Embedded Systems Europe Standards Column vol 1.1 Jan/Feb 2009 Standards: More Standards than you can shake a stick at! By Chris Hills

 

These are my own personal views and not those of my company Phaedrus Systems see www.phaedsys.com which is where the full version of this column, with links etc, resides under the Documents tab.

 

For those who are new to this column, now Tech Insights have merged the two magazines, it is mainly about standards. I am on the UK end of the ISO C and ISO C++ panels also the IEC 61508 Functional Safety and MISRA panels. However, I will discuss my view of standards in the broadest sense. As we all know standards of education training and programming have fallen and are not as good as when you and I did it. I am starting to sound like my father now! 

 

On the standards front BSI closed down the BSI-ISO C panel in 2008 due to some unprofessional behaviour on the panel that I hinted at in this column over the last few years. MISRA has been invited to join the parent BSI Languages panel as the main industry player for C and C++. However, the problem BSI has is that it needs to sort out the parent panel that permitted the problems in the C panel for most of this millennium otherwise things will not improve.

 

Whilst on standards and their panels another topic I have touched on in the past: the ECMA fast track in the ISO system.   This is where ECMA (European Computer Manufacturers Association) http://www.ecma-international.org/ produces many standards such as:

TC45 - Office Open XML formats,
TC46 - XML Paper Specifications
TC49 – TC1  ECMAscript  (better known as ISO/IEC 16262 Java script) 
TC49 - TG2 - C#, 
TC49 - TG3 – CLI
TC49 – TG4 – Eiffel
TC49 – TG5 – CLI/C++

 

ECMA does many other standard as can be seen by the languages being Technical Group 49 That is in the 2008 ECMA hand-book: prior to that they were all TG39. However, the “problem” people seem to have is with the ones listed above.  

 

ISO has a long and slow system. If can take many years to produce new standards. Also in many countries literally anyone can be on a local National Body panel (which was is part of BSI’s problem with the C panel).  This can lead to all sorts of things getting into the standards from small pressure groups and non-industrial people apart from the leisurely process slowing them down. This is partly why the current ISO C (C1999) has not been fully implemented by anyone nearly a decade later and many embedded compiler companies refusing to move off C90. 

 

Part of the problem in computing languages as opposed to lawn mowers, radio systems, networks etc there is no legal or commercial imperative to adhere to a language standard so no-one really cares to much. On top of this there are too many oddballs and non industrial people playing in the BSI/ISO languages standards committees slowing things up for serious players to get involved.

 

ECMA on the other hand is made up of committees of the members from the leading players in the industry all of whom have a commercial interest in developing standards rapidly.  Like the MISRA committees the ECMA groups are made up of qualified and experienced people. ECMA groups have an industrial/ commercial interest in getting a standard they can work to, and doing it quickly. In this industry five years is a long time and an ISO standard can take ten years.

 

Therefore ECMA has got a fast track at ISO by passing all the committees and normal procedure arriving with a final draft. So ECMA standards go in to the ISO system with a short discussion period and unless a lot of people vote “NO!” with reasons in a concerted effort the ECMA standard becomes an ISO standard.

 

The “problem” is that whilst Microsoft, Intel, Hitachi, IBM, Microsoft, HP, Adobe, Fujitsu, SGI, NXP, Sony, Toshiba, Philips, Siemens, Panasonic  etc among others have a seat on these panels there appears to be no small companies and individuals that you can get on the ISO panels. Thus what The Industry wants The Industry gets and when it wants it rather than at the gentlemanly pace of ISO. This can lead to say some of the ISO people standards that are done in hast and have holes in them… well given that no one has implemented C99 yet I don’t think the ISO languages panels have a leg to stand on! 

 

The problem is as has been discovered is that some companies can get languages like C# and CLI/C++ and ECMAscript (JavaScript) pushed through quickly without any “public” scrutiny. Or at least a very short time when the relevant ISO panels can look at them for comment.  This really has caused a hic-up recently with the CLI/C++ effectively changed the C++ language and the new OOXML standard it is fast tracking though ISO at the moment which is effectively not the same as everyone else’s XML.

 

Thus, cry the ECMA fast track detractors, at a stroke the “Microsoft” OOXML format becomes the world standard.  I can see their problem however if they were that worried why are the companies who administer the Open Office tools not joining ECMA and influencing?   On the other hand the GNU GCC has it’s own standard for C that differs from ISO C and they were not worried about compatibility.  That said commercial compilers are also becoming GCC compatible as much as C99 compatible.

 

So if it makes commercial sense companies will implement the standards no mater who does them. If there is no requirement from the customers it will not get done as per the current ISO c standard.

 

The real problem is standards are there for commercial or safety reasons. On one had the government

/law/insurance companies want things safe and on the other hand industry wants to make money, which is easier to do if some things are standardised. This is the way the world works. That said the safety standards have a financial implication too. Planes, cars, ships that work cost a lot less than ones that crash. Lost reputations, lost jobs, insurance payout, medical bills, lost market share etc all impact the bottom line. Industry, and therefore standards will follow the money. Always.

 

Talking of money, as some one said to me recently: It used to be “Money Talks!” ….  Now it goes without saying….  The Credit Crunch is having some interesting effects.  As with 9/11 and the airline industry I think many are using the current economic crisis to re-align businesses for the future.  Some companies who are shaky will go, others will streamline whilst the unions are too worried to say “no” because of the threat that the whole company might go away. But as with people saying after 9/11 all the budget airlines would go and cheap airfares were a thing of the past (and we are still inundated with budget airlines) I think we will see evolution not a disaster.

 

Tools companies I talk to (and not just the ones we distribute) are reporting that Q4 2008 was very good. Not just in the UK but Germany, and other parts of Europe also Canada and even parts of the US! So it is not all doom and gloom. The auto industry seems to have hot houses generating the green shoots of next years recovery whilst this years production lines are faltering.  Aerospace and defence projects seem immune to the crisis as they work on ten-year cycles and “National Defence” is always difficult to argue against.

 

So if anyone has any green shoots they would like to share to cheer us all up please send them in to me (anonymity guaranteed as I want to stay in business!)  For the next issue I have some items to make you think and cheer you up.

 

Author Details and contact

 

Eur Ing Chris Hills BSc CEng MIET MBCS MIEEE  FRGS   FRSA is a Technical Specialist and can be reached at This Contact

 

Copyright Chris A Hills  2003 -2008
The right of Chris A Hills to be identified as the author of this work has been asserted by him in accordance with the Copyright, Designs and Patents Act 1988