🚀 go-pugleaf

Submitted-by: henry@zoo.toronto.edu (Henry Spencer)

In article <1hjd0sINNobl@ftp.UU.NET> knighten@SSD.intel.com (Bob Knighten) writes:
>This is a commonly expressed claim ("standards should codify existing
>practice") and it may even be true for software, but I would certainly like to
>hear some justification.  Certainly it is not true for a great many other
>standards, such as many computer hardware standards, standards for electrical
>fixtures and building standards, where specification of a standard often
>preceeds the existence of any product...

Perhaps a better, if more cumbersome, way of phrasing the principle would
be "codify areas that are thoroughly understood".  The most successful
standards, even in the areas you mention, usually codify things that have
at least been demonstrated in the laboratory, and often they codify fairly
minor variations of existing practice.  The crucial point is that we either
know the specific formulation works, or we understand the design space in
the neighborhood well enough that we can confidently predict it will work.
This is emphatically not true of many of the well-meaning attempts at
software standardization currently underway.

-- 
"God willing... we shall return."       | Henry Spencer @ U of Toronto Zoology
       -Gene Cernan, the Moon, Dec 1972 |  henry@zoo.toronto.edu  utzoo!henry

Volume-Number: Volume 29, Number 100