriend” or “netbook,” new words to express new ideas.
The same is true for graphical modeling languages. Look at UML (Universal Modeling Language). It started as the unification of three fairly simple graphical notations (best known by their respective primary inventors: Rumbaugh, Coad & Yourdan, and Grady Booch). Once they unified their modeling languages and people started using them in earnest, they grew larger and larger, with new diagrams and new elements on those diagrams with each successive version. Sure there was always overlap in what could be expressed by different diagrams or different elements, but in each case, there were situations where one was more natural to the reader than the other. The fact that different constructs have imprecise overlapping meanings is of little concern in a language meant for people, since people are comfortable with choosing among a variety of ways of expressing the same thing, each with their own nuances and connotations.
But while notation creep is a useful way of expanding spoken languages or graphical notations, it is not such a good thing for a language that must be directly executable on a computer.
That’s because it is always a problem to take such a large language and give it formal executable semantics. The problem usually isn’t with a lack of rigor in the definition of any one construct. The problem is with the exponential number of combinations of those constructs.
Good programming languages typically add new fundamental primitives very cautiously. Consider how much hard preparatory work was done in the Java community before Java introduced generics into the language, or the hand wringing that is gripping that community as they grapple with the addition of closures to the language. The way it typically works is that some eminently-respectable, highly-credentialed expert (like Neal Gafter, in the case of closures) will make a seemingly very well-thought-out proposal that describes how the new construct will simplify the lives of so many programmers. Then another equally eminent expert (like Josh Bloch, in this case) will find unintended consequences of the new construct when it is used in combination with other things in the language.
That was just for one language feature. The BPMN 2.0 execution language has dozens of features that have never really been used together in an execution language. For example, the BPMN 2.0 execution not only has a variety of ways of handing the control flow for multiple incoming sequence flows, activities also can’t execute until all of the required inputs from one of the activities input datasets has become available. In other words, it has a fairly complex data flow model intertwined with its control flow model.
Another example is message correlation. BPEL has, in the past, been criticized for the complexity of its approach to correlation, but BPMN has two different correlation mechanisms. Key-based correlation is basically equivalent to BPEL’s correlation mechanism, although the standard has invented all new terminology for the various components. It then defines a new concept of context-based correlation. Rather than trying to convince you that it is complex, I’ll just include the complete explanation of it from the BPMN 2.0 specification (yes, in a 500-page specification, there are no examples or additional explanations for these concepts):
In context-based correlation, the Process context (i.e., its Data Objects and Properties) may dynamically influence the matching criterion. That is, a CorrelationKey may be complemented by a Process-specific CorrelationSubscription. A CorrelationSubscription aggregates as many CorrelationProperty-Bindings as there are CorrelationProperties in the CorrelationKey. A CorrelationPropertyBinding relates to a specific CorrelationProperty and also links to a Formal-Expression which denotes a dynamic extraction rule atop the Process context. At runtime, the Correlation-Key instance for a particular Conversation is populated (and dynamically updated) from the Process context using these FormalExpressions. In that sense, changes in the Process context may alter the correlation condition.
Confused yet? Are you wondering not just why BPMN 2.0 needed to define and redefine an important concept like message correlation, but also wondering how, precisely, to implement BPMN correlation?
These are just a couple of the ways that BPMN’s new execution language is more complex that using BPMN with BPEL. BPEL is now a known commodity. It’s widely implemented. Many production applications are running BPEL today. There are many people with experience with it and the concepts in the language are well understood. With BPMN 2.0, it now has a standardized notation, so there is no need to work with a new language that is a big bag of language constructs whose interactions have never been exercised together.
In episode #5 of our continuing webinar series on technical topics of interest to developers, architects and business analysts working with SOA-based business process management systems (BPMS), Dr. Michael Rowley, CTO, Active Endpoints compares and contrasts two different styles of message correlation. In episode #4, Michael outlined message correlation as defined by the BPEL standard. In this episode, Michael illustrates a different style of correlation, which relies on the execution engine to correlate incoming messages to specific processes. Michael also describes when and how each style (BPEL-managed vs. engine-managed) can be used and notes some pros and cons for each style.
There are two attached versions of the webinar replay (an iPod-formatted .m4v and a DivX-encoded .avi). As always, you can register for the next episode of CTO Tuesdays at http://www.activevos.com/ctot. We look forward to your comments, suggestions and feedback.
I have good news and bad news. The good news is that we (finally) have replays of episode #4 of CTO Tuesdays, our regular weekly webinar on BPM topics of interest to process designers and developers. The subject of this webinar is message correlation, an interesting topic that details how systems match up running processes and the messages for those running processes.
The bad news is that due to a technical issue, the audio for the host, our own Sonal Rajan, wasn’t recorded. This is shame because at the end of each topic, we always have an open Q&A session on the current topic to amplify the technical discussion. Unfortunately, these replays won’t have that Q&A because there’s no audio for the moderator. However, the actual presentation about message correlation was recorded just fine.
In the two attached versions of the webinar replay (an iPod-formatted .m4v and a DivX-encoded .avi), I have edited most of the silent introduction and the Q&A.
As always, you can register for the next episode of CTO Tuesdays at http://www.activevos.com/ctot.
In the latest episode of Dana Gardner’s BriefingsDirect Analyst Insight series, Dana covers the concept of business commerce clouds. Panelists commenting on this topic include: Tony Baer of Ovum, Brad Shimmin of Current Analysis, Jason Bloomberg of ZapThink and independent IT analysts Sandy Kemsley and JP Morgenthal. Have a listen to this podcast for these experts’ perspectives on SaaS, SOA, BPM, reliability, security and community as related to business commerce clouds.
In addition to the audio file, we have also posted a PDF transcript of the podcast, for your convenience.
BriefingsDirect Analyst Insights Podcas [50:31m]: Play Now | Play in Popup | Download (1870)
BriefingsDirect Analyst Insights Podcast : Download (177)BPMN 2.0 and WS-BPEL 2.0 are the two most important standards for BPM today. But why are there two? Can’t you just care about BPEL or just care about BPMN? In fact, both standards matter and the two should be used together. To back that up, I have to convince you both that BPEL needs BPMN and that BPMN needs BPEL. In today’s post, I’ll concentrate on the first: why BPEL needs BPMN.
First, lets assume that you are convinced of the value of BPEL. You see that it is a great high-level language for creating business processes and orchestrating services. Its service-centric approach is simpler and better for long-term manageability and reuse than other approaches to business process management. It is an accepted OASIS standard with multiple vendor implementations, so investments in BPEL processes are not tied to a single vendor and you can find people who already know the language without having to train them from scratch.
But if you are convinced you want BPEL, why should you care about BPMN? There are two main reasons:
1) To get the value of a standard notation;
2) To improve collaboration with a wide variety of stakeholders in the process, since BPMN is a significant simplification over existing notations used for BPEL.
When WS-BPEL 2.0 was standardized, the OASIS Technical Committee chose not to standardize a graphical notation for it. This was unfortunate, since no one creates a business process by writing BPEL in XML, which is the only standardized representation. Every vendor, and every BPEL developer, creates their processes using a graphical representation, but that representation is different for every tool.
And the notations used by these tools haven’t really been very good. They typically provide a one-to-one correspondence between control flow constructs in BPEL and things on the canvas. However, if you use the BPMN notation, it shows a notation that can mostly be understood without any knowledge of BPEL or even BPMN for that matter (as long as the labels are chosen carefully).
Let me make both of these points with the help of a trivial process example. Take a look at the BPMN representation of a process that I’ll call the “Question” process.
(Click on each image to see a larger version)
![clip_image002[4]](http://www.vosibilities.com/wp-content/uploads/2009/11/clip_image0024.jpg "clip_image002[4]")
It is trivial to follow what is going on, especially if you know the standard notation. You can’t tell by looking at this diagram, but I’ve used two different BPEL mechanisms for getting to the next activity. I use a BPEL link to get from “Receive Q” to the first diamond (the beginning of the BPEL if statement). I use a BPEL sequence to get from the second diamond (the end of the if) to the “Record Answer” activity.
The user who is looking at the graphical representation of the process doesn’t need to know about the distinction between these two mechanisms, so the diagram doesn’t show a difference. The developer may want to know about the difference, so ActiveVOS highlights them differently on mouse-over and shows them differently in the “process outline view”, but that isn’t really important for today’s discussion.
What is important is how different the process is represented in different tools due to the fact that no notation had been standardized. I’ll show what this process looks like in three different BPEL process designers.
Here is how ActiveVOS would represent this process in previous versions of the product (or using the optional “classic” style in 7.0):
![clip_image004[4]](http://www.vosibilities.com/wp-content/uploads/2009/11/clip_image0044.jpg "clip_image004[4]")
Here is how the Eclipse BPEL Designer represents it:
![clip_image006[4]](http://www.vosibilities.com/wp-content/uploads/2009/11/clip_image0064.jpg "clip_image006[4]")
And, here is how the designer for Oracle’s BPEL Process Manager represents it:
![clip_image008[4]](http://www.vosibilities.com/wp-content/uploads/2009/11/clip_image0084.jpg "clip_image008[4]")
In all three of these representations, each of the paths through the if statement are represented by a bounding box. The problem with this representation is that nested if statements can result in so many nested bounding boxes that it is hard to follow what is going on. BPMN simply has arrows through each path and the paths merge back into a single control flow at a gateway diamond.
Also notice the differences in the handling of links vs. sequences. Both ActiveVOS classic and Eclipse represent sequences with their own bounding boxes, then any arrow that is a direct child of a sequence box is known to belong to the sequence, rather than being a real link. Eclipse also draws the links in different color. The extra sequence icon and corresponding bounding box just interferes with the ability for non-technical users to follow what is going on in the process.
Oracle’s designer is odd in this respect. Sequences are not shown in a bounding box, so they don’t clutter up the control flow (a good thing in my opinion), but links aren’t shown at all! There is a link from the “Receive_Q” activity to the if statement, but there isn’t any representation of it on the diagram. It shows the “Receive_Q” and the if as if they happen in parallel. You have to look into the properties of “Receive_Q” to discover that it has an outgoing link, and further rummaging to find out where it goes.
The BPMN representation is, by far, the easiest version of this small process to understand. The process illustrates just three constructs whose representation is simpler with BPMN than with other approaches: ifs, sequences and links. The other BPEL constructs are generally as easy or easier for non-technical users to understand than previous approaches.
But, as valuable as the improvement in readability may be, the greater value that BPMN brings to be BPEL is probably consistency. Having different tools represent similar constructs in such different ways is detrimental to one of the key values in having a standard: skills portability. With a common notation, people will be able to carry their knowledge of how to understand and work with standards-based business processes between vendor tools. It will also create a greater incentive for people to learn these technologies and for schools to teach them. After all, people aren’t usually to thrilled about investing a lot of energy into learning proprietary technologies, and no school really wants to be teaching proprietary technologies.
This week on CTO Tuesdays Active Endpoints CTO Michael Rowley presented how events are represented in BPMN 2.0 and BPEL.
I think you will find Michael’s explanation of BPMN 2.0 event notation especially valuable.
I have attached two versions of the recorded webinar to this post. The first is an iPod-formatted .m4v. Also attached to this post is a Windows Media format .wmv file.
We have also made signing up for CTO Tuesdays and accessing the replays much easier. You can always sign up for the upcoming session of CTO Tuesdays at http://www.activevos.com/ctot. Replays are always available at http://www.ctotuesdays.com. And, an RSS feed of the replays is available at http://www.ctotuesdays.com/feed.
This week’s topic on CTO Tuesdays was an introduction to the new WS-HumanTask standard for workflow. In this informative session, Michael Rowley describes the importance of the new standard for workflow, how it separates tasks from processing and how WS-HumanTask enables human activities to be seen as services in a process application.
Attached to this post are three files. A PDF of the slides Dr. Rowley presented, an iPod-formatted .m4v file (which requires QuickTime or iTunes to be installed) and a more-or-less standard .avi file. The .avi is the larger of the two video files.
Due to a technical error (I didn’t press “show” on GoToMeeting), the first few minutes of the video show Michael’s slides, not the ones I am discussing. Since this is just an introduction, you won’t miss anything. I’ve put those “missing” slides into the .pdf file, so you can follow along if you want to.
We had a very lively panel discussion at the end of the presentation; I hope you’ll have the time to listen to the discussion that follows the presentation.
As always, we are very interested in your feedback, comments and topic suggestions.
One more note: you can always register for the upcoming CTO Tuesdays session by visiting http://www.activevos.com/ctot. We hope you join us for next week’s webinar.
ActiveVOS and SoftConEx have announced that their products have been combined to deliver a revolution in airline ticketing workflow for distributors, business travel service companies and travel portals. Details are in the press release attached to this post.
At Gartner’s BPM Summit in October, ebizQ’s Peter Schooff talked with me (Alex Neihaus) and Active Endpoints CTO Michael Rowley about ActiveVOS 7.0 and its new BPMN 2.0 modeler. A link to the podcast is below and it is included in our podcast feed in the iTunes Store.
eBizQ podcast:How BPMS Delivers Value to Today's Business [4:02m]: Play Now | Play in Popup | Download (134)We are very pleased to post the recording of the first episode of our new weekly webinar on BPM technology called CTO Tuesdays.
Every Tuesday, Active Endpoints’ CTO Michael Rowley, will present a topic of interest to BPM users. Our inaugural topic was an explanation of the meaning and uses of the BPMN 2.0 diamond symbol. If you are interested in learning BPMN 2.0 — or if you just want to brush up on some of the more advanced considerations in using this basic BPMN symbol — you will find this recording very instructive. Concepts are demonstrated in ActiveVOS 7’s new BPMN 2.0 modeler.
Attached to this post are two versions of the webinar: an iPod-formatted .m4v file our podcast subscribers will automatically receive and an H.264-encoded .avi file (which is much larger at about 113MB).
We welcome your input and suggestions for CTO Tuesdays. Contact us via email at editor at activevos dot com. Today, the best way to be notified of upcoming CTO Tuesdays is to be on our mailing list. And, the best way to get onto our mailing list is to download a trial of ActiveVOS. You can also register for upcoming CTO Tuesdays by clicking on the link in the right hand column of any interior page on www.activevos.com.
We are working hard on making registering for CTO Tuesdays easier. But because of the demand for education on topics like BPMN 2.0, we started the webinar series without waiting to dot all the “i’s” and cross all our “t’s.”
Update: You can now register for CTO Tuesdays by clicking the link in the right-hand column of any page on www.activevos.com except the home page. So, just navigate into the site a little and you’ll get a little reward: easy access to registration for CTO Tuesdays.
Updated update: You can now always register for the upcoming CTO Tuesdays at http://www.activevos.com/ctot.
We hope you enjoy this recording and that you will join us as your schedule permits for the live CTO Tuesdays every Tuesday at noon ET, 9am PT, 16:00 GMT (17:00 GMT after the end of US daylight savings time in November, 2009).
CTOT #1: The BPMN diamond [39:32m]: Play Now | Play in Popup | Download (4605)
CTOT #1: The BPMN diamond [39:30m]: Download (7409)
Scary big. That is how big Oracle is in middleware. After bragging that they had beat BEA to be #2 in the middleware market, they bought them. Then, with the acquisition of Sun, Oracle has control over the underlying technologies of Java and Java EE, plus the primary open-source challenge to their database dominance. Of course, IBM has also done its share of gobbling up middleware companies, so between them, the number of products and acquisitions has become overwhelming. We made exactly that point last week at Oracle “Open”World by dressing up actors as prisoners “shackled” to Oracle SOA Suite (Check out the hi-jinks here, here and here.)
Up to now, enterprises have typically had one choice that drove most other technology choices in the data center: .Net or Java. If you chose .Net, then you are a Microsoft shop and you’ve decided that the advantages of living in a single vendor world outweigh the disadvantages of being tied to that one vendor. However, if you chose to go with Java, you probably did so because you wanted to then live in the world of standards-based technologies, where for each technology purchase, you could separately evaluate products from a number of competing vendors.
But now, with much of the Java middleware world being absorbed into one of two vendors, the era of having a choices for each purchase is coming to an end. Instead, there will be just one big decision. Do you want to be a Microsoft shop, an Oracle shop or an IBM shop? All other decisions will flow from that initial decision. This is because the development teams in each of those companies will naturally be forced to give a high priority to getting any new software to work with existing software from the same company. Getting it to work with the other company’s software will be a “goal,” but as someone deeply involved in the development of software products, trust me when I tell you those are the kinds of goals that tend to slip as the ship date of any product nears.
Does it matter? Isn’t three enough? Yes, it matters. And no, three isn’t enough. Actually, the real problem isn’t the small number of choices; it is that the switching costs are just too high. If there were three good choices for each purchase, that wouldn’t be so bad. But when you are virtually locked into a single vendor for each new purchase once you’ve started down the road of buying from them, then it is a real problem. The lack of competition for individual products removes critical competitive pressures from individual product lines, so the products grow to become heavy, badly integrated, expensive beasts.
And, if enterprises have to pay more for lower quality software on their servers, this affects everyone. The ultimate consumer pays for the more expensive software in higher prices, but they pay even more for the lower productivity that comes from software that is hard to use, hard to manage, and unresponsive to the needs of the business.
We get up in the morning and think about what we can do to make ActiveVOS the antithesis of the bloatware from Oracle and IBM.
Well, it’s the day after our big push to liberate Oracle SOA Suite 11g users during Oracle “Open”World in San Francisco.
And I am almost at a loss of words to describe our experience and the effect we seemed to have had. But, I gotta try. Here are some semi-random comments.
We are extraordinarily grateful for IDG News reporter Chris Kanaracus’s story, which perfectly captured the reasons we stood on a street corner for two days to make sure people understand that alternatives to high costs and lock-in exist.
Our social media meetup was a great success…and a lot of fun. We shared photos and videos of the event. (Here’s a video of the main reason for the party. 
) I kid you not, the coolest people are the people who you befriend first online and then have the pleasure of meeting in the real world.
On Tuesday, October 14, we were hit with a rain storm that dumped a month of rain on San Francisco in about six hours. In spite of the high winds and Biblical downpours, we persisted in our mission of liberation from Oracle SOA Suite 11g.
You can check out videos on our YouTube channel (you have to see…and I mean you really have to see — the video titled “In the rain”), see stills in our Flickr photostream and, for our podcast subscribers, I’ve enclosed a short iPod-formatted video in the RSS feed. There’s also an HD-version of the video, for those that want to “be there” with us. Both are attached to this post.
Finally, you might find Otis Maxwell’s post about our attack on SOA Suite interesting. Otis’s description of how he found our meetup is very interesting. He calls what we did “suitcasing.” I think it’s simpler: we poked Oracle in the eye…and people loved it.
In case you are one of the folks who didn’t get the cards we handed out with the 11 things to consider before using SOA Suite 11g, here’s an image:
As you can imagine, pulling something like this off takes planning and dedication. I want to thank Sonal Rajan and Leslie Minasian, both of Active Endpoints, and Pat Flanders for their hard work and dedication.
Here are some of the people who “posed with our prisoners” today at Oracle OpenWorld. People were laughing and enjoying our street theater — and “getting” the message of ease of use and freedom from lock-in that ActiveVOS offers. We had a blast today…and we hope you did, too.
Tomorrow (Tuesday, October 13) at 4pm, we are hosting a meetup to pick three iPod nano winners from among all of the posted photos and videos. Lots of people were creating images today…remember to post them anywhere that’s convenient and then join us tomorrow for the meetup. It’s at the Thirsty Bear, 661 Howard Street. That’s just a block away from Moscone…down the street from the W Hotel.
(click on the image for a larger picture)
Today, we are at Oracle OpenWorld in San Francisco. We’ve dressed up actors as prisoners of SOA Suite 11g. We’ve handing out cards that contain the 11 things to think about when considering SOA Suite 11g and an invite to our social media meetup Tuesday, October 13. Here are the details.
We’re having fun. In the video below, you can see that people are enjoying the humor and taking the time to read our “11 things to consider before using Oracle SOA Suite 11g.” They’re also posing with our prisoners and shooting video and photos. So, we expect a great turnout tomorrow at the meetup. Remember: you can win an iPod nano for the best photo/video of the prisoners at the meeting. The judges are the audience. So, join us for some drinks, snacks and fun.
Check out some of the video and photos below.
First, we “cowboy up” at a Starbucks, then hit the streets before dawn to get a head start on liberation from SOA Suite 11g. Reactions are positive…from the woman who stands on the street corner reading the 11 things to consider about SOA Suite to the person complimenting our costumes:
The look of relief on a newly-liberated attendee:
After release from 11g prison, it’s all smiles on the streets of San Francisco:
