Wireless networks are ubiquitous these days. We are all, for better or worse, being bombarded by radio waves in all sorts of frequency ranges, and if we could put a 2.4GHz monitor in our heads, we’d be overloaded by the sheer volume of traffic screaming around on wireless networks using that frequency.

Despite the phenomenal security and reliability issues with it, users ranging from my parents to Fortune 100 companies are using it. What has driven the move from wired Ethernet to WiFi?

Clearly, convenience is part of it. Outside the enterprise, there is no doubt that this convenience is worth it, at least from my standpoint. Recently, I relocated from Cambodia to Santiago, Chile, where things such as acquiring Internet service can take weeks. Many businesses here provide free WiFi to customers, providing a valuable service for many and giving themselves a competitive advantage. I don’t really want to think about how much time I have spent at Starbucks for their magic combo – caffeine, no smoking, and free WiFi.

In the home, the case gets weaker. While staying in a temporary apartment during my move, I could see roughly 15 networks at any given time, of varying signal strength. Almost all of those were protected with either WEP or WAP, and I have to confess that I was sorely tempted to just “borrow” one of those connections with the help of Airsnort and friends.

Now that I've moved to a house, I only have two networks visible, one for each neighbor. Again, they are protected networks. But how many places do those people really use their computers? Is it worth it to them to deal with the hassle of setup, the knowledge that the wireless router will crash, and the interference with cordless phones and other wireless devices in the home?

Mache Creeger did a piece on this in ACM Queue, and I have to confess that I was nearly convinced. "Nearly" because I move so much and seldom live someplace that I own. If I had a more permanent home, I would certainly think twice about the utility of wireless versus a proper wired infrastructure.

Inside the enterprise, the case for wireless is even less clear. For the legitimate user, it's fantastic to be able to open a laptop anywhere in a building and know that, given the proper password, full access to the network will be available within seconds. No more worries about how many network drops a conference room has or how many people will need access, and no more cables draped over the edges of tables and running to the walls.

But what are the costs? Well, for one, managing a large wireless infrastructure is far from free, and the number of access points needed to provide good coverage throughout a large building is decidedly non-trivial. Then there will be dead zones and annoyed people that go along with them. “I can check my email in the bathroom!” is certainly not a phrase I long to hear in my life.

Add these to the security headaches of keeping synchronized encryption keys on all the devices, changing them regularly to limit the damage from compromise, trying to limit the network to the office’s perimeter, and adding the secondary access control mechanisms to prevent anyone with Airsnort from getting the run of the show, and the costs seem pretty high. The wireless network taxes not just the IT infrastructure, but the security group as well.

There are certainly products out there that provide reasonable control security for these networks, but it seems to me that the cost – both initial and ongoing – must be hard to justify. If you have experiences to share on either side of the wireless security argument, please post a reply to the message board attached to this column.

— Nathan Spande has implemented security in medical systems during the dotcom boom and bust and suffered through federal government security implementations. Special to Dark Reading