80/20(was Re: [xml-dev] modeling, validating )

[pulling a bit worth pondering out of the previous and largely dead thread]

At 12:28 AM 6/19/2003 -0400, Rich Salz wrote:
>To me it fits in the 80 part of 80/20, but maybe my viewpoint is too
>narrow.

The dangerous part of the 80/20 story is that people seem to focus on the 
minor losses involved in getting 80% of the capabilities, not the major 
sacrifices involved in reaching the 20% of work behind those capabilities.

Most people still seems to agree that recent specs provide roughly 80% of 
what they wanted, with some significant variation.  (Schemas and Web 
Services seem to be the primary places where people regularly say "That is 
a completely different thing from what I wanted.")

The other side of the equation is trickier, but I'd guess we've moved up 
from XML 1.0's 80/20 (80/30, perhaps?) to a range from 80/50 to 80/100, 
which isn't nearly as exciting.