Kendall’s latest, Community and Specifications includes

Megginson’s Four Laws of writing successful specs: 1. Simplicity succeeds

  1. Process is poison
  2. Code first, then specify
  3. Almost every new spec fails anyway

(as the author of SAX presumably he knows of what he speaks) In a similar vein, the David Clark quote I had as my sig for years (and am thinking of reviving)

“we reject kings, presidents and voting. we believe in rough consensus and running code.”

Been struggling with this of late, particularily in online organizing which exagerates communication problems; walking the line between enough process to stay transparent, and enough flexibility to stay open.