(j3.2006) Liaison to IFIP WG 2.5

[keith bierman]

On Aug 22, 2007, at 7:10 AM, Lawrie Schonfelder wrote:

> ...
> would be useful. An optional standard should exist. BUT <big but>  
> where

A core bit of tension is this notion of "mandatory+optional" vs.  
having a monolithic standard.

Clearly some feel that M+O is *good*. Having had to cope with the  
chaos that the original SCSI standard imposed (a small mandatory, a  
large set of optional bits, and any real hw needing some bits from  
the optional bits ... but often not the bits one had support for) I,  
for one, have a poor opinion of the M+O approach in practice.

It was, arguably, the heart and soul of the original f90 effort, but  
not what we actually deployed. I think there is a bit of a lesson  
buried in there.



---
Keith H. Bierman   khbkhb at gmail.com
<speaking for myself*> Copyright 2007