8 ms·

Foundations for the Study of Software Architecture (1992) [pdf]

• contingencies 3y ago
  Interesting historical artifact however the conceptual paradigm used seems to be literally "stuff, links between stuff, and why" which is a bit of a cop-out even if essential and timeless. The closing statement, proposing an interpretation of the social reality of computer based systems implementation at the start of the 1990s as "[deeply abd problematically] architect-less", is in severe contrast to present day reality in which "architecture" has run fully amok and any corporate project winds up wading through cloud architects, devops architects, etc. Here's a concluding proposal of my own: It seems the needle has swung too far. Maybe we only have the architecture-kerfuffle because the platforms people build on today are systematically overbuilt and over-complicated? How about going back to 1991 engineering, where the stack was comprehensible, owned, and did not require multiple tiers of organised religion-like specialists to access and utilise?
  • mpweiher 3y ago
    > "architect-less" [vs] "architecture" has run fully amok and any corporate project winds up wading through cloud architects, devops architects, The fact that these two statements actually both accurately describe reality is the crux of the problem, IMHO.
• lincpa 3y ago
  [dead]