Michael Oh et al try to bring business-sponsored Wi-Fi to larger area: The notion is that a number of businesses underwrite the nominal cost as a factor in bringing more customers into the area as a whole. Businesses don't charge customers for air conditioning, Michael Oh says, so why charge for Wi-Fi? Ah, but businesses don't let customers use their phone for free, either.
Nice straw man there with the phone analogy. Phones can only be used by one person at a time, and have somehow managed to remain metered (as opposed to bandwidth, which is pretty close to too cheap to meter). Furthermore, many small businesses use the phone line for credit card transactions, making the "one user at a time" thing far more problematic. OTOH, a/c is multi-user and too cheap to meter. So is milk, using RAIP (Redundant Array of Inexpensive Pitcher) technology.
I think it is a good analogy. bandwidth may be cheap now, but in a few years I believe you will change your mind. chip companies are shipping 40M wifi chips this year and growth is expected to double every 2 years or so..getting crowded fast.
Also, until some sort of mesh networking happens, backhaul connections to the net are still the bottleneck in the equation. The best way to handle scarcity in ANY situation is thru private property and trade.
802.11b is already faster than than the backhaul at most hotspots. Has it been a problem? Just because the chips get faster dosn't mean that bandwith to each user can't be throtled and/or allocated evenly at hotspots. Right?
And why won't bandwith continue to get faster and cheaper as well?
Fantastic question, D-Man! Backhaul over wire hasn't gotten much cheaper -- not at even a fraction of the same rate as LANs have increased in speed and dropped in cost. What's likely to happen next is that wireless backhaul starts delivering 5 to 10 Mbps at prices that you'd pay for T-1 today.
It is pretty easy to hook into a fat fiber pipe as long as you are willing to go to them, they rarely come to you. 802.16a will provide a very fast way of bypassing the Telco strangelhold and bridging the costly last 1-5 miles (depending on terrain) in towns and cities.
I can envisage hot spot operators buying a fat connection at a downtown fiber pop and then shooting in numerous directions off the roof, and in the process cancelling a lot of T-1 circuits.
Williams Fiber have done a deal with National Broadband who are going to do this exact thing at everyone of the Williams regeneration stations, and many of those are in strange and remote locations dotted every 70 miles along the fiber.
Cheers Nigel
Maybe companies should let people use their phones for free. I mean, why not... local calls don't cost them anything more, and if it brings in more traffic, all the better...
My point is simply that just because you might be able to make money on it, that doesn't necessarily mean that it's the most profitable business model for many of the players (or the consumer).