TechTV talks to EtherLinx: EtherLinx made a big splash two weeks ago with a New York Times article about the founders working a few houses away from Apple's starting garage building a wireless device that combined long-range networking with local Wi-Fi service.
Unfortunately, all the pieces the founders are discussing just don't add up, and we don't have enough additional detail to make determinations about their innovation.
What they're doing is not unique at all: only the price is. They told John Markoff of the New York Times they were using a software radio, but at the same time, they said they were modifying off-the-shelf equipment. This shrinkwrapped equipment doesn't use a software-defined radio, but standard baseband chipsets that are wired to specific frequencies.
They said they are using a unique antenna: if the FCC hasn't approved their modified devices for starters, or approved them with their antenna, or approved their antenna, they're violating the law. They're running a trial in Oakland, but without certification, which their description of equipment makes it sound certain they need, they are operating outside of the regulatory guidelines for 2.4 GHz devices. (Some of what they're doing is pure Wi-Fi, but they couple that with their own lower-speed protocol that is probably running frequency hopping, like the older 802.11 standard.)
They said they will sell the thing for $150, but they're describing at least $150 in retail equipment plus a substantial amount of labor. To sell something for $150, it must cost you, all in (post-EBITDA with all overhead) much less than $120, which is what you need to sell it to retailers or partners for. Those partners and retailers may want a much lower wholesale price to sell the units at $150.
They seem amazed that they're running 20 miles at 2 Mbps. I can point to several hundred ISPs around the country running several miles with gear that costs $500 to $1000 per installation. On (and off) the coast of Maine, Midcoast Internet Solutions has been offering multi-mile, mult-deca-mile connections since 1997. They serve an island 20 miles off the coast, the only other connection to which is underground phone cable. I can point you to a number of people in the Bay Area running legal and skirting-the-issue multi-mile links.
Here's the fax machine part. The fax machine isn't innovative in its components. It was innovative in taking a cheap modem (originally, very slow), a cheap printer (originally and often still, thermal), and a cheap scanner (297 dpi in its last revision many years ago), and turning it into a package.
Likewise, it looks like EtherLinx is taking commodity equipment, the price of which continues to fall, and combining it into what could be the fax machine of wireless networking. But it ignores current market and financial realities: how are they doing what they claim to be doing when tallying the numbers just doesn't work? And, are they so misinformed about the state of wireless ISPs around the country that they believe they have come up with something entirely new instead of something merely convenient and cheap?
Showering the media with over-optimistic early information lacking technical detail is a great way to put off your potential market, which includes these wireless ISPs. If this Maine ISP, just for instance, could go to customers with a $150 device and a $50 to $100 install instead of the $800+ price tag associated today, they'd be overjoyed. It would also make Alvarion, Cisco, 3Com, and a few other firms tremble in their sales boots.
I hope to get the answers. After an initial response from the founders expressing interest in talking, no follow-up and no further response. I understand that they're overwhelmed. But I hope they'll spread a little more information instead of hype about what they've developed.