CNET también está disponible en español.

Ir a español

Don't show this again

Tech Industry

Off the hype meter

Industry watcher Jon Oltsik offers his nominations for technology categories in which the elevation of hype has become an art form.

It doesn't often happen, but I recently found myself agreeing with Oracle CEO Larry Ellison. In a Wall Street Journal interview, Ellison was quoted saying, "We became the largest industry in the world by selling things that people didn't want to buy."

That's because the technology industry has elevated hype to an art form.

Chief information officers may have slowed down--or entirely stopped--buying computer products and services, but the industry hype machine is busier than ever. Here's my update from the fluff front.

Storage interoperability standards
Sometimes I wonder whether the folks running the storage industry are living in a world of their own. Organizations like the Storage Networking Industry Association can tap the engineering talent of every vendor that matters. Yet all that collective experience has produced only a couple of solid interoperability standards--Common Information Model (CIM) and Bluefin--that have barely made a dent in the marketplace.

Even though talk of standards, connectivity and application programming interface (API) swapping proliferates, products from EMC, IBM, Hitachi Data Systems and Hewlett-Packard can barely communicate with one another. Who truly represents the voice of an end user? Where's the engineering leadership? Where's the talk of a simple objects access protocol (SOAP) interface for storage?

Intrusion prevention
Those pithy scribes at Gartner decided that Intrusion Detection Systems (IDS) was not conducive to intelligent conversation so they came up with "intrusion prevention." When an intrusion prevention system detects an attack, it's supposed to take proactive countermeasures, shutting down network segments and alerting firewalls of suspicious Internet Protocol (IP) addresses.

Most IDS vendors now claim they do intrusion prevention and some indeed do. But the security world is still chockablock with immature products, limited interoperability and oodles of false-positive events. Note to the security industry: The job of analysts is to wrap visionary concepts around 10-dollar words. Yours is to build products that protect applications, data and infrastructure. The time to perfect IDS is now.

Business impact management
Get ready for another dopey concept, courtesy of the management software gang. Business impact management (BIM) is a management model for measuring and managing applications and technology infrastructure. The problem? For starters, BIM is loosely based on old standby models like the Telecommunications Management Network (TMN) and OSI-M (circa 1985 or so). It also is more about selling enigmatic software than it is about fixing problems.

CIOs know something that has escaped the attention of management software vendors: Solving complex system and network management issues has more to do with policies, processes, operations and skills than it does with software tools. So tone down the hype, focus on the real issues and throw BIM into the bin.

Tablet PCs
Anyone remember Go Computing, GRiD, Dauphin or the Apple Newton? If not, you're in trouble, because history is repeating itself--only this time, hype master extraordinaire Microsoft is leading the parade. When he unveiled the Tablet PC, Microsoft Chairman Bill Gates said the launch marked an "exciting new era of mobile computing that is limited only by the imagination of its users." Maybe.

Tablet PCs certainly include some neat engineering features like low-power central processing units (CPUs) and legacy-free basic input/output system (BIOS). But handwriting recognition is still weak and can anyone think of a killer app here? Tablet computers eventually may become popular in some areas, but isn't this the same blah, blah, blah that has surrounded every new technology?

Grid computing
Grid computing reminds me of hydrogen-powered cars. A neat thought, but still years away from practical reality. Grids are supposed to unify geographically dispersed systems into one common computing engine. The last time I looked, the existing technology, such as massively parallel processing (MPP) and server clusters, had a tough enough time performing this task when processors were in the same room or in just one system.

Between 70 percent and 80 percent of server clusters consist of two nodes and are designed for failover, not scalable computing power. If you really want to share processors, it's going to take a boatload of cash and a roomful of technology Ph.D.s. Since grids are a fantasy anyway, I say we skip them altogether and go straight to talking robots and ray guns. If nothing else, this would certainly inspire more entertaining vendor presentations.

Wireless
The hype just won't die. Two years ago, third-generation wireless (3G) was destined to revolutionize communications and deliver the wireless Web. Carriers spent billions of dollars to purchase wireless spectrums and build infrastructure, so that eight people in the world can today use their cell phones to photograph and e-mail pictures of pancakes that look like their mother. The current spin on wireless is wireless fidelity (Wi-Fi), a cool network technology that lets me share the digital subscriber line (DSL) line with my wife. Networking bigwigs like Cisco Systems have jumped on the bandwagon with enterprise hype about mobility and productivity. Please!

Network managers are scared to death of Wi-Fi, because it lacks security and good management. Honestly, how many people need to be connected all the time? It's bad enough that folks are checking e-mail on BlackBerry devices during meetings. Now you're inviting them to scan stock portfolios, sports scores and the latest news as well? Wireless networking vendors should abandon their hype and start to market a simple slogan: "Wireless technology--it has no wires." Let us judge what the technology can do.

.Net
I took some heat for failing to include .Net in my December article on the 10 biggest hype jobs of 2002. So let me make use of a second chance. Microsoft decided that .Net hype was so important that it named everything .Net. Talk about confusion squared. The other problem is that .Net is symbiotic with Web services, an original member of the Hype Hall of Fame. My personal belief is that Microsoft wants to confuse the market. When information technology (IT) professionals are perplexed, their natural reaction is to solve problems. This opens the door for Microsoft to come in and "help." Maybe I'm just paranoid. After all, I still believe that Microsoft is a monopoly.

Voice over IP
About 25 percent of the voice traffic is now IP-based, and some companies are realizing real financial benefits. Still, voice over IP (VOIP) remains quite buggy and proprietary. And while good old private branch exchanges (PBXs) lasted 12 years or so, VOIP equipment will need to be upgraded once every three or four years. How much will that save? VOIP is certainly the way to go in a green field project, but for existing facilities? No doubt, VOIP is the future but so is death--and you don't see anyone hyping that.

Linux
This freeware operating system generates tons of positive and negative hype. Most of the "hooray for Linux" hype comes from anti-Microsoft bigots, Intel server manufacturers or service providers. Not to be outdone, Microsoft's hype machine responds with its own flood of anti-Linux messages. The upshot is the creation of a cottage industry of Linux pundits, magazines and trade shows. Who cares who is right? It's an operating system, folks, not the solution for world peace. Linux is ready for some enterprise applications, while UNIX, Windows and Multiple Virtual Storage (MVS) better serve others.

Fast PCs
Remember how cool it was to move up from a 486 machine to a Pentium computer? Those days are over. PCs are now like cars--the best economic strategy is to simply beat them into the ground. This strategy may make "Click and Clack" happy, but not Intel, Advanced Micro Devices, Dell Computer or HP. Their combined hype machines will continue to present the yackety-yak on why you need a 4GHz machine. But unless you spend more time each day on 3D gaming, ignore the come-on.