I've just had a very interesting online analyst roundtable with the Systems Engineering wing of IBM Rational. The latest news is that Meg Selfe is moving on to start up a Smarter Physical Infrastructure Group, which (I guess) means another step towards the practical fulfillment of IBM's 'autonomic computing' vision, first put forward a few years ago. I'm sure that 'smart' computing (which I see as being based on the "internet of intelligent interconnected things", more or less) is the future, that IBM has a good understanding of this area and that Ms Selfe has a good chance of making it work in terms of commercial product.
Bret Greenstein is now Vice President, Rational Complex & Embedded Systems at IBM Rational, replacing Ms Selfe. He has a background in grid computing, which is a good basis for engineering the smarter planet.
IBM seems to be making a bit of a success of systems engineering, especially in its natural home, the defense, aerospace and automotive industries. RELM (Rational Engineering Lifecycle Manager; see the original announcement here) is going to be built into everything, tying together requirements in Doors, Rational Composer, impact analysis, test management and so on; and it seems to be getting some traction - although the pioneer adopters aren't ready to go public yet.
The current focus seems to be on building systems of systems that bring together
mechanical, electrical engineering, and the software which increasingly embodies the
functionality of electronic devices. IBM has good standards for interfacing different tools - as well as RELM, there are Jazz and OSLC - in pursuit of this.
This is all pretty impressive, and if you believe in systems engineering (as everyone at the roundtable, including me, seems to) you will be very positive. Nevertheless, I am left with some concerns that Systems Engineering is becoming, or remaining, a bit of a silo, although I'm not sure how many of my concerns are real issues and how many simply a pragmatic result of IBM managing real-world issues with development culture amongst its clients and their use cases. Engineers are happy thinking Systems Engineering; ordinary commercial IT developers often aren't. In the longer term, it may all work out
for the best although we will need to see how much the transformational aspects of the smarter
physical infrastructure impact the back office integrations.
So, to reiterate, I think the position is that IBM Rational Complex & Embedded Systems is telling a strong story around open tool integration; and this is especially strong in the integration of electrical systems, software and electronics to deliver smarter business outcomes. But I also think that these approaches could be adopted more widely, in business IT, and that there are cultural silos in development that IBM may not be doing much to remove in the short term. I'll explore these issues in my next research note.