Scott Francis has written an excellent post about the core message of our presentation at bpmNEXT. He called it the "Zero Code Hypothesis". Though Scott's summary is completely right in its essentials, I would like to put some details straight: We don't think that "Zero Code BPM" does not work at all - Zero Coding and our approach are both reasonable, depending on what you have and what you need. The real problem is that most of the zero coding BPM vendors claim that you should use their solution for situations where you actually shouldn't.
Yesterday we released version 2.0.12 of our modeler with a number of bug fixes as well as layout and usability improvements. One highlight is splitting layouted sequence flows to insert new flow elements between two connected nodes. Other things worth noting include Improved integration of the model wizard Proper support for event subprocess / non-interrupting start event Diagram image generation handles spaces in diagram file names correctly Event Subprocess Wizard integration We also fixed some crucial bugs, reworked the layouting of participants and the resize behavior of pools, lanes and subprocesses.
bpmNEXT has definetely been a Home Run, as Bruce Silver puts it. Bruce has impressively succeeded in creating a real think tank event for BPM thought leaders sharing their ideas, visions and quite a lot of crazy new stuff around BPM. For those of you who do not know him: Bruce Silver is the BPMN Guru in the US, and I really admire him for both his competence and all that he has done for the standard.
Along with the launch of our open-source BPM platform we made the camunda modeler available to the public, both as a software and as source code. With the camunda modeler, you get at free modeling tool that integrates in your Eclipse IDE and focuses on seamless modeling of process and collaboration diagrams. But see yourself: We invite you to try out the modeler, give us feedback and contribute to it. Present, past and Future The camunda modeler is based on the Eclipse BPMN 2.0 Modeler that integrates into the Eclipse IDE.
The camunda bpm stacks currently includes three web apps: cycle, cockpit, tasklist. All of them are rewrites from a JSF 2.0 ancestor version and with this post I want to explain the decision to built them on a HTML5 plus REST architecture and not with <insert java web framework here>. Its clear that the web itself is based on the client-server principle. Many Web frameworks like JSF, Vaadin etc. implement it like this: Provide a abstraction layer to define the HTML + JS + CSS Code to generate (Java Code, Facelets etc.) which in the end is your application On request, generate the code sent to the browser, initially create a session model for data binding etc.
camunda BPM comes with a fresh REST API based on JAX-RS. Its goal is to expose the process engine services as broadly as possible. That means we aim to enable you to interact with process engine services via REST with similar expressiveness as in plain Java. With 7.0.0-alpha1, we provide methods such as task querying that already realize our desired degree of detail (similar for process definitions and instances). For future releases, we plan to broaden the scope to reach the afore-mentioned goal.
Independent BPM Analyst Sandy Kemsley writes about the camunda BPM launch on column2: http://www.column2.com/2013/03/stick-a-open-source-fork-in-it-camunda-bpm-splits-from-activiti/
I am proud to announce that today camunda launches a new open source BPM project: camunda BPM. At this juncture we part ways with the Activiti project which we have contributed to since the first days. Leaving Activiti is a sad but necessary step for us. Starting as a consulting company, we have built a customer base of 250+ in little over 4 years. Last year, we entered the BPM vendor business with the camunda fox BPM platform.