Friday, April 07, 2006
The Second School of Semantic Science
ontologyMapping Glass Bead Games
Continuing discussion on the SOS CoP forum
Communication from Andrew
Rebekah,
I think these observations are great (the community is working, I think)! One of the problems I've had is that a lot of these things were either internal assumptions ("well, of course that's the way it is") or just more hazy concepts (for me).
Between Paul and yourself putting some words down, I think it provides some valuable, concrete points to use when discussing SOA and Web services, not only with business stakeholders. It's also a way to increase the awareness of people designing and implementing solutions that we can't afford just to bury our heads in religious arguments about technologies and pretend these problems either aren't there or that someone else will fix them.
Masochist that I am, I'm reading Boehm's "Software Engineering Economics" based on references to it in some other things I was reading. Even though it was published in '81, this quote seems like a lesson that the industry has yet to learn:
"From the standpoint of
_practice_, this phrase [useful to man] places a responsibility upon us as
software engineers to make sure our software products are indeed useful to
people. If we accept an arbitrary set
of specifications and turn them into a correct computer program satisfying the
specifications, we are not discharging our full responsibility as software
engineers."
All of the things Paul and yourself mention fall squarely into that part of our responsibility that is often ignored because, as you pointed out, it's the hard stuff.
Writing code and designing software in isolation are relatively easily, and you're in control of the environment. Thinking about the implications of what you're doing is too often seen as the "hard stuff", "not necessary" or "people/manager stuff" that we (collectively as the software industry) have allowed to become acceptable (and ingrained, even).
As you have alluded to with the potential global aspects and interactions resulting from service-oriented architectures, we can no longer allow these things to happen. Maybe this is part of the process of "growing up" for the industry.
Thanks,
Andrew