Why it was invented
Although used only to compute stuffs for academic or military institutions in 1940s (eg ENIAC), businesses started to realise that they can use computers to manage their operation as early as in 1950s (eg LEO I). As computers became more powerful, its computing power could be made available to multiple staffs at the same time starting in 1960s. This was made possible through the use of terminals linked to the computer via early communication technology like dial-up or serial communication. In 1981, IBM entry into the Personal Computer (PC) market lent credibility to the then rising industry, and IBM PC compatible units started to deluge the market from 1982 onward with the likes of Compaq. With wide-spread acceptance of business productivity tools running on IBM PC compatible like Lotus 1-2-3 (released in 1983), companies started using PCs more and more too. Then entered Novell NetWare in 1983 with the promise of connecting a company’s PCs together. Out of so many ways of sending signals between computers, Ethernet ultimately became very dominant by end of 1980s. Local Area Network (LAN) was born.
Even as early as 1985, computer manufacturers like Toshiba already started to imagine portable PCs for the relatively few people who want to be able to use their own PC anywhere they go. Early on, all the portable PCs that a company owns must be connected to the LAN via network cable. Understandably, this sort of limited the true portability nature of the portable PCs within the company’s premise. A wireless LAN sounded like a good idea. With the release of IEEE 802.11 in 1997, this was becoming possible. In 1998, a competing technology called HomeRF also appeared
Obvious commercialisation routes
The obvious opportunities of this new technology lied on the hardware side, both for the manufacturers (which tend to be larger corporations) as well as distributors (some may well be small start up opening up new distribution channel in previously un-touched geographics)
The new wireless networking technology required new electronic components that can process the signal, hence opening up a new field in business. Existing semiconductor company like BroadCom started a new business line to make Integrated Circuits to handle wireless LAN. Some prolific researchers started a new company to cater this new wireless LAN chip set market, for example Atheros in 1998.
Aside from the need for new electronic components, the new technology also opened up new market in network hardware, as in access point and Network Interface Controller (NIC). Traditional network hardware makers for routers, switches, and NICs branched out to commercialise this new technology. Companies like 3Com (Ethernet’s co-inventor, Robert Metcalfe, founded 3Com earlier in 1979) & Linksys from America, D-Link from Taiwan, or TP-Link from China. Existing distributors of these brands also got to enjoy the opening up of new market.
Less successful attempts
Creativity abound in new market. Some people imagined that the wireless LAN market can be easily stretched to remove the need of wired network. To that extent, they created wireless LAN NIC for desktop computers. There are few problems with that. Firstly, desktop computers are not mobile, so the freedom of movement that wireless LAN offers is meaningless there. Secondly, wireless LAN NIC is considerably more expensive than wired NIC. Thirdly, wireless LAN is generally slower than wired LAN. Combine those, and all you get from using wireless LAN NIC in desktop is paying more to get rid of network cabling and enjoy slower network. That said, there is still small demand for such thing in the market up to the time of the writing of this article.
Creative uses emerged
Creative entrepreneurs dreamt of striking it rich by providing ways for people to work with their computers from virtually anywhere. How do you do that? By putting wireless LAN connected to wired broad-band connection in as many places as possible. And why would that business model work? Because every visitor who would want to use their wide-spread wireless LAN has to subscribe to them and the revenue will be split between this wireless Internet Service Provider (ISP) and the venue owner. It can be easily seen that this game would require huge capital outlay. The wireless ISP had to establish broad-band internet connection to every single location it wanted and provided the wireless LAN hardware on that site at the same time. And they had to do this on large number of locations. Without enough locations, this business model just wouldn’t work.
One of the early wireless ISP was MobileStar, which was founded in 1998 and was very successful in signing up great locations like Starbucks. Initially deploying its network using HomeRF technology, MobileStar then had to change its network to IEEE 802.11 as it was becoming more popular, racking up yet another huge capital outlay, no doubt. MobileStar popularised the term hot spot. By 2001, the huge capital requirement to survive in this business had wrecked MobileStar, and it was resurrected as T-Mobile HotSpot division. And there were few more entrepreneurs trying their luck in this segment, like Surf & Sip, which was founded in 2000, or WayPort which re-fashioned themselves to take on this new field.
Seeing that the market for such service is large enough, few companies tried to be creative by looking for ways to provide similar service while circumventing the large capital requirement. Thus born the so called wireless LAN aggregators, such as Boingo in 2002. Instead of building up its own network, aggregators simply give access to existing wireless LAN network already put up by venue owners or even other networks. iPass also re-modelled itself as wireless LAN aggregator. Technology giants like IBM, Intel, and AT&T was also interested in similar business model. In 2002, together they formed Cometa, aiming to do build extensive hot spot network. The story was captured very well in Fortune article from 2003 titled “The Really, Really Messy Wi-Fi Revolution“. The concept even caught on later in developing market, for example the creation of CBN HotSpot service in few Indonesian locations.
In 2000, the IEEE 802.11 technology was given a sexier name: Wi-Fi. Aside from the increase in population of Wi-Fi-enabled laptops, other mobile devices begun to sport Wi-Fi support too. In business setting, another class of device started to utilise the availability of Wi-Fi in 2001: mobile computers. Prior to Wi-Fi, these devices, which have inherent need for mobility within the premise of the business, was relying more on point-to-point 2.4 GHz Radio Frequency (RF) communications. The availability of Wi-Fi meant that it was simpler to set up connectivity to multiple mobile computers roaming the warehouses and shop floors. Starting from 2002, another kind of device started to sport Wi-Fi capability: smart phone.
Even with so many old network hardware players piling up into this market, the pie is still large enough for people to found a new company catering to un-served needs like enterprise-grade access point, as in the case of Aruba in 2002. In this case, new opportunity is also created for distribution of their products to new geographics, like what Helios is doing for Aruba in Indonesia. Later on, even the world’s largest chip maker, Intel, created new product and jumped into this new market, using creative bundling branded Centrino in 2003.
Aside from the obvious fit for businesses, Wi-Fi was also finding new place to call home, and that is home itself. The proliferation of home broad-band internet connection, the increase in number of computers per house hold, and the rise of inexpensive laptops were making people to want to establish a computer network at home. This market was tapped very well by the network hardware manufacturer by bundling the broad-band internet connection (e.g. ADSL router), Wi-Fi access point, and Ethernet switch in single product, greatly simplifying the efforts needed to set up home network.
Market consolidation & evolution
As is always the case with any new industry, not all players who entered early on stayed on. The home network market grew so well that Cisco, itself a long-time network hardware manufacturer in the enterprise market, acquired LinkSys in 2003 to quickly get foot-hold in this new market. In 2004, facing really large capital requirement, Cometa was closed. In 2008, WayPort was acquired by AT&T. In 2010, 3Com was acquired by HP to bolster its networking division. In 2011, QualComm acquired Atheros. In 2013, Cisco divested its LinkSys operation after owning it for 10 years and sold it to Belkin.
Parallel to the development of wireless LAN, the telecommunication industry was also ramping up their data services. As data service became faster and faster (2.5G EDGE to 3G UMTS to 3.5G HSDPA and so on), suddenly users had simpler option to work from anywhere with the added benefit of seamless freedom of movement. The hot spot business model has no hope in catching up with the coverage of the data network that the telco players were rolling out. So, for the hot spot business model, some has transitioned out to be, effectively, loss leader. That is, products or services that a business provides to its customers at a loss so that it can attract larger crowd to its more profitable products or services. Many cafes & restaurants now provide free Wi-Fi for its patrons. In fact, not having free Wi-Fi is now a great detriment for cafe business. Air ports as well as shopping centres around the world also provide free Wi-Fi for its mobile customers. With the rapid increase in mobile data connection causing stress in its network, telecommunication carriers themselves are actually experimenting with rolling out hot spot network. The intention is to make it functions as data communication off-loading channel, for example Telkomsel FlashZone or IndosatNet