SMS is a cash cow. It's the only thing that really makes money for some carriers. After voice, it's generally the biggest contributor to ARPU (average revenue per user) -- more than selling ring tones or 3G service.
This isn't really a big secret; read any quarterly report from a carrier or, heck, read the press release about the report.
A terabyte costs only $100? Someone had better tell NetApp.
I dispute that so-called professor's assertion about the size of a text message making it cheap; for transmitting data over long distances the limiting factor is packet latency; for tiny messages proportionally greater of the system's time is going to be spent waiting for packets to move on the wire. It's quicker and easier to move a gigabyte file 3000 miles than 10,000,000 100-byte files. Per byte of data, SMS traffic is probably the most expensive traffic to route and switch.
SMS messages are much smaller than a single TCP packet is capable of carrying (or, if you want to roll your own re-transmit logic, a UDP datagram). They also have effectively no QoS guarantees attached, so you can queue them for transmission whenever bandwidth is unavailable (or would degrade service for more critical streams, like active voice calls).
Since the "last mile" bandwidth is effectively free, by virtue of being embedded in the control channel, there's no reason other than "consumers will pay it" that justifies the cost of SMS traffic.
The Internet is full of services that blast millions of tiny packets between networks (routing, DNS, etc.) without incurring significant per-transaction costs. There's nothing special about SMS that should drive costs up as much as they have been.
You know what's way harder than transmitting an asynchronous 160-byte max message? Transmitting a real-time several-KB-per-sec stream of voice data.
No matter how you slice it, text messages are a scam. If it was even remotely a load of any significance on the network, we wouldn't be able to make voice calls.
Even if SMS is the most expensive traffic per byte, for the sake of argument, it's still insignificant.
Who cares? Prices aren't based solely on cost, they're based on competition and demand. The author seems to miss the point of capitalism.
From their point of view, why shouldn't AT&T, Verizon, Sprint, and TMobile charge whatever the market will bear? What's the incentive to charge less?
Of course, there's an opportunity there. Should one of them, or some new company, start charging $5/month for unlimited texting, it could be pretty popular.
This isn't really a big secret; read any quarterly report from a carrier or, heck, read the press release about the report.