The question posed was, “What does the passing of Google Wave mean for BPM and ACM”?  When Wave was announced, I was personally very interested, and am sad to see its demise.  I suspect however, that Wave has made its mark, and that mark will continue to influence design for many years to come — among those who understood what it was.

What Was Wave?

First and foremost, Wave was a protocol.  It can best be described as a competitor of HTML, SMTP, and IM.  That is what was interesting about it, the way it blended the synchronous capabilities of IM, with the asynchronous capabilities of HTML, and the way that SMTP allowed individual messages to be forwarded and ultimately assembled into a structure like an online forum.

It was a fresh take as trying to tackle both synchronous and asynchronous.  It was an outline structured document with forum like capabilities, but it also recorded the entire history of building that document.  Thus you could fetch the current document, and then see it continue to evolve in real time as others modify it.  This ability for mass modification of a single “document” is unique.  And the document itself is designed to convey a conversation between people as well, a kind of structured conversation reminiscent of gIBIS from 1988.

Is Wave a replacement for BPM or ACM?

I am not sure where this idea came from.

Early after the Wave announcement, SAP did a demonstration of collaborative process design using Wave.  This was a good demonstration because it leveraged the merged synchronous/asynchronous nature of the wave protocol.

I think then many people got confused between collaborative design of processes, and the actual running/management of those projects.  This mistake is common among those who view all work as something automatable with a BPMN process. That is probably the reason that at several talks I was asked if Wave was a “competitor” of ACM.  Wave might have been a good tool to help a system communicate, but just go look at the list of required capabilities for an ACM system (Mastering the Unpredictable, Chapter 5) and you will easily see that Wave has almost none of the required capabilities by itself.  However, Wave, or something like it, could be a useful ingredient of an ACM system.

Most case management today, with unpredictable processes, is run using email, with a large part of the remainder run using paper files.  It is true that Wave would offer a new option for both communicating (synchronous) and document storage (asynchronous).  As a system architect I looked forward to leveraging this new protocol for its advantages, but there was never any consideration that it would replace the entire category of Case Management, nor would Case Management become exclusively dependent upon it.

In any case, let me state clearly that the passing of Wave will not effect in any way BPM or ACM efforts, neither positively nor negatively.  It is simply a different category of thing.

Why did Wave Fail?

From the very beginning, the fact that Wave was so separate from other channels of communication was a problem.  No integration to email was a common complaint.  You had to get a whole new ID for wave.  There was no graceful degradation that allows part of a group to use wave, while another part sticks to the old email.  Thus I could only discuss on Wave things where I knew all the participants would be on Wave — a very small list of my contacts.

Group oriented technologies are always sensitive to this.  If you have a group of 20 people, and 1 of them fails to have access, then the entire thing fails.  You have to have >95% of the group on the medium for it to be useful for actual work.  There needs to be a way to ease people into it, first getting your toes wet, then your foot, and so on.  Wave required that you jump in completely.

As an experiment, however, Wave was a huge success.  It demonstrated a specific capability which was surprising.  It is like a vision of a better world that you never dared to believe might exist.  There it was, demonstrating amazing performance and integration.  Now that we have seen what is possible, systems architects will not forget.

Had Wave compromised to provide the requisite email and IM integration, it probably would have compromised the functionality to a level that would not have demonstrated the full potential.  This is hard to describe, but occasionally the high-tech world need a “purist” demonstration for a capability; one that is impractical for for actual use, but still a demonstration of the possible ultimate goal.  Wave accomplished this consummately.

What about the Future?

Google now has their “buzz” which looks quite similar from the outside, but better integrated to email.  It is not clear to me the relationship that Buzz has with Wave, but I am supposing it achieves a similar effect, except using more traditional protocol infrastructure.

While sad to see it go, my guess is that memory of Wave will still drive system design for many years to come, and possibly, some day, an equivalent combination of synchronous and asynchronous communications will be available, but somehow better integrated to the legacy infrastructure.