Wireless LANs in Public Places

This category of WLAN deployment is the least well-defined, and has the most potential to change the way we use networks in our everyday life. As I write this, I am sitting in a coffee shop accessing the Internet and my corporate IPsec-based VPN via a WLAN. Many are attempting to get WLANs deployed in as many public spaces as possible, hopefully to garner revenue from such deployments. Currently, WLANs may be found in coffee shops and other restaurants, airport gate waiting areas, bookstores, and other places. One of these "other places" in which WLAN access will soon be available is on airplanes. Boeing has already tested a flying WLAN, with satellite-based Internet connectivity. In fact, the service has been tested live by Lufthansa. The satellite link limits the responsiveness (due to the roundtrip time^[35] of the signal between the plane and the ground via the satellite).

^[35] The delay, in one direction (i.e., from the ground station to a satellite in geosynchronous orbit, and back down to the plane, or vice versa) is approximately one half of one second.

The practical downside for WISPs is that by far most people's interactions with computers occur either a) at work, or b) at home. Personally, I only lug my laptop around when I have to have it with me. I'd love to be able to use my WLAN card on the train when I am commuting to and from work. As of yet, that service is not available. At best, WLAN hotspots will provide highly localized coverage. While in a hotspot, you could enjoy Internet access at speeds on the order of 1 to 2 Mbps, but when you leave the hotspot your access speed drops to zero Mbps.

Some influential telecommunications market analysts think that WLANs are the wave of the future, and that eventually APs will be densely distributed (at least in densely populated areas) simply because they are so cheap to buy, deploy, and operate. It's not clear to the author if there will be a business case for these wireless islands (archipelagos?), or if it is all supposed to be free. The author fully supports people building shared community networks, but networks need maintenance, and that costs money. I'm rambling a bit here, but I know that some Bell Labs researchers proved that you can make an arbitrarily reliable system out of arbitrarily unreliable components, as long as it was possible to characterize the "unreliability" of the components and ensure that the system was designed to compensate for it (i.e., by increasing redundancy in certain aspects of the system design). One might suppose that if enough people in a given area were operating overlapping WLAN "cells," that connectivity could be available as long as at least one of the cells was still operational. Time will tell whether a "service" that is, at its most fundamental level, best-effort can succeed.

The current deployment platform of WLAN STAs is primarily laptops, which are not exactly so portable that people take them everywhere they go. However, as PDAs with IEEE 802.11b capabilities become more common (even cell phones with Internet access via integrated WLAN hardware have been announced, especially if 3G and 4G cellular doesn't take off), the potential user base for WISPs will get larger, since more people will tend to have a WLAN device with them. The WISPs' main competition, ultimately, will be the public-access WLAN deployments that many activists are setting up. Whenever WLAN users have the option of choosing free access over paying for it, it's easy to presume that they'll go with the free one. The WISPs will need to offer more than just vanilla Internet access for people to be willing to pay for the service (and the price will have to be right).

The ability of WISPs to derive revenue is obviously tied to the number of people who are carrying a device with them through which a WLAN could be accessed. Clearly, business travel is the one case where people will be likely to carry their laptop along with them (hopefully they won't take it on vacation!). Thus, WLAN deployments in airports, hotels, and on planes make a lot of sense, since these are places in which larger-than-average concentrations of business travelers may be found.

Despite "cherry picking" the best locations that are frequented by businesspeople on the go, the total user base of traveling WLAN users is still going to be small, compared to those using WLANs on a regular basis at home or at work. Besides the limited user base, WISPs are also under price pressure because many ad hoc organizations are also setting up low-cost or no-cost WLANs in many of the same places as the WISPs, so it's not clear how many people will be willing to pay for this service (or how much they will be willing to pay, which is a separate question altogether).

A well-known example is a certain popular coffee shop chain that has signed up a nationwide WLAN access provider, but in certain branches of the chain, cheaper alternatives are available. The price structure of the nationwide provider is seen by some to be too expensive. If so, simple laws of supply and demand should tend to equalize the price such that the provider is able to attract sufficient customers to cover their operating costs. The nationwide plan does have some inherent advantages that would be worth paying a small premium for, in that a customer can count on finding a WLAN-equipped coffee shop in most places in the country, even in less populated areas that may not be served by multiple WLAN access providers. In the cities, more alternatives are likely to exist, so perhaps a tiered pricing structure will evolve that will reflect where you are when you use the WLAN.

Today, roaming and staying connected (via a laptop) is not exactly what you might call "seamless," at least not from an ease-of-use perspective. The advantage of this cellular phone based approach is that it is going to work over a wide geographic area so, it is slow but widely available.

The cool thing about WLANs in public places is that they enable end users to be truly mobile, not only in the sense of being able to move, but in the sense that they can remain connected, at high speeds, while moving (or while sitting in a coffee shop, after they have stopped moving temporarily). The bandwidth that a roaming WLAN user will have access to is much larger than other current mobile alternatives, which may be limited to 128 to 256 kbps (perhaps up to 1 Mbps in a couple of years, if 3G wireless technology is deployed as promised), but probably much less, via a cell phone connection to carry data communications to and from a PC, either over a wire between the PC and the phone, or over a wireless short-hop connection such as Bluetooth.

The real trigger that would get people to carry devices with them is not just making them smaller and lighter, but also having them fully leverage mobile-related networking technologies. When these devices fully leverage so-called "presence-based" services, the users may want to be able to be reachable whenever their PDA (or whatever) is within range of a WLAN. Today's presence-based applications, such as Instant Messaging, rely on "buddy lists" that enable the service provider to tell each user when his or her friends are online. One can imagine a service that not only knows that a friend is online, but also knows the spatial distance between the two of you. If you knew that a friend was several blocks away, you might want to meet him or her at a location in between, and a service that could help enable such meetings might be very popular, especially if it was package-able into cellular phones that are already being used as text messaging "terminals" but that really don't rely on presence (the way Internet-based Instant Messaging does). Many forms of presence-based applications present themselves; for example, you could find out what movie start times were within walking distance of your closest (in the sense of distance) friends. The service could even buy the tickets for you, to save you from standing in line.

Public Access WLANs: Universities Exploring the Future

A clue to the transformations that such pervasive WLANs might have on society in the next 5 to 10 years is described in the October 2002 issue of Wired magazine. A case study on Dartmouth College is presented, in which the effects of the pervasive WLAN technology are analyzed. Later in this chapter, I will summarize some of their findings. One of the more amusing is that the bell tower on campus can be controlled wirelessly, allowing students to change the tune. The bell, for whatever reason, was attached to the university network, allowing it to become a resource available to any wireless user; in the case mentioned in the story, the wireless user is actually a "bell-hacker."

A university campus is more like a hotspot than a typical corporate deployment, since the university is likely to experiment much more with unexpected applications based on pervasive wireless connectivity. One user even modified the software that triggers reminders in his PDA, so that the distance from the appointment would be taken into consideration (based on the location of the user at a given point in time).

But is a university campus a wireless hotspot? Such deployments as have happened at Dartmouth are happening all over the country, at perhaps hundreds of schools, including Carnegie Mellon, Stanford University, the University of Texas at Dallas, the University of Minnesota, Virginia Tech, and so on. It is indisputable that at those universities where WLANs have been deployed, the WLANs have transformed the way the students and faculty work, play, and communicate.^[36] The impact on society once these über-wired people get into the workforce is difficult to estimate, but today's trends toward "mobile lifestyles" will almost certainly accelerate. The "mobile professionals" of today, with their cellular phone and laptop will surely seem old-fashioned in a short time.

^[36] When one student user was asked to name the biggest difference between Dartmouth and another school that he was attending, it was that at Dartmouth, no one knows anyone else's phone number! All communications are via email or instant messaging, and soon VoIP over the WLAN.