I can't tell you how many times in the past few months someone has stopped me at a conference or other event to ask what we're going to do with all that performance in today's 54-Mbit/second wireless LANs. Now, where have we heard that before? Remember 10-Mbit/s Ethernet? What were we going to do with all that?
We did a lot with it-and it was so slow that we all readily upgraded to 100 Mbits and probably will move to 1,000 when we can afford it.
There has been one great truth in networking that I think will always apply: More performance is never enough. And performance, along with price, keeps driving the market, the two together defining the curve that one needs to be ahead of to remain competitive.
In recent months, the entire focus of the wireless LAN industry has been on price. We've seen chip set prices essentially collapse, rendering worthless business plans produced only a short time ago. It's pretty clear that this trend is going to lead to a shakeout in the WLAN chip business, and that's probably a good thing.
The trend is also resulting in dirt-cheap consumer and even enterprise WLAN gear-again, a good thing in terms of expanding the market. WLANs will likely be standard equipment in well over half of the notebooks shipped next year. Cell phones and PDAs aren't far behind.
But at some point, the great truth noted above reasserts itself: Even 54 Mbits/s is too slow. And, of course, 54 Mbits/s is the speed that the marketing department guarantees the product won't exceed, not necessarily the speed it will meet: Actual performance is usually much, much lower. I suppose we should be accustomed to this by now; consumer ink jet printers can't really print 19 pages per minute, and yet we continue to tolerate such metrics as, if nothing else, a pleasant delusion.
But the reality is that we can-indeed, we need to-go much faster than 54 Mbits/s. Some two-channel 108-Mbit products are available, and 802.11n will provide at least that in one channel. Meanwhile, a whole crop of startups, including Airgo, Bermai, Cognio and Meru, continues to push the WLAN performance envelope. I recently attended a lecture at MIT on gigabit wireless LANs!
Once today's price wars are over, it's back to a fundamental emphasis on performance.
Craig J. Mathias is principal of Farpoint Group (Ashland, Mass.).