I definitely see 'mobile' being forefront in the future, but wireless technology will have to improve a bit more to support it.

At the main campus of the school I work at, we've abandoned our computer lab several years ago to make room for another classroom. We have computer stations in the hallways for public student access, but our 'lab' has been replaced with ~60 laptops with two wheeled carts for storage and charging. They are connected to active directory with wireless authentication prior to logon for all the benefits of the active directory domain which we enjoyed when we still had the lab. The portability of laptops is extremely convenient, even borderline necessary for today's classroom environments. These laptops get used quite frequently. Additionally, most of the teachers have their own laptops, a few laptops and ipads for learning assistance. Our wireless is used quite heavily.

Embracing the future, our campus created an "open technology" policy in regards to mobile devices and personal laptops. Unlike other schools and businesses, we allow people to bring their mobile devices and use them with our wireless network. Everybody and their dog has an ipod and/or smartphone connected to our wireless. This naturally means that everybody's ipod/smartphones are taking up valuable connections/bandwidth on our wireless access points. Granted, most campuses and businesses don't go this far, but it will become much more commonplace in the future.

Our access points are Netgear WNDAP350 (6 of them) with a WMS5316 management controller. This is a mid-range system, roughly $350-400CDN per, as we are a small school (~170 students, 15-25 teachers and other staff at this campus). Despite load balancing via the controller, strategic placement of the WAPs to account for the architecture, simultaneous b/g & a, and channel planning to avoid frequency or zone overlap, we still get connection and/or bandwidth issues if too many users are clustered in one area. A high end system like Xirrus or Ruckus would would perform better, but cost significantly more (~$1500-2000 per unit).

Our wireless is secured with WPA2/Personal PSK for the best balance of security and user convenience. I would use radius to further secure the network, but a lot of handheld devices don't support it. At the time I wrote this post, according to the management controller, there were 90 wireless devices connected to our network (all ours, no rouge devices). This is average. On a bad day, we'll have around 120 devices.

Now, with the knowledge that each wireless radio can only handle so many connections at a time, consider that:
1) wireless devices are going to become more popular in the future, and local area network access is going to be expected, or even required
2) that wireless access points can only handle so much connections/bandwidth per radio. A consumer grade router/WAP can handle 5-15 connections, and a high-end WAP ~40-50/radio.

Ignore my ranting, and look at the pattern. As mobile devices become more popular and replace traditional computing, and more companies embrace the BYOD ideal, wireless budgets will get strained more and more heavily. The radios in common APs will have to improve in quality quite a bit or be lowered in price for this future to be as painless as possible. Most organizations don't have budget for the high-end stuff.

My experience tells me that unless wireless technology improves, or drops in price, everybody will soon be looking at same issues I face. A copper network is stable and fast. Wireless may work fine at home with only a few connections, but in a larger enviroment, it's messy at best. A good wireless network costs money, and not everybody can dish-out for a good system.