Ivan Kornienko and I presented "BPM for Mobile, Mobile for BPM" at bpmNEXT on March 20th.? We walked through our thought process around mobile as it relates to BPM, shortcomings of the current approaches (as we see it), and some discussion of how it could be better.? We also emphasized how our solution to BP Mobility integrates into the IBM Process Designer - allowing BPM designers to stay within their IDE of choice while they build out mobile UI alongside desktop UI in their BPM models.
This is one of the talks that we wanted to give at IBM Impact as well - but it is up to IBM to choose it (or not) as content for their conference!? I trust that we'll be able to get the video or link to it at some point down the road.? IBMers, if you're listening, the talk went over quite well with industry experts from across a number of vendors.? I expect it would be interesting to IBM customers as well!
Sandy Kemsley was in the audience and wrote the following on her blog:
Problems with existing mobile BPM: an unprioritized activity stream as an interface for business processes is no better than email: everything is mixed together in a single view, with no process context; using desktop HTML instead of native apps ignores device constraints and capabilities. Rethinking that requires a more responsive UI to adjust to size/orientation/touch aspects of mobile screens, hybrid apps to leverage native capabilities while sharing some code with the desktop, and REST APIs to provide process-specific context. Demo shows their add-ons to IBM BPM, allowing new UI components to be added to UIs created within the IBM BPM design environment that better support mobile devices through automatic form factor (size/orientation) adjustment, gestures (e.g., iPhone swipe), and use of device capabilities (e.g., camera). Although the trend is towards native HTML5 and responsive UIs, hybrid and native apps are bridging the gap in the interim; having a common IDE for all development means that there don?t have to be separate desktop/mobile development teams that use different tools. They also advocate specific apps for major process-related functions rather than a common task list, reducing the number of clicks required to complete a task.
Thank you Sandy for capturing notes on our talk!
We were asked some really good questions afterward:
- is every activity a good candidate for a mobile device (or, vice versa)?
- shouldn't all BPM processes be in one app?? We think there's a tension between putting "everything in one sock" and realizing you have more than one sock.? There isn't one right answer, but we think one app isn't the answer, and perhaps "one app per process" isn't the right answer either (although, in one case, we wrote three different "app" solutions for one process...)
- a few questions about how it works - what's configurable, do I really have control over the right pieces, etc.
- Also talked about why our thinking had changed-? that a year ago, or two, native apps seemed the only logical way to address the iOS devices.? Addressing different form factors with HTML meant writing 3 different HTML apps - but now, with responsive UI techniques pioneered by startups, we can write one app, with a bit more work, and have it work on a range of devices.? It is more work, but it isn't 3 times more work to support three form factors.
The best part about giving the presentation was the conversations that it initiated afterward - there are a lot of kindred spirits in the BPM community at this conference, and many of them are thinking about the problem of extending their work to mobile platforms. I hope we can be of some help in that regard - either by illustrating our journey along the way or by directly lending a hand with other efforts.
I want to thank Nathaniel and Bruce for giving us this opportunity to speak.? Hope to do it again at a conference near you!? Here are the slides, however, as you can imagine, it loses something without the voice-over -