Using open source to build the ultimate walled garden
AT&T moving to dominate user experience?
Much has been made about AT&T's surprise entry into the OpenStack project earlier this week. Most notably, why the major U.S. carrier decided to sign up for OpenStack in the first place.
That's no slam on OpenStack, mind you: the Rackspace-owned cloud computing project is much-beloved in the open source community for the technology and the Apache license that covers the project. The fact that governance will be shifting from Rackspace proper to a planned OpenStack Foundation definitely helps, too.
[ Free download: Selecting the right cloud: A step-by-step guide ]
But why would a carrier want to make use of OpenStack? There are some clues, most notably AT&T's planned AT&T Cloud Architect program, a service that will provide "public, private, and bare metal cloud servers or choose dedicated options."
In other words, AT&T is about to take on Amazon's EC2 and S3 cloud services, and OpenStack's technology is going to be the engine that drives it.
It's not going to be a straight head-to-head match-up, either; AT&T appears to be sweetening the pot by announcing HTML5 support as well as enhancing its API platform to provide such things as in-app billing.
For me, this is an important clue in itself: by trying to attract developers, it looks like the AT&T--the carrier--is trying to move up and down the stack to capture the servers on the back-end with OpenStack and the user space with some sort of Web-based interface. If this is so, that would represent a huge move: because it would be the first time a network provider would theoretically own the user's eyeballs from the desktop/device all the way to the servers.
It won't be 100 percent, of course, since users will be able to zip around the Internet at will. But imagine this: AT&T providing customers with an interactive interface on their TVs, devices, or desktops that has apps (free or fee-based) provided by AT&T and running through AT&T's mobile and land-based networks. There'll be a browser, too, so users can surf; perhaps set up like Amazon's Kindle Fire browser Silk, which will handily pre-load pages for users. All of this will run quite well on OpenStack-managed cloud servers, and provide AT&T with a gold mine of user-generated surfing, commerce, and other data.
An analogy for old folks like me: it would be like CompuServe, if CompuServe owned the phone company.
Leaving aside the potential problems for user privacy here--and oh, there are many to be addressed to be sure--a plan such as this would represent a stunning coup for AT&T, since they would be able to provide the one thing Apple and Google have not been able to have in their respective plans to own the entire stack: the network on which all communications must flow.
It would also be a huge validation for open source, because every single level of such an offering would be driven by open source software: from the interface and the browser to the bazillions of Linux-based servers managed by AT&T's OpenStack-based cloud.
As a starry-eyed geek, such an offering intrigues me, because it may mean that AT&T might be able to provide users with a la carte programming… something that it doesn't do now and is driving users like me to devices like Roku that enable my family and me to purchase and subscribe to only the entertainment content we want.
As a more savvy consumer, the thought of AT&T or any one network having that much involvement in my online experience is more than a little problematic.
There is a lot of work out about AT&T's plans, since I am offering little more than speculation. But I offer you this: AT&T's desire to have a hand in all levels of the user stack could be reflected in Google's own plans to own the part of the stack they're missing: the network. For this, I give you Google's pilot fiber program in Kansas City, KS.
If Google wants the whole stack, you can bet your bippy that others will want it, too. AT&T already has the hard part, if you'll pardon the pun: the wires. Thanks to open source, moving up and down the stack is a much easier and cheaper proposition than Google wiring the country.
Let's see if this call goes through as dialed.
Read more of Brian Proffitt's Open for Discussion blog and follow the latest IT news at ITworld. Drop Brian a line or follow Brian on Twitter at @TheTechScribe. For the latest IT news, analysis and how-tos, follow ITworld on Twitter and Facebook.