The second edition of my book BPMN Method and Style is now available. The first edition was published in June 2009, based on the BPMN 2.0 “beta” specification. The new edition is based on the final specification of August 2010, officially adopted by OMG in January 2011. The graphical notation did not change between beta and the final, but there were a number of significant changes in the metamodel, XML schema, and “rules” of the spec. These are all reflected in the new edition.
The first half of the book is addressed to process modelers – business process analysts, business architects, BPM project team members, as well as developers – and focuses on the graphical notation. It explains Method and Style’s driving principle – that the process logic should be described unambiguously, completely, and consistently from the diagram alone – and presents a methodology and style rules to achieve it. It is a complete rewrite of the original edition, mostly based on the experience of delivering BPMN Method and Style training for over two years.
The new edition explains how BPMN concepts like “activity” and “process” are aligned (or not) with those same terms as used in BPM Frameworks and similar aspects of enterprise architecture. It takes advantage of the fact that BPMN style, described in the first edition as “best practices”, is better viewed as a set of rules that can be used to validate models in a BPMN tool. It continues the levels-based approach of the first edition – Level 1 being a basic working set of shapes and symbols familiar to business people from traditional flowcharting, and Level 2 a somewhat expanded set supporting event-triggered behavior – but makes slight adjustments to each level to match the Descriptive (Level 1) and Analytic (Level 2) Process Modeling Conformance subclasses in the final BPMN 2.0 spec. (A few aspects of Method and Style ultimately did make it into the final specification!) Even if you bought the first edition, there is enough new material in the Method and Style section alone to justify getting the new one.
The second half of the book, the BPMN Implementer’s Guide, is brand new. It is primarily addressed to developers and tool vendors, although analysts and architects will benefit from it as well. It explains the BPMN 2.0 metamodel and XML serialization. Based on my look at the first round of so-called BPMN 2.0-based tools, this is something that few BPMN tool vendors understand. The first part of the BPMN Implementer’s Guide focuses on non-executable models using elements of the Analytic subclass (Level 2 palette) only, explaining the meaning and usage of the various XML elements and attributes. It describes the BPMN-I Profile, a set of constraints on the serialization that guarantee that a given BPMN model has a unique XML representation, important to automating model interchange between tools.
The second part discusses executable BPMN 2.0 and the serialization of data mapping, service tasks, human task assignment, and similar details. It presents an example using Bonita Open Solution to illustrate the relationship between executable design in a real BPM Suite and its representation in the BPMN 2.0 export. It concludes with recommendations to implementers for aligning executable BPMN 2.0 with Method and Style. The BPMN Implementer’s Guide is logically a separate book, but a number of its ideas flow naturally from Method and Style, so the simplest thing was to combine them.
Hi, Bruce, you’ve sold me, I presume if I study the heck out of the first edition first it won’t be contradicted by anything in the second, yes?
Phil
Phil,
That’s a loaded question. In first edition I say reserve Error for technical exceptions, but in second edition I say ok for business exceptions also. This is just convention, not rule of the spec. There were a few changes and clarifications to some of the rules of the spec between the versions, and pretty much all the XML details are new in the second edition, as this is what most of the changes between beta and final spec involve. The Method changed a little bit, and more style rules, but most of the first book is still good today.
I ordered yesterday at amazon.de, it will be unfortunately
delivered in 5-8 weeks. I hope that it goes faster.
Dear Bruce: I’d like to take one of your lessons on line… what should I do?
Jimena,
Live-online Click here for more info on Nov 28-30 class. Click here for info on 2012 classes or to register for January class. For self-paced online/on-demand training, go to http://www.bpmessentials.com.
Dear Bruce,
It would be great if this book would be available on iBooks store! (or in Kindle format would be ok too). Please consider e-book format for your books:)
Ludmila
Hi Bruce!
The Ludmila’s suggestion is intesting.
I also would like a digital version of the book.
I am Brazilian and the costs of purchasing the book are prohibitive…
Please publish this as an iBook or Kindle book so I can add this to my iPad library.
I got the book at the first january week. For me as non native reader/speaker its very understandable and I use the book in combination with the OMG bpmn specs.
I’m trying to serialize a Visio BPMN Diagram and I am struggle with the BPMNDI part. I don’t know, how can I convert/transform the shapes position data.
In Visio, for example, the x/y coordinates are in inches or mm.
Considering, I have a shape with X: 57 mm, Y 45 mm, Width and Height 10 mm. How schould I represent/convert these data to BPMN coordinate system?
@Tihomir,
Each page in bpmndi has a resolution attribute, which (for some reason) is always specified in pixels per inch. Let’s say you want resolution = 96. If your Visio coordinates are in mm (this will depend on which template you use), then you simply need to do the math: VisioMM x 1 in/25.4 mm x 96 pixels/in.
Of course you also need to convert between Visio coordinate system (origin at bottom left, position of shape center) and BPMN coordinate system (origin at top left, position at shape top left).
Bruce, thanks for the advice with the resolution attibute. But for the coordinate system convertion(actually I only need to convert the Y values) I have not an clue point since the Y values go to infinity.
Maybe I could first determine the max Y value from all shapes: Y_max=Max(Y1, Y2, Y3,…., Yn)
Then for each new Y value I could say Yn = Y_max – Yn
Ok, but then the value for my highest shape would be e.g. Yn= 100 – 100
I don’t understand your issue. I created xslt to map Visio Premium vdx to BPMN 2.0 myself. If that is what you are doing, you do need to map the x coordinates because Visio reports the center of the shape and BPMN references top left corner. On y, don’t worry that BPMN has no page size. It’s the Visio page size you need. The vdx gives distance from page bottom, so distance from page top is PageSize – y[reported].
Oh, sorry I didn’t bear in mind, that I only need the VisioPageSize and not BPMNPageSize. For X and Shape’s PinX its clear. Cheers!
Tihomir, Good luck with your project. I found it a real challenge to deal with the complexities of extracting BPMN from the vdx, and I posted about some of my methods. I will be interested to compare your result with my tool (xslt 2.0).
Yes, for me too. Actually for me is also a big challenge since I am new in BPMN and programming
I work programmatically with an Visio drawing (and visio object model), not with vdx.
Hello all & Bruce – I am new to BPMN 2.0 and was hoping to find a low-cost or free “basic” reference guide (preferrablly an e-book). I appreciate any recommendations. Bruce – I plan on purchasing your book, however I need to get OK from my manager. Before doing so, are there any plans of publishing it electronically? If so, I would like to wait for the e-version. Thank you – Nicole
It is now available on Amazon Kindle $9.99.
Hi Bruce.
Great book. I am reading the 1srt edition and I’m learning a lot.
I’m trying to istantiate the OMG metamodel of BPMN2.0 with some of your examples.
There is one thing that puzzled me. In your examples you have message flows coming in and going out from collapsed subprocesses. However in the OMG standard I saw that message flows only can be exchanged among instances of InteractinNode (IN). As subclasses of IN we have Event, Task and ConversationNode. So apparently an instance of Subprocess cannot receive/send a message.
Is this a bug in the spec or I’m missing something?
Anacieto,
Good catch. From my perspective it is a bug in the metamodel. Note that the spec also says that subprocess may not have dataInput or dataOutput, which I also consider a bug. Possibly from an executable process standpoint, the intent is that the message (and data flow) is sent/received by a task or event only. If so, in that case the visual representation of message/data flow to/from subprocess must either be suppressed altogether – not good from method and style perspective – or should be serialized differently, i.e. source/target of message/data flow is not the subprocess but the child level node. I think the latter is too difficult to implement, and if it serves any purpose at all it is just for the benefit of BPMN 2.0 execution engines… so I choose to disregard that part of the metamodel. To me it is a bug. I would not expect OMG to change it, either.
Thank you for answer.
In page 42 of the 1st edition of your book you said:
“…in BPMN 2.0 it is illegal to draw a pool unless the diagram contains more than one.”
However in the spec, pag 113, refer that “…a Process that represents the work performed from the point of view of the modeler or the modeler’s organization can be considered “internal” and is NOT REQUIRED to be surrounded by the boundary of the Pool, while the other Pools in the Diagram MUST have their boundary (see Figure 9.5).”
Are you referring only to specific case of subprocesses?
1st edition was based on the beta spec 2009., which had minOccurs=2 for the participant element in schema (participant is the semantic counterpart of pool shape). In the final spec they removed that constraint – since it made most BPMN 1.x diagrams schema-invalid. Your p113 quote is still valid in the final spec, although I should be more precise in the terminology. It says only one process in a collaboration diagram may be unenclosed.by a pool shape. In BPMN 1.x we said it had an invisible pool, but in BPMN 2.0 it has no pool. The participant element is still there (not actually required by schema) but the bpmndi shape pointing to it on that page is not present. The minOccurs issue re participant and the process-without-pool-shape issues are completely separate.
Thanks Bruce. It’s nice to know the story behind the covers.
Hello Bruce and all,
I bought the electronic version from Amazon three days ago (February 20). I really like the textual content, but I am a bit disappointed by the diagrams: the resolution is just too low, so that e.g. task type icons are blurred and some labels are barely readable. Too bad because the text is really great. Any hope to get an “upgrade” with higher-resolution diagrams?
Best regards
Jérôme
Jerome,
Sorry but impossible with the Kindle format. I spent a lot of time trying to get more resolution, but the Kindle format conversion always limits the pixel count. I haven’t seen other Kindle books with anything better on the graphics side.
Hi Bruce
I was trying to create the instances of the metamodel for a little example of Collaboration between 2 pools/participants (each of one with one process).
However I was obliged to create also an instance of ConversationAssociation (because there is an association 1 to 1 between Collaboration and ConversationAssociation. See Figure 9.1 page 109 of spec). Do you think it makes sense when we are creating a collaboration diagram to have to create instances that seems to be related with other kind of diagram (Conversation)?
Regards
I can’t speak to the UML diagram, but in the XSD all references to conversation-related elements from collaboration are optional. So you are not required to use them. The metamodel diagram may be in error. Conversation diagrams are mostly for showing, at a high level, the interactions among a network of processes or business entities. Unfortunately for executable BPMN, they stuck the whole key-based message correlation mechanism inside this special-purpose model that nobody uses.
Hi Bruce
I saw the online workshop you gave for Bonitasoft and downloaded the software.
I’m familiar with BPM but not quite well with the graphic notation.
I wat to start experimenting right away with Bonitasoft and see where I run into problems.
My goal is to have a working Process, to simulate and automate the process. More as a pilot to see if Bonitasoft is the right tool.
What do you recommend to take this agile experimental path?
Later on I might take a refreshment course.
Hello Bruce,
thanks a lot for publishing your great book also as ebook-version at amazon!
I have got a wish for the next edition of the ebook: Could you provide also the table of contents, so that it is possible to navigate to the chapters by using the “favorite” button (the button is shown as an “open book”. You may find it in the middle of the toolbar of the kindle-app for iPhone/iPad) At the moment it offers only navigation to title-page, start-page and position …
I found your workaround on the on the “Table of contents” page with the links … well it helps to get around. Anyway would be great to use it also in the “standard” way too
Thanks a lot in advance
hello
Here in Iran based on political problems with Economic sanctions i can’t buy this book by internet would you please guide me how can i access this book?
Sorry, I am at a loss. I don’t know.
Bruce,
Does your 2nd edition book has a spanish version?
Thanks a lot!
Working on it currently.
Thanks, our team really will benefit from it!!
Please keep us posted!
Is there a spanish version on the first edition?
[...] it is necessary to nail down its scope, that is, “where it starts and ends,” according to Bruce Silver in BPMN Method Style. “You don’t want your modeling efforts, after weeks of interviewing [...]
Dear Bruce
I’m interested in buying your book BPMN Method Style but I would like to have it on my iPad rather than having the physical book. Can you let me know how should I proceed?
Thanks a lot in advance
Kind regards
Julio
Julio,
There is a Kindle version $9.99 on Amazon. Use the free Kindle app on ipad to view. The graphics are not as good as the printed version, but adequate I think. This is a limitation of the Kindle publishing software.
Thanks a million Bruce!
[...] the notation. If you want to go into BPMN in a little more depth, I can highly recommend the book BPMN Method and Style by Bruce [...]
Hi Bruce,
I’m interested in a digital copy too, but I don’t use Kindle. I guess I could register with Kindle, but I’m not highly motivated considering the graphics issue. Would you consider publishing in Google Books? I don’t know how much effort this would be, nor if this would offer any better graphics, but it is where I’d prefer to purchase if possible.
Cheers.
Kindle has free apps for iPad, PC too. You don’t need a Kindle device.