Interesting article, but I think it is a bit off base. Comparing Skype and other computer based VoIP to phone service provided by my cable provider is like comparing the first kit pc's to the Digital VAX you used in college at the same time. Today's computer VoIP apps are like those kit pc's, the test bed/playground of tomorrows Microsofts and IBMs. Skype will look a lot different in 2 years and cable phone service will too. The one thing I will almost guarantee is that they will be much more alike in 2 years. Skype will be scaled to a carrier grade service, and with that add additional services. Cable phone service will probably add some interactivity as cable providers continue to give customers the ability to self configure the other services (data, video, etc.) they push through their cables.
The author does expose a challenge with marketing VoIP services. Users expect their phone to work, ALWAYS. If you pick up the phone and don't hear a dial tone, you fear that someone cut the lines, or some other nefarious act. You don't expect the phone company to have a problem. Also, you want to be able to pick up the phone, push 7 or 10 keys and then talk to someone. With VoIP being more a child of computers than telephony, it is being overburdened with features and thus complexity. While cable phone service is VoIP in that it is your voice transported over a data network, most people will look at VoIP as a service, and thus look for the unique features and calling price plans. Where features are pushed ahead of simplicity, complexity will be the norm.
Eventually our concept of simplicity will broaden a bit to include some VoIP processes while VoIP processes will improve in the areas of user interface and functionality. Remember our current land line phone process is MUCH more complicated than it used to be. In the past I only picked up the phone and told Marge who I wanted to talk to. Now I have to know all of the numbers to dial. Man progress sucks.