--- In [email protected], "Nick Gall" <[EMAIL PROTECTED]> wrote: > > On Thu, May 29, 2008 at 11:02 AM, Rob Eamon <[EMAIL PROTECTED]> wrote: > > I'm still wanting for a distinction between architecture and > > design. I think it is the aesthetics/creativity aspects but it > > doesn't seem that anyone else is getting behind that notion. The > > search continues.... > > Let us know what you find. Steve Jones is looking as well. I still > think that the only solid distinction is the one between high-level > vs. detailed design.
http://www.sei.cmu.edu/news-at- sei/columns/the_architect/2003/1q03/architect-1q03.htm I think I posted a different link earlier in the discussion but the material is the same. This paragraph from that paper captures my quest exactly: "In suggesting typical "architectures" and "architectural styles," existing definitions consist of examples and offer anecdotes rather than providing clear and unambiguous notions. In practice, the terms "architecture," "design," and "implementation" appear to connote varying degrees of abstraction in the continuum between complete details ("implementation"), few details ("design"), and the highest form of abstraction ("architecture"). But the amount of detail alone is insufficient to characterize the differences, because architecture and design documents often contain detail that is not explicit in the implementation (e.g., design constraints, standards, performance goals). Thus, we would expect a distinction between these terms to be qualitative and not merely quantitative." Alas, the proposed resolution, referencing the notions of intension and locality, leaves me wholly unsatisfied. Here is another paper by the same authors that goes into more depth. http://www.sei.cmu.edu/staff/rkazman/ICSE03-1.pdf The conclusion I've come to is basically in agreement with Nick, that the only agreed upon distinction is the level of abstraction. Which to me seems utterly pointless. I worked at a Fortune 50 tech firm some years ago, and all job titles tended to include "engineer" somewhere in the title. And when they didn't, there was usually an effort to change the title to do so. I used to joke that the administrative assistant title would eventually change to administrative engineer. Eoin Woods points out in this presentation: http://www.sei.cmu.edu/architecture/saturn/2008/presentations/Woods_Ke ynote_SATURN08.pdf that the use of architect within the IT profession boomed in the 90s. Our industry has successfully proliferated the use of "architect" presumably since it sounded cooler than engineer or analyst or "Analyst II." It seems a shame to me that we swiped a term from another discipline and then reduced it to nothing but an equivalent to high-level design. Thanks for indulging this little academic exploration. -Rob
