Posted by John Keller
It's difficult to overestimate the importance of aerospace and defense software in ever-more-sophisticated military electronics. Software is perhaps the most crucial enabling technology, as well as the riskiest vulnerability in military weapons, communications, navigation and guidance, and most other applications that give U.S. warfighters a crucial technological edge over their adversaries.
Nevertheless, it never fails to astonish me the difficulty that software companies have in explaining how their tools, operating systems, integrated development environments, services, and other expertise represent the enabling technologies they truly are.
When I talk to software companies, as I did this past week at the Embedded Systems Conference in Boston, I always want to hear in a clear, concise way, what their software engineers bring to the table for the aerospace and defense systems designer and the military platform integrator.
More often than not, however, I get a tortured and long-winded explanation of software capabilities, new or upgraded tools, and why some such software widget is better than the one offered from competitors. I rarely get a straight answer to my question of how this particular piece of software can benefit the guy designing a communications system, electronic warfare suite, radar or sonar system, or avionics flight control.
For this, however, I can't always blame the folks at the software companies. Explaining why software is important is hard. You can't pick up a piece of software, hold it in your hands, turn it over, and feel its heft like you can with hardware.
Software, by nature, is an abstract thing, and explaining its importance to people like me who aren't software engineers is a daunting task. Still, it shouldn't feel like you have to be in the club to get it.
I'm sure software companies do a great job of explaining themselves to other software people. On many levels, I think it takes a deep knowledge and appreciation of the challenges of developing software that works, doesn't hog processor and memory resources, can be maintained and upgraded easily, and can't be hacked by the bad guys to understand the latest software products.
Still, I'm always frustrated when I come away from a software show, as the Embedded Systems Conference has become, because I have a nagging feeling that I've missed something. There just has to be some understandable explanation between "my software makes electronics work better," and the gory details of what the software actually does.
That's what I'm after, and it's a big challenge to get anywhere close, it seems.
Related stories
-- Military use of consumer computing like iPads and Android software raise concerns for safety and security;
-- Safety- and security-critical avionics software; and
-- Software middleware providers acknowledge the need for security and safety.
You hit it on the head - the concept that software is "abstract" because you can't touch it. You can't detect ("touch") and know properly hardened metal or correct fuel additives - but they can kill you just as dead. A bad drawing can create a killer part, but the line on that drawing is as ephemeral as the software that might have created it.
ReplyDeleteThe time is here to create a software 'product' as a fixed thing, not a document that can be forever re-edited til the money runs out.
Software is a 'thing' - and you pay for it. Just because it's easy to change doesn't mean it should be and thats the problem with the entire cricital software industry: every line of code is fungible and flexible as though they didn't have impacts far beyond their existence as bits and bytes.