Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!munnari.oz.au!mtiame!marvin!david
From: david@marvin.jpl.oz (David Magnay)
Newsgroups: comp.realtime
Subject: Re: Software primitives for real-time programming languages
Message-ID: <743@marvin.jpl.oz>
Date: 5 Oct 90 09:22:35 GMT
References: <1851@tuvie>
Organization: Johns Perry Lifts, Melbourne, Australia
Lines: 28

The Japanese TRON project has a spec called I-TRON, which is for a (rather
basic) real time operating system. Rather than standardise the primitives,
they are using a higher order of abstraction.

At the end of the day, the goals need clarification before setting off to
solve the problem. Will a set of standardised primitives help anyone ?. If the
goal is to maximise portability, there are several levels of abstraction that
can be ( and have been) chosen. OS level ( a la UNIX/posix,..), the
library call level ( a la C language), the BIOS level (MS-DOS) or the hardware
level (PC's). I suppose other levels could be postulated. Also to be
considered is who will administer the standards, and what are their
motivations in all of this. The cost of supporting a standard, even a PD one,
will effect peoples attitude to it, and the level of support you get.

Perhaps all we need are "generic interfaces", with the implementation details
up to each vendor.

The goals you set out with will select the choice. Lets clarify these a little
further.

Remember, most projects go awry in the first 5 minutes. ( It just takes a
little longer to work out) Lets spend 10 minutes setting the direction.

..........................................................
David Magnay				mail	david@marvin.jpl.oz.au
Boral Elevators				was: Johns Perry Lifts
45 Wangara Road, Cheltenham 3192	phone	(03) 584-3311
Victoria, Australia			O/seas	+61 3 584 3311