Slashdot Mirror
SMS 4x More Expensive Than Data From Hubble
paradoxSpirit writes "Physorg has a paper comparing the cost of text messaging versus the cost of getting data from Hubble Space Telescope. From the article: 'The maximum size for a text message is 160 characters, which takes 140 bytes because there are only 7 bits per character in the text messaging system, and we assume the average price for a text message is 5p. There are 1,048,576 bytes in a megabyte, so that's 1 million/140 = 7490 text messages to transmit one megabyte. At 5p each, that's £374.49 [$732.95] per MB — or about 4.4 times more expensive than the 'most pessimistic' estimate for Hubble Space Telescope transmission costs." "Hubble is by no means a cheap mission — but the mobile phone text costs were pretty astronomical!""
And don't forget that both the sender *and* the recipient pay for a text message for every one sent.
Sprint's charging $0.20 each for these now-a-day (unless you have another plan of some sort). It's just the latest ripoff in the mobile phone industry.
TFA is talking about the transfer of data, not how many little bits are actually involved in the transaction. Headers and transmission overhead are not data. If you downloaded a CD ISO, you would not say that you downloaded 946MB and include "overhead" in your figure. Did you include your name and the class number in the word count for your papers in college?
I'm sure they have a way of putting a block on a particular service -- I've been with Cingular (before the AT&T merge) and Sprint and both of them can block SMS, you just have to call and push the CSR to do it. And if they tell you they can't, ask for next-tier support. If they tell you no, threaten to take the matter up with your state's AG and the FCC. That usually gets a quick response. (If Sprint can do it, and their customer service is quite possibly the worst I've ever encountered, then T-Mobile can do it.)
In-flight calls or data are a poor example. You're talking about putting in equipment that costs more than typical network equipment because of requirements like low EMI, light weight, minimal maintenance, ground stations to handle the data, programming to manage handoffs at 500+ mph, and the process of getting FAA approval when you integrate it with a jillion other systems on a commercial airliner. It genuinely is expensive. Even at $10/hour that Boeing was charging for their Connexion internet service, they lost huge amounts of money on it (I think partially because they over-engineered the system, but I'm not very familiar with the details).
The cost of using it are very low, but the costs to initially add the feature are very high. Then you add in the fact that usage rates are typically low (only a handful of passengers buy it, only "full-service" airlines install the equipment), it can be hard to make it pay for itself.
Of course, they do add a high margin on top of their projected costs because they can without affecting the demand much, but the fixed costs still dominate (at the moment...data services will be much better integrated in the coming generation of airliners, and we may be moving towards allowing cell phones in flight, too).
SMS is the opposite. They aren't seeing low usage on new, expensive infrastructure. They're seeing high usage on existing, paid-for infrastructure.
The SMS scheme really isn't a very good one. SMS messages get multiplexed into the control channels on the mobile phone network, and it's really a 2nd generation technology. The size of the control channels is fundamentally limited, but each slot is big enough for a text message. So the providers squeeze the SMS into it because it fits and it doesn't require re-engineering their protocols to fit it in the voice channels. This is also why SMS is limited to so few characters: That's what fits in a time-division on the control channel.
Unfortunately, it proved to be a popular service. The limited extra space fills up quickly. In fact, it's theoretically possible to launch a relatively wide-area DDoS attack by sending only a couple hundred messages per second from zombie clients. To get the best return on their existing capacity, providers raise the price to discourage excessive use.
The puzzling thing in my opinion is that it's taking so long for this service to shift from being side-banded in the 2G scheme to being normal data packets on 3G networks. As that happens, the capacity for text explodes (text is way more compact than voice, pictures, video and other planned 3G content) and the providers can leverage the genuinely low cost of text to undersell their competitor's plans. A pricewar ensues and the consumers win.
But it hasn't happened yet. My best guess is because the companies realize that the first one to make a substantial move in this direction will only enjoy success for a short time before the others all catch up. Then the competitive advantage is gone and profits have dropped close to zero.
No, I haven't sourced much of this. It's mostly conclusions from discussions with friends who work in the mobile industry. Feel free to correct the parts I got wrong.