Navini Networks

» see navini.com as it lived, on the wayback machine

~ the obituary ~

Around 2005 the question of how a house gets broadband still had more than two plausible answers. Cable and DSL were winning, but laying wire was expensive and much of the planet had never been wired at all. Fixed wireless was the alternative that kept almost working, and Navini Networks bet its entire business on it.

Wu-Fu Chen and Guanghan Xu founded it in Richardson, Texas in January 2000, roughly two months before the market that funded companies like it came apart. It got through that and shipped a product line covering both ends of the link, the base station at the tower and the modem in the customer's house. The base station was the Ripwave MX8. The modem was the part that mattered commercially, because it meant an operator could mail a box to a customer instead of sending a crew with a ladder.

Cisco announced a definitive agreement to buy the company on 23 October 2007 for roughly $330 million in cash and assumed options, called it its 124th acquisition, and completed the deal on 19 December 2007. Navini went into Cisco's Wireless Technology Group. The base stations were relabeled as the Cisco BWX 8300 series, and after that the only documents that used the word Navini were the ones announcing that something was no longer for sale.

~ the case file ~

~ the antenna that aimed itself ~

The engineering claim was beamforming on a small adaptive array. Earlier fixed-wireless systems needed line of sight to the tower and an installer on the roof to get it, which put a hundred dollars or more of truck roll into every single customer and killed the economics before the first bill went out. Navini's array worked out where the signal was coming from and steered toward it, indoors, from wherever the customer had put the box.

That idea is why the company existed at all. Self-installation is what turns a wireless system from a niche rural product into something an operator can actually scale, and the industry eventually agreed: the same thinking runs through every home mesh node and 5G fixed-wireless gateway sold today.

~ the bet on wimax ~

Navini started before WiMAX was a standard and had to retrofit itself onto one. Its early systems ran a proprietary scheme Cisco later documented as SCDMA mode, and the later ones moved to 802.16e. Operators who had bought the first generation got a software upgrade path rather than a replacement, which was decent of everybody involved and also a sign of how small the installed base was.

The larger bet did not come in. WiMAX lost to LTE over the following five years for reasons that had little to do with the radios and a lot to do with which standard the big carriers and handset makers lined up behind. Cisco marked the BWX line end-of-life, with the last day to order falling on 29 November 2010, less than three years after it bought the company.

~ inside cisco ~

Navini was Cisco's 124th acquisition, by Cisco's own count, so the pattern was thoroughly worn in by the time it arrived. The purchase buys an engineering team and a product that fills a gap in a roadmap, the brand goes in the bin on day one, and the question of whether it worked gets answered quietly in an end-of-life bulletin three years later.

navini.com still resolves and nothing answers on it. That is a slightly different kind of dead from a parked domain: somebody is paying for the registration while no machine anywhere is configured to serve the name.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »