Possible Correction: AT&T Says "All-Fiber," Not Fiber to the Basement and G.fast
If so, my article that AT&T will use 400-700 megabit G.fast to five million is a mistake. AT&T's press release I just read (below) promises "Within 4 years, AT&T will offer its all-fiber Internet access service to at least 12.5 million customer locations, such as residences, home offices and very small businesses." It may well be that the pr people at AT&T don't understand the technology being used and much of the build will in fact be Fiber to the Basement and copper, probably G.fast, to the apartment. But if I may have an error, it's my job to issue a correction.
Market for mobile chips disappearing. When Henry Samueli shut down Broadcom's cellular baseband business, I happened to be visiting one of the founders. Many of the 3,000 engineers affected were his friends and the pain was apparent. However, I could see why the decision was necessary. Apple and Samsung dominate the mobile phone business and each made their own chips. Those two companies had growing market share and there just wasn't enough merchant business to sustain the chip vendors. Qualcomm's scale gives them a major advantage and the very efficient Mediatek has also grown.
FT/Orange Spain going to 10M & 14M. Without much publicity, Spain has pulled far ahead of other large nations in fiber homes. The U.S., Italy and France have passed fewer than 25%, Germany and England less than 10%.
The center of the Internet has moved South. Carter Horney of Forward Concepts has just published a
and Verizon could easily triple their network capacity without breaking the capital budget. The technology is obvious but the change in business models is unlikely. Sprint has remarkable holdings around 2.5GHz. Dish spectrum is totally unused.