GPRS Prices: The Readers Have Their Say

Dan,
Regarding your recent article on the high price of GPRS services -- do the algebra. Compare the cost of 1 Meg of data to the equivalent voice time. If you want GPRS to be low cost then the implication is that voice would be virtually free. As it is, the carriers have bandwidth constraints (and quality of service issues accordingly). I think $7-$10/Meg is a huge improvement over the pricing we've seen for earlier services (i.e., CDPD [cellular digital packet data]), given the substantially faster and more ubiquitous GPRS.
Rather than pan the service, why not focus on the new capabilities it offers? GPRS is a powerful enabler for connecting mobile workers to their enterprise systems. With the right software, mobile workers can interact with these systems as if they are in the office -- for just a few megabytes of consumption a month.
Regards,
Jeff Kleban
Executive VP Strategy & Product Management
Syclo LLC
What's happening is fairly easy to understand if not to excuse. A coded voice call takes about 8 kbit/s; if I download at (say) 32 kbit/s, I've used up capacity for 4 phone calls. Thus 1 Mbyte (8 Mbit) ties up 4 phone circuits worth of capacity for about 4-5 minutes, or 20 minutes worth of chargeable phone calls. At $0.20/premium minute that's $4: This is what the service provider would like to charge to stay revenue-neutral, and of course they hope to make extra money from data.
The problem is that a normal human being wants to pay by the minute of use, and as we know, 1 Mbyte is just a Web page with a few nice graphics on it. In 20 minutes of normal Web usage one will probably download 10 Mbytes of data: $2.50 if you're hooked up via Voicestream/MobileStar 802.11b connections on their most expensive pay-as-you-play plan, or $70 via GPRS (with a longer wait). Fundamentally, the problem is that the price we're willing to pay for data bits is much less than for voice bits, because human beings want to pay by the minute of useful information, not by the bit. The problem only gets worse for "3G" services, where the data user kicks off more potentially-lucrative phone calls for a serious download but expects to pay the same per-minute price.
The solution, of course, is a fundamentally cheaper pipe -- i.e., Ethernet-based, like 802.11 -- for data services.
Dan Dobkin
WJ Communications Inc.
Also, in our testing on Voicestream’s network of our PC card, we ran it for 24 hours of continuous downloading, and our bill was >$1000 for that one session! Wow! Well, at least we know the upper limit (sort of) for GPRS users!
Prices have to come down. I fear for the content providers too -- corporate email might just about scrape through as a justifiable GPRS expense, but how can a consumer app developer (game studio say) convince VCs that anyone is going to pay such high airtime rates and pay for their content???
Perhaps operators will fix this by having device-specific rate plans, but this is pure speculation.
Ben Gibbs
— Dan Jones, Senior Editor, Unstrung
http://www.unstrung.com
Like what we have to say? Click here to sign up to our daily newsletter
Cable Next-Gen Technologies & Strategies – Making cable faster, broader, deeper, better
If you want to know where the cable tech space is heading, we've got you covered. Join the biggest names in the industry in Denver, CO on March 14 & 15 for the latest edition of Light Reading's Cable Next-Gen Technologies & Strategies conference.
Back for a record 16th consecutive year, Cable Next-Gen is the premier independent conference covering the broadband technology market. This year's edition will tackle all the top tech topics, including 10G, DOCSIS 4.0, Distributed Access Architecture, next-gen PON, fixed wireless access, network virtualization, the Digital Divide and more
Get your free operator pass here.