Devil's Advocate: Handcuffed by the industry standard

Consultant Martin Brampton again questions what too many of us have grown to accept...

The phrase 'industry standard' is nowadays bandied about ad nauseam. Yet what does it mean, and should we always be adhering to it?

The history of the term should give us pause for thought. IBM, as everybody accepts, invented the PC as we know it. The very first PC had a basic 8-bit bus for plug-in hardware additions and was soon followed by the IBM PC AT with a 16-bit bus. (AT stood for 'Advanced Technology' - the IT business has always been creative with names.) This was widely known as the AT bus and many plug-in cards were made for it.

When a few years had gone by, IBM realised the PC was becoming rather more important than it had anticipated and that other people were making money from it. In an effort to reassert control, IBM created a new range of machines, most of which used a new 32-bit bus designated MCA. The licensing for MCA was much tighter than for the earlier bus standards and offended other PC makers such as Compaq.

In what turned out to be a classic marketing coup, Compaq renamed the old AT bus the 'Industry Standard Architecture'. Thus was born the ISA bus, which is only now disappearing from use. The 'industry standard' sobriquet was one of the factors that eventually led to the demise of the MCA bus. Far more of the rival 'Extended Industry Standard Architecture' 32-bit bus machines were shipped, including some by IBM.

Compaq's move did at least have the saving grace that both ISA and EISA were supported by a number of vendors, leaving purchasers a choice of where to buy their 'industry standard' products. Others have been less generous. Ever since the term was coined, vendors have sought to label their products the 'industry standard' regardless of their acceptance by rivals.

Unfortunately, purchasers of IT have a lemming-like tendency to find comfort in travelling the same path as the others. When everyone has rushed to embrace the same product the cries of anguish start. Why are we paying so much for this technology? How can there be such a lack of competition? Why is there so little innovation in this area?

In most markets, the idea that one product must dominate would be absurd. The Ford Focus is produced in vast quantities. Does that mean anyone wanting to buy a new car should choose the 'industry standard' Ford Focus? Is it ridiculous to want a Mercedes, a marque that has only an odd per cent or two of the market?

But, you may say, this is not fair. We need to adopt standards in IT because otherwise things do not work together. If we choose products with a tiny market share it is difficult to find staff to support them. Vendors that fail to achieve market penetration may withdraw products or even go out of business.

All that is true but we must not forget that vendors use proprietary standards to prevent freedom of choice and to pre-empt future purchase decisions. When they are able to speak of their strictly licensed exclusive technology as 'industry standard' they are at an advanced stage in their marketing programme.

To be effective, we do need strong policies. Decisions have to be made that take account of issues like the general acceptability of products, the possibilities of interoperation and the efficiencies that may come from limiting the number of vendors involved.

Conversely we also need to allow for the excess costs that may result from the inability to adopt new competing technologies, the difficulties of dealing with monopoly suppliers and the lost opportunities that may result from bypassing vendors who are not currently market leaders.

All these issues need careful examination. Reliance on 'industry standards' is just a sloppy way to make decisions.

Do you agree with Martin's assessment? Post a Reader Comment below or mail editorial@silicon.com with your thoughts.

** Martin Brampton is a director and founder of Black Sheep Research (www.black-sheep-research.co.uk ), an independent consultancy providing research, writing and speaking services on a wide range of business and technology subjects. Martin was previously a director at Bloor Research, and has worked with IT as a user and analyst for over 20 years. He is a frequent contributor to silicon.com's Behind the Headlines TV programme and can be contacted at silicon@black-sheep-research.co.uk