Monopoly! Monopoly! Monopoly?

So, over the summer there was talk about investigating deals between wireless carriers and phone manufacturers for antitrust violations.

Now the DoJ is looking into investigating IBM for anticompetetiveness. They refuse to license their software to potential competitors.

Finally, though I don't have a link, I swear I heard a story last night on the radio about potential action to force wireless providers to open their networks to competitors.



These three stories, to me, run the whole spectrum of antitrust viability. On the one hand, selling the first really cool peoples' phone exclusively through one carrier isn't a monopoly. Apple didn't create any barriers to entry in the cool phone market and they didn't prevent consumers from getting a cell phone. Did they get some monopoly profits through signing an exclusive deal with ATT? Probably, they're trendy but not stupid, I expect Apple signed the most profitable deal they could get for their new fangled cool phone. But really, they were the market leader, they were the first to come up with the cool phone concept in a package people would buy, they deserve to reap the benefits until companies like Google and Blackberry catch up and start competing. I'd personally much rather see Apple catch a couple years of possible monopoly profits than give them a patent and monopoly rights for a long time on the "cool phone" concept.

On the other hand, you have wireless providers. Infrastructure costs are a huge barrier to entry, just as they are in the long distance carrier and local LEC markets. Our system of forcing local and long distance providers to sell access on their networks currently works pretty well, keeping prices down and providing access to just about anyone. I don't really see a distinction forcing Verizon to sell access to their cable allowing competitors to offer me long distance service and forcing Verizon to sell access on their cell towers. This seems to be a natural extension of current telecomm policy.

Finally, IBM. I don't know enough details to really have an opinion either way. Is your mainframe more like a long distance line carrying traffic between Chicago and Denver or is it more like a cool phone? To be sure, once you have a mainframe in your data center, it's exceptionally difficult to get it out of there. And IBM is no champion of open software (Or even standards: A lot of those applications that "are estimated to handle 50 billion transactions a day in such areas as automated teller machines, health records and accounting." still don't even talk to the mainframe via TCP/IP, using proprietary IBM protocols instead. Much to the chagrin of network engineers the world over.)

But, they're IBM's products. And why should IBM have to share their ball? The applications that those mainframes run aren't special in some what, they don't need to run on mainframes, we have the computing power to run them on other architectures now. Plenty of accounting, HR and database systems run on everything from Big Sun and HP boxes down to Intel based Windows or Linux servers. Companies stay with their mainframes in part because it's a tremendous pain to move their applications off them, and therefore very expensive. They also probably stay in part because those mainframes and applications that they've been using for 20 or more years are rock solid, they just don't go down. In other words, companies are staying with IBM mainframes because it's more profitable for them to do so, not because mainframes are some magical tool and only IBM can provide them.

I'll have to keep track of the case and see if it's built on anything more than IBM refusing to share it's ball with the rest of the kids.

(Yes, this is my version of brief.)