Everyone likes standards, but standards mean compromises. Tim Dean looks at whether they're worth the trade off.
Standards. It's one of those things where the industry is damned if they do and damned if they don't. There is a clear tension in the technology industry (and other industries, such as telecommunications) between the need to appeal to the public desire for common industry standards, and the relentless march of technology, which is often developed internally, and can't always wait for the standards bodies to catch up.
The problem occurs because we, the consumers, want the best of both worlds, and are quick to criticise the industry when we can't always get what we want. For one thing, we want technology that doesn't bind us to a particular manufacturer in the future. This is an interesting phenomenon as it seems there is a considerable amount of brand loyalty out there, with people sticking with the brand they know and trust until such time as it fails them - and only then does a new deliberation process start to find a replacement. Then again, vendors such as Sony have long been the target of ire because of restrictive and proprietary standards, like ATRAC and MD.
Basically, we want to know that if we buy something from one vendor, we won't be restricted in our choices from there on in. Even if we never change vendors, we like to have the option open to us if we ever need to use it.
On the other hand, standards often take a long time to develop and ratify. Usually this is because there are so many stakeholders, and each one has its own slightly different vision of what the standard should be. This results in compromises being made, and there are plenty of times when the resulting standard is not as good as it could have been if the compromises hadn't been formed.
One common compromise that is forced on to new standards is backward compatibility. Heck, it was only with Windows XP that 640 x 480 VGA was no longer the default base resolution. That means up until the end of 2001 all graphic card manufacturers had to spend time ensuring their hardware was VGA compliant, even if only a tiny proportion of the market had any need for it. Backward compatibility also introduces many limitations to progress, as Microsoft will vehemently argue when talking about the internet.
|
'...we want to know that if we buy something from one vendor, we won't be restricted in our choices from there on in.' |
There are other issues as well. One is simply the relentless march of technology. Many large vendors around the world invest billions in their own research and development. This can often result in great technological advancements, but then there's a dilemma of whether to unleash the technology as a proprietary standard, or whether to collaborate with other vendors, usually competitors, and make an industry standard.
For an interesting case study, we can look at our trusty old 56K modems. In the days of 33.6Kb/s modems, Rockwell and Lucent developed a new technology, called K56Flex, to make use of pulse code modulation to bump up speeds (downstream) to 56Kb/s.
At the same time US Robotics was working on its own X2 protocol to do the same thing. The two developers refused to work together at first, and the two protocols were then unleashed on an unsuspecting market. Shenanigans ensued.
Many an ISP cursed the two companies for forcing them to choose between supporting one technology, or installing two sets of hardware to support them both. Finally, the ITU (International Telecommunication Union) stepped in and used bits from both technologies to create a new standardised protocol: V.90.
|
Bluetooth
For a case study in how not to run a standard, we have Bluetooth. Originally developed by Ericsson in 1998, it was then passed to a Bluetooth Special Interest Group (SIG), which involved over a thousand other companies. The technology itself was decent enough, but when all these companies came together, each with different visions, the SIG got bogged down. Vendors then started releasing their own Bluetooth products before the standard was locked down, and the result was embarrassing for everyone. Interoperability was out the window, and Bluetooth failed to live up to its lofty expectations.
www.bluetooth.org |
This scenario clearly disadvantaged users, but there are times when proprietary standards can benefit users. Microsoft is one of the companies that argues this point very strongly. Microsoft is known for developing upon existing industry standards, such as HTML or Java, and adding its own improvements. These improvements are quite often very technically beneficial, but the down side is they no longer conform to industry standards and can raise compatibility problems with users that don't run Windows.
Another example of vendors not waiting for official standards to release a technology is with WPA, the new wireless security standard (see Tech Horizons, page 22). 802.11i was in draft format, and was in the process of being ratified by the IEEE, but the Wi-Fi Alliance was forced to respond to market demand for better security, and it released WPA. This is a case study of how to do it right. WPA was not released by just one or two vendors, but was supported by just about all wireless vendors. It was also backwards compatible with old hardware, and forward compatible with 802.11i. At the end of the day, everyone's happy.
Ultimately, in the fight between standards, with their compromises and slowness to market, and proprietary standards, with their cutting edge technology but lack of compatibility, open industry standards win. The coming convergence between the IT and consumer electronics industries is sure to abound with standards issues, but we just have to sit tight, and encourage the industry to do 'the right thing', then we'll all benefit in the end - consumers and vendors.